1. Trang chủ
  2. » Kỹ Thuật - Công Nghệ

Servo magazine 07 2008

84 443 0
Tài liệu đã được kiểm tra trùng lặp

Đang tải... (xem toàn văn)

Tài liệu hạn chế xem trước, để xem đầy đủ mời bạn chọn Tải xuống

THÔNG TIN TÀI LIỆU

Thông tin cơ bản

Tiêu đề Servo Magazine 07 2008
Trường học SparkFun Electronics
Chuyên ngành Robotics
Thể loại Magazine
Năm xuất bản 2008
Thành phố Corona
Định dạng
Số trang 84
Dung lượng 15,41 MB

Các công cụ chuyển đổi và chỉnh sửa cho tài liệu này

Nội dung

Live Test Researchers tested the Lewis robot photographer on a group of 5,000 subjects over a period of 40 hours.. During this 40 hour run, people guests at a large technology event eith

Trang 2

Let your geek shine.Meet Leah Buechley, developer of LilyPad—a sew-able microcontroller—and fellow geek Leah used SparkFun products and services while she developed her LilyPad prototype

The tools are out there, from LEDs to conductive thread, tutorials to affordable PCB fabrication, and of course Leah’s LilyPad Find the resources you need to let your geek shine too.

©2008 SparkFun Electronics, Inc All rights reserved.

»Sharing Ingenuity

S P A R K F U N C O M

Trang 3

Free Book

with Kit

Trang 4

08 Robytes by Jeff Eckert

Stimulating Robot Tidbits

12 GeerHead by David Geer

Lewis, the Robot Photographer

16 Ask Mr Roboto by Dennis Clark

Your Problems Solved Here

58 Twin Tweaks

by Bryce and Evan Woolley

There’s a New Humanoid on the Block

64 Robotics Resources

by Gordon McComb

Stocking Up with Surplus Electronics

67 Dif ferent Bits

77 Then and Now by Tom Carroll

Robotics — A Historical Perspective

PAGE 12

THE COMBAT ZONE

Trang 5

Tech Zone.

by Robert Doerr

Scaling and inverting encoder values

to fit your particular application.

by Fred Eady

Learn what it takes to design, build, and code a heavy duty DC motor driver module.

— Part 2

by Alan Marconett

This final installment examines the QwikFlash controller board and the software that runs Loki.

SERVO Magazine (ISSN 1546-0592/CDN Pub Agree#40702530) is published

monthly for $24.95 per year by T & L Publications, Inc., 430 Princeland Court, Corona,

CA 92879 PERIODICALS POSTAGE PAID AT CORONA, CA AND AT

ADDITION-AL ENTRY MAILING OFFICES POSTMASTER: Send address changes to SERVO

Magazine, P.O Box 15277, North Hollywood, CA 91615 or Station

A, P.O Box 54,Windsor ON N9A 6J5; cpcreturns@servomagazine.com

PAGE 44

PAGE 39 PAGE 34

Features & Projects

Trang 6

Published Monthly By

T & L Publications, Inc.

430 Princeland Court Corona, CA 92879-1300

(951) 371-8497

FAX (951) 371-3052 Webstore Only 1-800-783-4624

www.servomagazine.com

Subscriptions

Toll Free 1-877-525-2539 Outside US 1-818-487-4545

P.O Box 15277 North Hollywood, CA 91615

PUBLISHER

Larry Lemieux

publisher@servomagazine.com ASSOCIATE PUBLISHER/

Jeff Eckert Tom Carroll Gordon McComb David Geer Dennis Clark R Steven Rainwater Fred Eady Kevin Berry Robert Doerr Ted Larson Alan Marconett Kym Graner Bryce Woolley Evan Woolley Heather Dewey-Hagborg Nick Martin Mike Jeffries Bryan Ruddy

CIRCULATION DIRECTOR

Tracy Kerley

subscribe@servomagazine.com MARKETING COORDINATOR

WEBSTORE

Brian Kirkpatrick

sales@servomagazine.com WEB CONTENT

Michael Kaudze

website@servomagazine.com PRODUCTION/GRAPHICS

Shannon Lemieux Joe Keungmanivong

ADMINISTRATIVE ASSISTANT

Debbie Stauffacher

Copyright 2008 by

T & L Publications, Inc.

All Rights Reserved

All advertising is subject to publisher’s approval.

We are not responsible for mistakes, misprints,

or typographical errors SERVO Magazine

assumes no responsibility for the availability or condition of advertised items or for the honesty

of the advertiser.The publisher makes no claims

for the legality of any item advertised in SERVO.

This is the sole responsibility of the advertiser Advertisers and their agencies agree to indemnify and protect the publisher from any and all claims, action, or expense arising from

advertising placed in SERVO Please send all

editorial correspondence, UPS, overnight mail,

and artwork to: 430 Princeland Court,

Small is Big

When it comes to robot

components, small is big If you’ve

followed the robotics news lately, you

know that academic and military

R&D communities are busy at work

developing robots that mimic – in

form and function – small crawling

and flying insects Need to locate

survivors in the rubble of a collapsed

building? Simply release a swarm of

heat-seeking crawling robots that can

squeeze through cracks without

disrupting the rubble and

endangering trapped victims Need

an up-close view of a hostage

situation? A swarm of flying

microbots with photosensors could

provide police with a composite,

real-time image of the victims and

their captors

Despite ongoing advances in

research laboratories, there are

numerous challenges that must be

overcome before practical

autonomous insect swarms can

become a reality There are issues

of how to provide communicationsbetween each insect-sized robot andtheir human masters, local

computation, sensors, power, and

of course, powerful, lightweight,controllable micromotors And there’sthe underlying issue of cost

A recent advance in the area

of micromotors has been thecommercial availability of linearmicromotors from New Scale

Technologies (www.newscaletech.

com) Their series of Squiggle motors

fills the void between the microscopicnanomotors and the miniature servosand electronic/pneumatic linearactuators popular among roboticsenthusiasts

I had the opportunity to evaluateNew Scale’s mid-sized offering — theSquiggle SQL-1.8-6 linear motor —shown in the photo As the namesuggests, the motor is a mere 1.8

mm in width The rectangular motorbody is 6 mm long, with a 12 mmaxial screw running through itscenter The 160 milligram SQL 1.8 iscapable of handling a 30 g loadwhen driven by a 400 mW, 40V, 171kHz pulse The even smaller SQL 1.5linear motor can work with a 20 gload As illustrated in the photo, theelectrical connection to the Squigglemotor is via a flex printed circuitstrip

With a PC-based controlapplication and USB-to-Squiggleinterface, I was able to vary the travelrate from micrometers per second tomillimeters per second, with animpressive 0.5 micrometer resolution

Although the relatively fragile motorwas glued to a polycarbonate mount

Mind / Iron

by Bryan Bergeron, Editor Œ

Mind/Iron Continued

Piezoelectric Squiggle micro motor on

polycarbonate mount shown next to a

six-pin DIP for size comparison.

Trang 7

for evaluation purposes, I could easily envision a

spider-sized eight-legged walker, powered by 16

skeleton-mounted Squiggles

The size of the peripherals that accompanied the

motor — a wall wart power supply, a USB driver card, and

a three-foot USB cable — not to mention the desktop PC

and software — explains why the robotics shops aren’t

offering autonomous robots sporting Squiggle-based

grippers and actuators Even a six-pin DIP dwarfs the

Squiggle, much less a PIC or BASIC Stamp However, the

control issue should be partially solved by the time you

read this New Scale has a miniature ASIC driver under

development that could form the heart of a Squiggle

spider robot

Power issue is another matter The smallest battery

packs that I’ve used are thin-film lithium-polymer cells

designed for miniature indoor R/C aircraft The thin,

dime-sized cells power a single-motor aircraft for about

five minutes As such, an autonomous eight-legged

Squiggle spider would likely have a lifespan measured

in seconds with current battery technology Even so,

in some applications, 20-30 seconds of operation

could be worth the cost of a swarm of insect-sized

microbots

On the topic of microsensors, with the exception of

Hall-effect devices, I haven’t seen any commercial sensor

offerings that come close to the level of miniaturization

required for an insect-sized microbot I’d like to have an

affordable ultrasonic or IR rangefinder comparable in

relative size to the Squiggle However, consider the

challenge in creating a suitable IR rangefinder withstandard components A typical IR LED alone is about thesize of an insect’s head And the available ultrasoundrangefinders require even more volume Clearly, when itcomes to microsensors for autonomous microbots, it’s timefor a new generation of SMT devices

Although autonomous microbots made completely ofcommodity — read affordable and readily available —components may be a few years away, there are myriadapplications of micromotors in other areas of robotics Themost obvious applications range from the manipulation ofcamera optics and R/C mini helicopter control surfaces, tocontrol of microvalves in implantable drug delivery devices

to surgical robots Although I expect to see the first large-scale applications of micromotors in the consumerelectronics industry, the medical applications will likely have the most profound effect on quality of life

Consider that current surgical robotics rely onstandard-sized motors connected to scalpels and otherinstruments through cables Although these roboticsystems enable surgeons to operate with greater efficiencyand effectiveness than traditional methods, because of thephysical arrangement of cables and instruments, theworking area is constrained to only a few inches across.The use of micromotors connected directly to instrumentswould allow for a much larger work area for tele-surgeons,

as well as lighter, mechanically simpler surgical robots Sizeand weight can be critical factors if the remote patienthappens to be an astronaut on Mars, or a critically injured

US soldier in a remote area of the world SV

Dear SERVO:

The “analog” servo block diagram, Figure 5, of theServo Buddy article in May 2008, is missing the velocityfeedback path from the motor to the local pulsegenerator Without this damping feedback, the servo willoscillate After the stretched drive pulse has ended, themotor back EMF is used to modify the next local pulse

In servos that use the NE544 IC, this feedback is from pin 9 to pin 1 via a resistor For the NJM2611 IC, frompin 11 to pin 15

It is interesting to note that years ago what isnow called an analog servo was called a digital servo.Back then, an analog servo required an analog VOLTAGEinput

— William J Kuhnle

RESPONSE: While I tried to keep the diagram simple,

it might have been good to include that Thanks for pointing it out.

— Jim Stewart

SchmartBoard Is Looking for

Beta Testers

New Website will be Social Network

for Electronics Enthusiasts

SchmartBoard is looking for people to beta-test a soon

to be opened web space call Solder By Numbers™

The website, which is due to launch in late summer,

will be a social network for electronics enthusiasts

SchmartBoard is looking for all levels of testers from

professional engineers to novices who have an interest

in electronics They are looking for people from around

the world

According to SchmartBoard’s VP of Sales &

Marketing, Neal Greenberg, “SchmartBoard is not yet

ready to reveal specific details about the website,

except that it is web 2.0 for electronics enthusiasts

Solderbynumbers.com will be a place to design and

build your electronic circuits while you create a worldwide

network of peers The site will be much more than a

social network It will be a place to collaborate, create,

communicate, and learn.”

To sign up to be a beta-tester, go to www.solder

bynumbers.com.

Trang 8

The Vulture Seldom Comes

Home to Roost

On a more celestial level, DARPA

is also funding a competition to

develop an unmanned aerial vehicle

that will shatter endurance records

The bird will draw 5 kW of power,

carry a 1,000 lb (450 kg) payload,

stay aloft for at least five years, and

remain in its assigned airspace 99

percent of the time while fighting

winds encountered at operating

altitudes, reportedly ranging from

60,000 to 90,000 ft (18,000 to

27,000 m) The goal is to provide

long-term intelligence, surveillance,

reconnaissance, and communication

missions over locations of interest

Contractors for phase one are

Aurora Flight Sciences (www.aurora.

aero), Boeing (www.boeing.com),

and Lockheed Martin (www.lock

heedmartin.com) A variety of

propulsion approaches — including

solar and internal combustion — will

be considered; however, nuclear and

lighter-than-air designs have been

ruled out The winning design must

comply with space — not aviation —

industry standards, because only a

“pseudo-satellite” will handle the

demanding requirements A

supervi-sory engineer at NASA observed,

“What you don’t want to build is afragile, expensive pain in the butt.”

The Aurora offering will be based

on its “Odysseus”

design, which uses solarpower during daylighthours and stored energy

at night It combinesthree “constituent aircraft” in a 500 ft(150 m), intriguing Z-wing configuration

Boeing is expected tofield a design based onthe existing British-builtZephyr high-altitude,long-endurance UAV,from partner QinetiQ

(www.qinetiq.com).

Lockheed Martin is stillmum on the subject

The competitors have 12 months

to come up with their initial designsfor DARPA review Phase two will endwith a three-month flight test of asubscale demonstrator, and the finalphase will require a 12-month test of

a full-scale vehicle

Mini Network Bots

Also pulling down governmentfunding — in this case, up to $3

million over three years from theDefense Advanced Research Projects

Agency (www.darpa.gov) — is iRobot (www.irobot.com) Under

the grant, the company will developthe LANdroid robot, a portable com-munications relay device According

to the contractor, “This robot will besmall enough that a single dismountedwarfighter can carry multiple robots,inexpensive to the point of being disposable, robust enough to allowthe warfighter to drop and throwthem into position, and smart enough

to autonomously detect and avoidobstacles while navigating in theurban environment.”

The objective is to enable networking in urban areas wherebuildings and other pesky objects can block wireless operations Inoperation, each of the little guys willwander around until it finds a goodspot to function as a node and thenjoin the rest of the swarm to form the network If one is destroyed, theothers will adjust their positions tokeep the system up and running

New Touch Technology

One of the perennial problems inrobotics is improving the machine’s

Aurora’s Odysseus design: A possible configuration of the Vulture UAV.

Photo courtesy of Aurora Flight Sciences.

Sneak peak at what the LANdroid robot will look like Photo courtesy of DARPA.

by Jeff Eckert

Trang 9

sense of touch, and what could be a

better way to solve it than to learn

from our touchy-feely friend, the rat?

Enter BIOTACT (BIOmimetic Technology

for vibrissal ACtive Touch, www.

biotact.org), a project funded by the

European Union and involving nine

research groups in seven countries

The goal is to emulate how such

mammals as rats and Etruscan shrews

can rapidly sweep their whiskers

back and forth to gather information

about their surroundings Thus, a bot

fitted with hundreds of whisker-like

sensors may be able to seek, identify,

and track fast-moving target objects,

even in poorly lit places where machine

vision doesn’t get you anywhere

The challenge is to develop new

biomimetic computational methods

and technologies that enable the

technology But the consortium

has been granted four years and

$11.6 million to do it, so the odds

look good

Showcase of Robotenders

It’s beginning to look like the

Germanic tribes have a curious

fetish about linking robotics with

such ostensibly unrelated fields as

sociology, philosophy, and art (see

last month’s Robytes) In this vein, the

upcoming 10th anniversary of the

RoboExotica conference recently

came to light According to the

event’s Vienna-based

creator (www.robo exotica.com), “Until

recently, no attemptshad been made topublicly discuss therole of cocktail robotics as an indexfor the integration

of technological innovations into thehuman Lebenswelt[environment], or

to document theincreasing occurrence of radical hedonism in man-machine communication.” Imagine that

But you can stop worrying, because

“RoboExotica is an attempt to fill this vacuum.”

RoboExotica generally consists

of a series of events (exhibition, conference, workshops, music, andfilm presentations) held at variouslocations in Vienna But this year,sometime after the December 4thkickoff in Austria, it will be presented

in San Francisco, as well, “thus

facilitating the already existingexchange of ideas between the WestCoast’s very much alive technology/art scene and the RoboExotica mother ship in Vienna.”

Unfortunately, the US incarnationwill not include the annual cocktailrobot awards, where you can enter amachine in one of five categories:serving cocktails, mixing cocktails,bartending conversation, smoking culture, and other achievements inthe sector of cocktail culture To participate, you’ll have to show up atthe Rote Bar/Volkstheater Wien

(www.volkstheater.at/rotebar html) The program is still under

development, so check the websitefrom time to time for details

Bot Assists Endoscopy

This month’s device for tauntingthe squeamish is EndoAssist, a roboticendoscope manipulator offered byProsurgics Ltd Used in invasivethoracic and abdominal surgery, it isparticularly useful for ardiothoracic,urological, bariatric, ob/gyn, andgeneral surgery Perhaps the mostinteresting feature is that the surgeoncontrols camera angles simply bymoving his head Glance left, and thecamera moves left, and so on You

R o b y t e s

Artist’s concept of the “ScratchBot”

employing the BIOTACT sensor.

Photo courtesy of the BIOTACT project.

The EndoAssist robotic manipulator Photo courtesy of Prosurgics.

A contestant from RoboExotica 2007.

Photo courtesy of Roboexotica.com.

Trang 10

can also pan, zoom, or modify the

view in any direction For a video

demonstration that may affect your

ability to keep lunch down, visit

www.prosurgics.com/prosurgics_

endoassist.htm.

Uribot Tends Kobe Airport

Finally, the strangest application

of robotics of late would be Dasubee,

a bot designed specifically to clean

urinals One is already operating in

the Kobe, Japan airport An astute

observer will note that it resembles

an elephant Designer Susumu Kanai

revealed that this design was inspired

by the pachyderm’s trunk, which

resembles the powerful water cannon

employed by the bot The ears are

handles, the eyes are the start and

stop buttons, and its little yellow hat

is the filler cap for the 13 gal (50 l)

tank Reportedly, usingspecially developedantibacterial detergent,Dasubee can shine up

a fouled privy in only

10 seconds

If you’re stillreluctant to buy one,consider that Kanaihas included “avacuum function tobreathe in a scraperand the water of thefloor to be able towash the dirt scattered

to the floor together

on the function side.”

(Something may havebeen lost in thetranslation.) You canpick one up for onlyone million yen (about

$9,500) SV

R o b y t e s

Dasubee, the urinal bot and its proud operator Photo courtesy of Impress Watch Corp.

Perform proportional speed, direction, and steering with

only two Radio/Control channels for vehicles using twoseparate brush-type electric motors mounted right and leftwith our mixing RDFR dual speed control Used in manysuccessful competitive robots Single joystick operation: upgoes straight ahead, down is reverse Pure right or left twirlsvehicle as motors turn opposite directions In between stickpositions completely proportional Plugs in like a servo toyour Futaba, JR, Hitec, or similar radio Compatible with gyrosteering stabilization Various volt and amp sizes available.The RDFR47E 55V 75A per motor unit pictured above.www.vantec.com

STEER WINNING ROBOTS

WITHOUT SERVOS!

Order at (888) 929-5055

Trang 12

When Lewis the Robot

Photographer first enters acrowded room, it gets atten-tion But, once people have adjusted

to its roaming around, looking hereand there, they forget all about it

After all, it’s just a machine, anotherobject in their environment

Lewis’ ability to blend in keeps

it from creating the kind of apprehension that comes with a live photographer who roams aroundsnapping candid pictures of people (as a wedding photographer might

do, for example)

Because Lewis captures people atease, it can take a much higher quality

of photos — no blinking, phony smiles,

or stiff or awkward poses BecauseLewis recognizes faces and quicklysnaps only the best photos, it takesmany more quality pictures in thesame period at gatherings, functions,and on special occasions

Looking for Faces

Lewis starts by scanning the roomfor pairs of what appear to be legs

This way, he can identify people andthen look up to find and identify their faces Then, Lewis uses face-recognition technology that identifiesparts of images with lots of skin tonesgrouped closely together

Lewis separates real faces fromthings that may look like faces to therobot eye Lewis eliminates anythingthat is too big, too low, or the wrongshape Anything left is assumed to

be a face

Lewis can take front-on and sideangle pictures of people It continuallyscans images for the criteria that predict a face or group of faces Once

it has detected a suitable image, itadjusts the camera to take a qualityphoto, moving it into position via

a series of zooming, tilting, and panning

The robot uses object avoidancetechnology to guide itself aroundobjects and people, and maintains itsposition within the mass of subjects

by recognizing a given object and centering itself in the group based onthe position of that object

Forming Pictures

How does Lewis form pictures offaces? By following rules One suchrule is the rule of thirds The rule ofthirds says that if you split a pictureinto thirds, first horizontally, then alsovertically, the primary point of visualinterest in the photo should be wherethe lines cross Lewis makes humanfaces the points of greatest interest,placing them at these cross points

Contact the author at geercom@alltel.net

by David Geer

Lewis, the Robot Photographer

At first brush, a robot that snaps people’s pictures might not imbue the mind with

a novel image But, a photographer that sets its subjects at ease, circumvents their shy and self-conscious natures and related facial reactions, and captures the essence

of the subject unawares, now that’s a wonder to see!

Full side view of Lewis.

Trang 13

Photographers try to avoid empty

space in their photos This helps

ensure that photos contain as much

relevant visual information as possible

Lewis weighs the rule of thirds and

the rule around empty space one

against the other whenever they

conflict, to take the best pictures

Lewis can also think for himself

when taking photos He is free to

break the rules altogether and take

feedback about his images He uses

this information to learn which rules

to break and when in order to deliver

great photos based on a sort of

photographic instinct

Live Test

Researchers tested the Lewis

robot photographer on a group of

5,000 subjects over a period of 40

hours Lewis took 3,000 pictures in

that period During this 40 hour run,

people (guests at a large technology

event) either ignored the robot

completely or tried to interact with it

Because the robot wasn’t instilled

with the ability to interact, people

quickly dispensed with it and began

socializing with other people in

the crowd Because people ignored

the robot, they relaxed and acted

naturally, enabling the robot to take

candid, natural pictures

Results

Among other things, researchers

determined that the robot should

have a sort of bi-modal capability If

someone is trying to interact with it,

it should stop what it is doing and

interact with those people, taking

their pictures where possible If no

one is trying to interact with the

robot, it should blend into the

background and continue to take

candid shots This version of the robot

is only capable of blending in So, the

robot will ignore people who want to

interact with it or who specifically look

to have their picture taken

People will be more likely to

interact with the robot on some level

if they know what it is up to What’s

it there for? In this version

of the robot it made a noise,sounding an alarm or signalwhen it had taken a picture However,the sound wasn’t loud enough formost people to hear

If the signal were louder, thiswould communicate to people in therobot’s proximity that it had just taken

a picture This would form some level

of communication between the robotand those people, and provide somesimple basis for interaction

The robot had no way of tellingpeople to hold still or say cheese Ittook four seconds for the robot to line

up shots, in which time people might

hold still to get their picture taken, orthey might move around After therobot’s test run, people suggestedthat the robot actually say cheese orshow a picture of a “birdie” (as in,look at the birdie) to signal that it wasabout to take a picture

People waiting in front of therobot hoping to have their picturestaken were often disappointed whenthe robot was navigating, getting itsbearings, or homing in on a landmarkinstead of taking pictures at that particular moment However, the

Sharing a more whimsical moment with Lewis are members of the Media and Machines Lab (from left):

Assistant Professor William Smart; Assistant Professor Cindy Grimm (seated): two of the lab’s founders, Shannon Lieberg, Engineering Class of ‘04 and research assistant Michael Dixon, B.S./M.A ‘03; and Nik Melchior, a fifth year B.S./M.S.

student in computer science and engineering Members are showing off the “playful props”

used by Lewis in the lab.

Lewis close-up head shot with camera Lewis has worked

photographing at a real live wedding reception.

GEERHEAD

Trang 14

robot was programmed to take

frequent pictures of human faces even

when it may not have had the

opportunity to focus in for a good

picture To better interact with

subjects, the robot should have the

ability to communicate which mode or

“state” it is in to the subjects So, if

the robot is available to interact, it can

communicate that, and so on

Photographic subjects expectedthe robot to respond when waved at,like a human being would However,the robot didn’t have this capacityeither When the robot did seem toreact — because it turned towardsomeone by sheer coincidence whensomeone had waved at it, for example

— people thought this meant it wasmore intelligent than it actually was

In particular, when the camerapointed in their direction coinci-dentally in response to trying tohail the robot, this was mistakenfor eye contact

Lewis seemed intelligent topeople when he did whatappeared to be a double take.Because the robot face detectioncode was not optimized, the cam-era panned past the faces andbeyond by the time the softwaredetermined it had detected aface The camera then returned tofocus on the face for the picture.This apparent “double take”humanized the robot in the eyes

of on-lookers, attracting people tointeract with the robot

Likewise, other robot behaviors made the robot appearnot so smart, even though thesebehaviors were quite intelligentfor a robot in what they

accomplished One such behavior was looking at the wall (pointing itscamera toward the wall) or movingalong a wall in order to aid in its navigation While the robot was trying to get its bearings, it appearednot to “see” anyone around it, and solooked dumb

Conclusion

Continued research based onLewis should address whether Lewistruly functions in two separate modes,whether the level of sophistication ofthe people interacting with Lewis has

an impact, and whether the robot orthe people around it should drive itsinteraction SV

Lewis, the Robot Photographer

www.cs.wustl.edu/MediaAnd Machines/Lewis

The Media and Machines Lab

Full frontal view of Lewis.

Here, Nik Melchior (left), a fifth year B.S./M.S.

student in computer science and engineering,

helps create the programming framework that

allows others to command Lewis Shannon Lieberg

(center) of the Engineering class of ‘04, works Lewis’

controls with Assistant Professor Bill Smart.

GEERHEAD

Trang 15

New breed of robots could soon

wander Antarctica

By GREG BLUESTEIN, Associated Press Writer

Robotic rovers have patrolled deep space and the deepest seas,

but scientists are still struggling to create drones that can

overcome the multiple challenges of exploring Antarctica.

Georgia Tech researchers think the SnoMote — a small robot

designed like a snowmobile — will be able to deal with the nasty

weather and with slippery terrain that constantly cracks and

shifts They envision dozens of SnoMotes roving Antarctica's

vast expanses to add to data already collected by satellites

and a handful of weather stations and sensors Ayanna

Howard, an associate professor at Georgia Tech in Atlanta,

has worked for two years under a NASA grant to perfect the

two-foot-long robots.

Her initial designs with spider-like legs proved too

cumbersome to navigate snowbanks So, she and her colleagues

leaned on others' designs, outfitting a snowmobile designed for

kids with sensors, gauges, and cameras, and then programming it.

She developed a program that lets the SnoMotes negotiate with each other and “bid” on which site to investigate, allowing them to decide for themselves how to dole out their assignments The next challenge, though, was to come up with navigation for the rovers Other probes tend to use distinguishing characteristics like rocks to chart their paths But such features can be hard to come by in vast icy expanses.

On a field trip to a Colorado glacier, Howard's team discovered they could use microscopic fissures in the ice and snowbanks to guide their way.

“If you can come up with a way to classify these uniquely, you can come up with a way to navigate,” she said.

Simulations so far have proved her team's formula effective, but plenty of challenges await when the robot is put to the test

on the glaciers of Alaska.

With Penn State University researcher Derrick Lampkin, Howard has designed a shell that weighs 60 to 70 pounds, can withstand harsh winters, and eventually could include heaters to keep computers and wiring running in the cold.

Lampkin said his goal is to develop a "scale-adaptable, autonomous, mobile climate-monitoring network."

The researchers hope the robots will ultimately cost around

$10,000, relatively cheap for governments, researchers, and others seeking to document changing conditions in the world's most remote places.

The more the better: Howard said in order for scientists to say with certainty how climate change is affecting the ice, they need plenty of accurate data points to create climate models She envisions a field of 40 to 50 of the SnoMotes wandering icy plains, a small army gathering data to shed light on global warming and other quandaries without breaking the bank.

“The whole concept is: How do you do this in the most affordable way?” she said.

3DLQH6WUHHW_3RZD\_&DOLIRUQLD___ZZZKLWHFUFGFRP

>_j[Y Ej^[h8hWdZ

.EW

+656*

6SHHGVHF 7RUTXHR]LQ 6WHHO*HDUV

+656*

6SHHGVHF 7RUTXHR]LQ 6WHHO*HDUV

+657*

6SHHGVHF 7RUTXHR]LQ 7LWDQLXP

$//63(&,),&$7,216$792/76

The Mechanicrawl • Saturday, July 12, 2008

Spend a summer day exploring the mechanical marvels along

San Francisco's North Shore! See giant running steam engines, turn

of the century automata, mechanical computers, an eight foot high

mechanical planetarium, and more You'll be able to map your own

route for the event and spend as much time at each location as you'd

like You can walk, bicycle, or use public transport for Mechanicrawl;

maps, routes, and additional info are listed under Map Your Crawl on

the website For all the details, go to www.mchanicrawl.com.

Trang 16

Q.Every once in a while there is a question floated

around in my local robotics group that (I think) is

such an epiphany (read: slap forehead) that I think

it deserves a wider audience This first question is one of

this nature:

Does anyone have any recommendations for an

inexpensive DC-DC converter that has 11-14 VDC in, 5 VDC

out (up to 3-5 amps)? I need a heftier supply for the

low-level control system on my robot Currently, I’ve got a 12V,

12 Ah lead-acid battery with a 5V-1A converter, and I’m

reaching the limit of the converter when I add my sonars

and IRs later, I would prefer an off-the-shelf solution

— Daniel Herrington

A.My first thought was the TI PT78ST105, which is a

1.5 amp 5V switching regulator that works with up to

38V This nifty part has the same pin-out as the

venerable 7805 regulator and is way more efficient.

Another suggestion was the TI PTN78020 which has similar

input voltage maximums and a 6 amp output with high

efficiency since it too is a switching DC/DC converter

However, this part has lots more pins; it still needs no

external components These parts are easily found at places

like Mouser, pricing depends upon various options As good

though, as these solutions are, they are not “off-the-shelf”

and would need a circuit board to use Don Clay put forward the solution of using a BEC that the R/C airplanehobby crowd commonly uses on electric aircraft that uselarge battery voltages This is a very cool idea because itcan be connected to the robot’s main battery and will efficiently give the needed 5V in a small, self-containedunit, and it already has easily usable cabling The one thatDaniel selected was the Castle Creations CC BEC which sells

for about $22 at good ol’ HobbyTown USA (www.hobby town.com); see Figure 1 This device is very useful because

it can be set to output voltages from 4.8V to 9V by usingthe Castle USB link adapter (not included) In case you werewondering, BEC stands for Battery Eliminator Circuit In the

“old days,” electric R/C cars had a battery for the motor andanother for the R/C electronics The BEC “eliminated” one

of those batteries

Q.I want to send commands to my robot using an

IR remote I don’t want to build another IR remote;

I want to use one of the bazillions that I have lyingaround the house How do I use these? How can I decodetheir output?

— George S.

A.Oh boy, I feel a marathon answer

coming up! I’m not going to go intohuge detail about every kind of IRremote out there — there are a ton of webpages that you can Google and find thosekinds of details I’ll provide a selection ofthem at the end of this answer for those curious though I did not find a lot of pagesthat connected the dots between the variousformats and how to write a program to readthem, either So, in this answer I’ll provide thenitty gritty details of how the most popular IRcodes are created and provide some PIC codethat allows you to decode them First, theugly details there are many, MANY different

Tap into the sum of all human knowledge and get your questions answered here! From software algorithms to material selection, Mr Roboto strives to meet you where you are — and what more would you expect from a complex service droid?

Trang 17

IR encoding schemes out there Wikipedia tells us that there

are literally hundreds of IR protocols! Fortunately for us, there

are three that are by far the most common: SONY SIRC,

NEC, and Phillips RC-5 If you pick up any odd remote at

your house, it will most likely be using the SONY or NEC

formats Let’s discuss each of the “big three.”

SONY SIRC Format

First off, here are a couple of links that discuss this

format There is a lot of useful information in these; I

disagree with a few things that they say (more on that later),

but since some of the data that I used to write my code

came from these links, most of what they say is accurate:

www.hifi-remote.com/infrared/IR-PWM.shtml

This site gives a lot of information on how to read the

universal IR remote entries that you use to program your

universal remotes The author calls this “PWM,” or pulse

width modulation It is more accurately called “PPM,” or pulse

position modulation, rather like R/C radio communications and

what we deal with when controlling hobby servos

www.edcheung.com/automa/sircs.htm

This site gives some hints on how to write code to send

IR signals Not what I was looking for, but you might be

interested in this if you wanted to have a robot wander into

your TV room and take over control of the TV set!

The SONY format has three timing values that we are

interested in, shown in Table 1 The Lead In is the header

before the data that tells us that the code is coming

Everyone says this is to set up the AGC (automatic gain

control) in the receiver, but quite frankly, if it didn’t look

different from the rest of the transmission how could we

tell that a new command was coming? The rest of the data

is an asynchronous data stream of 1s and 0s like the serial

data in RS-232, but coming in over modulated IR radiation

instead of a pair of wires When I say “On,” I mean that the

IR carrier is detected An “Off” means that there is no IR

detected by the sensor

The actual SONY IR command can come in three

variations: 12 bits, 15 bits, and 20 bits In every variation,

we first have the Lead In header, then we have a seven-bit

command Next, in the 12-bit protocol we have a five-bit

Device ID The Device ID tells us what is being controlled,

for instance, a TV or a DVD In the 15-bit format, the

Device Code is eight bits; in the

20-bit code, we have a five-bit

Device ID and then an eight-bit

Extended Command byte All

of my remotes were either

12-bit or 15-bit All of the

codes come in LSB first This

means that the Least

Significant Bit is first and the Most Significant Bit (MSB) islast The packet format includes some kind of a Lead Out aswell, but it is really just a guarantee of some time spacingbefore the next packet is sent Don’t bother to look for itsince its length depends upon the data just sent; it doesn’tseem to be consistent (see Packet 1)

When you code for this protocol, you need to remember to assemble the bytes by putting the bits inreverse order! Remember, they are LSB first

When you press a button on a SONY remote, it willsimply repeat the command packet every 45 ms for as long

as you hold the button Some buttons on my remoteswould only send a command once when you let up off ofthe button, but the rest just kept repeating endlessly

The NEC format is a bit different It includes some errorchecking and has a different way of dealing with repeatingpackets The NEC protocol sends a Lead In, 32 bits of data,and a Lead Out Since we can count on the Lead In and 32bits of data, don’t bother to look for the Lead Out hereeither The NEC format sends its data LSB first like SONY,

[Lead In] [0 | 1 | 2 | 3 | 4 | 5 | 6| 7] [0 | 1 | 2 | 3 | 4 | 5 | 6| 7] [0 | 1 | 2 | 3 | 4 | 5 | 6| 7] [0 | 1 | 2 | 3 | 4 | 5 | 6| 7] [Lead Out]

12 Bit Code [Lead In] [0 | 1 | 2 | 3 | 4 | 5 | 6] [0 | 1 | 2 | 3 | 4]

Command Device ID

15 Bit Code [Lead In] [ 0 | 1 | 2 | 3 | 4 | 5 | 6] [0 | 1 | 2 | 3 | 4 | 5 | 6 | 7]

20 Bit Code [Lead In] [0 | 1 | 2 | 3 | 4 | 5 | 6] [0 | 1 | 2 | 3 | 4] [0 | 1 | 2 | 3 | 4 | 5 | 6 | 7]

Command Device ID Extended Command

Description ON Time OFF Time

Lead In 2.4 ms 0.6 ms Logic 1 1.2 ms 0.6 ms Logic 0 0.6 ms 0.6 ms

Table 1 SONY timing values.

Description ON Time OFF Time

Lead In 9 ms 4.5 ms Logic 1 0.56 ms 1.68 ms Logic 0 0.56 ms 0.56 ms Repeat 9 ms 2.25 ms

Table 2 NEC timing values.

PACKET 1

PACKET 2

Trang 18

but there the similarity ends The packet format looks like

that in Packet 2 By ~Device ID and ~Command, I mean

that it is inverted, or 1’s complement of its respective

Device ID and Command To check for a proper reception,

all you need to do is AND the byte with its respective

complement; if it comes up all zeros, then you have a good

reception When you hold down a button on the NEC

remote, it does not send out the same command over and

over; it sends out a special signal called the Repeat Code

The NEC protocol has four timing values that we care

about, and one we don’t (the Lead Out) Table 2 shows the

ones that we pay attention to

Phillips RC-5 Format

Some useful information on the Phillips RC-5 format

can be found here:

a logic ‘0’ and vice-versa is shown by a change in the bitphase If you are like most of us, your eyes just glazed over

at that description This is one time that a picture is prettymuch needed to explain what I mean (see Figure 2).Note that it isn’t the timing that shows what the bit is,but rather the phase of the timing But how — you ask — doyou know what a ‘1’ is and what a ‘0’ is? You know becausethe bit stream of RC-5 starts out with two 1 bits and then atoggle bit, which on the first press of the button is a 0 Figure

3 shows how a transmission is formatted An RC-5 packetconsists of the preamble of 1 1 0, then a five-bit address andthen a six-bit control RC-5 packets are encoded with the MSB(Most Significant Bit) first Because you know that the firstbit is a 1, then any bit transition from that point onwardyou can track to know what the current bit is — a 1 or a 0.Figure 3 may look a little odd because I put both sameand switch phases in every bit cell except for the start bits.Just to make it more confusing though, I’ve read that thestart bits can be either 1 1 or 1 0 I hope not; it’d make ithard to figure out the 1 startup The RC-5 protocol willrepeat the button press every 113.8 ms, but every packetafter the first one will have the Toggle bit toggled differently than the last start bit

How to Decode IR Transmissions

Okay, now that the explanations are over with, how

Figure 2 Bi-Phase encoding.

Figure 3 RC-5 packet.

Figure 4 IR detector schematic.

Part Description

C1 110 μF 25V capacitor C2-C4 1 μF 25V capacitor R1 10K 1/4W 5%

R2 2.2K 1/4W 5%

R3 1K 1/4W 5%

D1 Green LED S1 N.O button U1 PIC16F688 DIP U2 7805 regulator U3 PNA4602 IR demodulator J1 Power connector J2 Four-pin male 1” connector J3 RJ-11 six-pin connector

Table 3 IR detector parts list.

Trang 19

can we use the IR transmissions from our remotes in our

robots? To see what is going on, I built a special IR

decoder board that has a Panasonic PNA4602 (38 kHz) IR

demodulator, a PIC16F688, a power plug, regulator,

programming header (Microchip ICD2 six-pin type) and a

plug for the Acroname RS-232 converter board Figure 4

shows the schematic and Figure 5 is a photo of the finished

board with everything plugged in This board was wired

together with a bunch of wire that I had lying around, so it

isn’t very sensitive to how you put it together I just wanted

an experimenter board that would show me the IR

codes and allow me to experiment with decoding the IR

transmissions Different remotes use different modulation

frequencies, but the Sharp PNA4602 will work with

frequencies between 36 kHz and 40 kHz just fine I chose

the PIC16F688 because it had an interrupt line, a hardware

USART, two timers, and an internal RC oscillator that runs

at 8 MHz, all in a 14-pin package These were all of the

hardware interfaces that I needed Unfortunately, for some

reason Microchip requires some kind of dongle to debug

this chip which I didn’t have, so I debugged using “printf”

statements in the code

The board isn’t particularly sensitive to the components

that you can use My previous list just came from my parts

bins The power connector was a barrel socket that fit the

various wall warts that I had lying around I chose the pin

out of the RS-232 connector to match the Acroname Serial

Interface Connector that you can get from www.

junun.org/MarkIII for about $10 Finally, the

programming connector is chosen to fit the cable on

my Microchip ICD2 so that I can just plug in and program

the board without constantly taking the part out of a

programmer board and plugging it back into my test board

If your programmer allows In Circuit Serial Programming

(ICSP), then get a connector that matches its cable

Figure 5 shows what my board looks like I put little

rubber feet on the bottom to make it extra spiffy looking I

put a small three-pin socket on my board so that I could

swap around IR demodulators that have the same pin-out

as the Panasonic units, should I so desire

Now that we understand the formats and we have a

board to look at the signals, how can we decode them?

The answer is software, of course I like to use C as my

programming language, but the logic that I use here can be

ported to any compiler language that you feel comfortable

with I should qualify that statement — any compiler that

allows interrupts to be used is required In this case, my

compiler of choice is the CCS C compiler for the Microchip

14-bit cores (the PIC16Fnnn series.) I am not going to go

into the gory details about how to program a PIC; whole

books have been written on that subject and that isn’t my

intent with this column However, you can benefit from my

line of reasoning for doing things how I did them and move

these procedures to your compiler or even your other

microcontroller if you wish I’m only going to show SONY

and NEC decoding in my code This is because I don’t have

any RC-5 remotes; all of mine were SONY and NEC formats

My test code has two sections in it The first is the

set-up of the interrupt routine that will capture the timesbetween pulses by measuring between falling edges on the INT line When the IR transmitter is On, this will bedetected by the PNA4602 and it will drop its output low.This is why I chose falling edges; the output of the device

is normally high, or Off Before I go into my logic for detecting pulses and parsing them out, let’s get an idea ofhow I set the PIC up to look for these transmissions To dothat, let’s look at how to initialize this PIC The code snippetbelow shows how I set up the timers:

//Turn off comparator setup_comparator(NC_NC_NC_NC);

//We will be doing interrupts setup_timer_0(RTCC_INTERNAL | RTCC_DIV_128);

//16ms timer setup_timer_1(T1_INTERNAL | T1_DIV_BY_4);

//Gives 131ms max timeout ext_int_edge(H_TO_L);

//IR IRQ on falling edge

CCS does a good job of hiding the ugly details of thehardware from you, but you still need to read your datasheets to understand what they are hiding! Here we turnoff the comparators because these pesky things default to

on in the PIC and will get in the way of those I/O lines.We’re going to use TMR0 to time our pulses because it

is an eight-bit timer and we don’t want all that much resolution in our times This “slop” allows us to quicklycheck a bit time and save it away; the coarseness of themeasurement allows us to read remotes that are close, butnot perfect to the specified times We know that the pulses

in SONY and NEC are between 1.2 ms and 13.5 ms (seeTables 1 and 2 again) so we want this timer to have a maximum time before it rolls over that is near that maximum time Since our PIC is running on its internal

8 MHz clock (which is divided by four internally), we knowthat if we use the prescaler on TMR0 at divide by 128, wewill have a maximum time of:

(1/(2MHz/128))*256 = 16.384 msFigure 5 IR detector board and Acroname serial connector.

Trang 20

That is pretty close to 13.5 ms Why are we using

TMR1? When we start saving times to measure for our

data bits, we need some way to stop looking if the signal is

interrupted, otherwise we’ll just appear to “hang” and do

nothing TMR1 will be set to values that are just a little

larger than the inter-packet repeat rate For NEC, this is

108 ms; for SONY, it is 45 ms TMR1 is a 16-bit timer and it

only has four prescales to choose from, so I took the one

that got me close to 108 ms Finally, I set the INT interrupt

to trigger on a high-to-low transition for my measurements

Here is the code for the actual interrupt service routine:

#int_global

void isr(void)

/*

* Lets handle all ISR save/recovery functions, the

default isn’t lean enough for

* an ISR that is highly time critical.

//We only have two interrupts, so if it isn’t

this one, it’s the other.

if (bit_test(PIR1,0)) //Timer1 overflow IRQ

{

bits[wBits++] = 255; //timed out bit_clear(PIR1,0); //Clear TMR1 int flag bit_clear(T1CON,0); //Turn TMR1 off until

bit_set(T1CON,0); //turn TMR1 back on so

//we can time out TMR0 = 0; //clear out the timer }

if (wBits == MAXBIT) //rollover the bit

//buffer wBits=0;

to be saved during an ISR call This means that we need to

do it Really, the only reasonable way to do this is withsome simple assembly code This function needs to be FAST and we do that by keeping it lean I combine this lean ISR with the definitions at the top of the program thatlooks like this:

unsigned char save_w; //These next 3 bytes

//are saved on interrupt

#locate save_w=0x7f unsigned char save_status;

#locate save_status=0x7e unsigned char save_FSR;

#locate save_FSR=0x7d unsigned char wBits=0; //To make access to

//these variables fast

#locate wBits = 0x7c //Keep them in common

//memory for ISR use unsigned char t1h;

#locate t1h = 0x7b unsigned char t1l;

#locate t1l = 0x7a /*

* Give me direct access to several SFR’s that CCS doesn’t handle the way I want.

Now look at the ISR from above When we get aninterrupt from INT, we record the value in the TMR0 register

in the next available data slot in our ring buffer (called ringbecause it rolls over to 0 when it reaches the end of the

“ring”) If we time out on TMR1, then we stuff a 255 intothe buffer telling the decoding routine that we timed outand we should ditch the entire set of numbers and wait forthe next start to be detected

The second part of the program will decode the timessaved in the buffer and turn them into 1s and 0s It uses itsown pointer into the ring buffer and knows when to lookfor a value when the read pointer is different from thewrite pointer I’ve put lots of comments into this code, soI’m not going to go over it line-by-line Here is what the bitdecoding routine looks like:

Trang 21

unsigned char DecodeBit(unsigned char time)

/*

* This will determine if the bit is a 0 or a 1,

* by whichever standard is in being used

* Returns a 1 if a logic 1 bit, 0 otherwise.

*/

{

unsigned char ret = 0;

unsigned char val = 0;

val = time; //Can be used to “fuzzy”

//the time for rounding

if (whichOne == NEC)

{

if ((val <= NEC_ONE+FUZZY) &&

(val >= NEC_ONE-FUZZY)) ret = 1;

else if ((val <= NEC_ZERO+FUZZY) &&

(val >= NEC_ZERO-FUZZY)) ret = 0;

There is more magic in DecodeCode() that shows how

to recognize the start of a transmission and how to end

one This little routine above simply shows the decoding of

a single data bit Notice the + and – FUZZY settings IR

specs allow for about a 10% slop in the standard times

This FUZZY setting gives us that You can experiment with

how large you want that FUZZY to be since it is a #define

at the top of the program I’ve found that a setting of 2

works well

My code will “auto” detect NEC

code if you hold the button down

This is because NEC uses the repeat

code frame that is distinct from any

other transmission You don’t have to

do that if you don’t want to, and

you’ll need to reset the PIC to get it to

pay attention to SONY codes again

regardless I have two modes of

operation that can be selected by the

setting of the TEST define If this is set

to 1, then the program will only print

out the times This is useful for when

you are trying to understand a new

format If TEST is set to anything else,

then the program will decode either

SONY or NEC, and print out the

device and control codes when they

are received The entire program

source can be found on the SERVO

website www.servomagazine.com).

It is called IRdecoder.c

Conclusion

Figure 6 shows my collection of IR remotes that I used

to test my program I found interesting departures from theestablished standards in some of them; no doubt you will, too.I’ve given you a powerful tool that you can use to discover IR codes and a basic template that you can use

to embed the ability to control your robot using a commonhousehold device: the IR remote Have fun and be creative

If you have any questions about this program or how I

“figured it all out,” send your questions to roboto@servomagazine.com — I’m happy to answer! SV

NEWS FLASH! At the last possible moment I discovered

this site It is an excellent compendium of various IR

remote formats: www.rhoads.nu/bjorn/hp48/remote.

Figure 6 A selection of IR remotes.

Trang 22

Know of any robot competitions I’ve missed? Is your

local school or robot group planning a contest? Send an

email to steve@ncc.com and tell me about it Be sure to

include the date and location of your contest If you have a

website with contest info, send along the URL as well, so we

can tell everyone else about it

For last-minute updates and changes, you can always

find the most recent version of the Robot Competition FAQ

at Robots.net: http://robots.net/rcfaq.html

— R Steven Rainwater

J uly

7-11 Africa Championship Robotics Competition

Pretoria, South Africa

Students from various countries and continentswill compete in several robot challenge

events

www.nydt.org/home.asp?pid=9638-10 European Micro Air Vehicle Competition

Research Airport, Braunschweig, Germany

Tiny, autonomous flying robots compete forprizes Every year, these things get smaller and smaller

www.mav08.org8-11 Botball National Tournament

Norman, OK

Educational robot contest for middle and high school students designed to use science,technology, engineering, and math to solve real world problems

www.botball.org13-17 AAAI Mobile Robot Competition

www.livingjungle.com22-25 FIRA Robot World Cup

Shinan Software Park, Qingdao, China

This competition has events for every kind ofrobot soccer imaginable, ranging from thehumanoid robot league down to the tiny Khepera robot league

www.fira.net

26 RoboBombeiro

Polytechnic Institute of Guarda, Guarda, Portugal

Autonomous fire-fighting robot contest

http://avdil.gtri.gatech.edu/AUVS/IARC LaunchPoint.html

29 AUVS International Underwater Robotics

Suzhou, China

Soccer Simulation — teams demo and test theirrobots; Small-size Robot Soccer — F180 robotsplay soccer; Mid-size Robot Soccer — larger robotsplay soccer; Sony Legged Robot Soccer — legged

Send updates, new listings, corrections, complaints, and suggestions to: steve@ncc.com or FAX 972-404-0269

Trang 23

robots play soccer; RoboCup Junior — small robotsplay soccer; Humanoid Soccer — humanoid robotsplay soccer; Rescue Robots — NIST StandardRescue Robot Test Field; RoboCup@Home — realworld robot event.

http://www.robocup.orgTBA War-Bots Xtreme

Saskatoon Saskatchewan, Canada

“Robots” (RC vehicles) attempt to destroy eachother

The Science Place, Dallas, TX

Autonomous robots demonstrate their talents

www.dprg.org/competitionsTBA Robot Fighting League National

Minneapolis, MN

“Robots” (RC vehicles) attempt to destroy eachother

www.botleague.comTBA Robots at Play

City Square, Odense, Denmark

Robots compete to demonstrate playfulness andinteractivity

Extreme Robot Speed Control!

6 Dual 2.5A (6A pk) H-bridges

6 Plus 12A fwd-only channel

6 5V - 18V

6 1.6“ x 1.6” x 0.5”

$79.99 Scorpion Mini

6 2.5A (6A pk) H-bridge

6 14V - 50V - Dual 80A H-bridges - 150A+ Peak!

6 Adjustable current limiting

6 Temperature limiting

6 Three R/C inputs - serial option

6 Many mixing options - Flipped Bot Input

6 Rugged extruded Aluminum case

6 4.25" x 3.23" x 1.1”

e Sidewind r

$29.99

$399

Introducing Dalf

6 Closed-loop control of two motors

6 Full PID position/velocity loop

6 Trapezoidal path generator

6 Windows GUI for all features

6 Giant Servo Mode!

6 C source for routines provided

6 PIC18F6722 CPU

25

$ 0

MADE IN THE USA

H-bridges: Use with Dalf or with your Micro/Stamp

Trang 24

Robotic Operations Result in

Significant Data Collection

from Mars

IEEE (Institute of Electrical and

Electronics Engineers) has named

Paul Backes, Eric T Baumgartner, and

Larry Matthies recipients of its 2008

Robotics and Automation Award The

three are being recognized for their

contributions to different robotics

technologies used in space flight

systems including the successful Mars

Exploration Rover (MER) mission

rovers Spirit and Opportunity, which

to this day are still functioning on

the surface of Mars The IEEE is the

world’s leading professional

association for the advancement of

technology

The award, sponsored by the

IEEE Robotics and Automation

Society, recognizes Backes,

Baumgartner, and Matthies for

contributions to robotics enabling

effective autonomous operations of

science investigations under extreme

conditions on the planet Mars The

award was presented to the three

on May 23, 2008 at the IEEE

International Conference on Robotics

and Automation (ICRA) in

Pasadena, CA

The works of Backes (distributed

and remote operations),

Baumgartner (manipulator control),

and Matthies (navigation systems)

have advanced robotic technology,

particularly rover operations, and

made possible the scientific

exploration of Mars MER is the first

long-term mobile autonomous

robotic exploration in an unknown

space environment

An IEEE member, Backes is the

technical group supervisor of the

Mobility and Manipulation group in

the Mobility and Robotic Systems

section at the Jet Propulsion

Laboratory of the California Institute

of Technology in Pasadena Heconceived and led the development

of an interface system to allowscientists and engineers tocollaborate in generating activitysequences, which was used as theprimary science planning tool in the

2003 MER mission The interface alsoenables the public to view missiondata and simulate their own activitysequences Backes holds sevenpatents, has won several awards,and has published numerous bookchapters, articles, and papers He

was associate editor of the IEEE Robotics and Automation Society Magazine from 1993 to 1998.

Baumgartner contributed tothe MER project as the lead systems,test, and operations engineer forthe MER Instrument PositioningSystem This system was responsiblefor the robotic deployment andplacement of four in-situ — meaning

“in place” — instruments onto theMartian surface through the use

of a five degree-of-freedom roboticarm Presently, Baumgartner is thedean of the T J Smull College ofEngineering at Ohio NorthernUniversity in Ada He has publishednumerous papers in the area ofmobile robotics and vision-guidedmanipulation and has receivedseveral awards for his efforts on theMER project

Matthies’ work on autonomousnavigation of robotic ground andair vehicles led to the development

of algorithms for descent motionestimation, visual odometry, andreal-time 3D perception with stereovision These capabilities wereincorporated into the MER mission,providing landers with the ability toestimate horizontal velocity androvers with the ability to detectobstacles and measure slip Hiswork can be found in terrestrialapplications including off-roadautonomous navigation and roboticvision systems An associate member

of the IEEE, Matthies is an adjunctprofessor at the University ofSouthern California and a member

of the editorial boards of the

Autonomous Robots Journal and the Journal of Field Robotics He has

received several awards, holds twopatents, and is widely published

Blo w Out Special

$9.95!

WWW.SERVOMAGAZINE.COM

Trang 25

Ask for our FREE 96 page catalog

VISIT OUR ONLINE STORE AT

www.allelectronics.com

WALL TRANSFORMERS, ALARMS,

FUSES, CABLE TIES, RELAYS, OPTO

ELECTRONICS, KNOBS, VIDEO

ACCESSORIES, SIRENS, SOLDER

ACCESSORIES, MOTORS, DIODES,

HEAT SINKS, CAPACITORS, CHOKES,

TOOLS, FASTENERS, TERMINAL

STRIPS, CRIMP CONNECTORS,

L.E.D.S., DISPLAYS, FANS,

BREAD-BOARDS, RESISTORS, SOLAR CELLS,

BUZZERS, BATTERIES, MAGNETS,

CAMERAS, DC-DC CONVERTERS,

HEADPHONES, LAMPS, PANEL

METERS, SWITCHES, SPEAKERS,

PELTIER DEVICES, and much more

Trang 26

Baby Orangutan B-168

Robot Controller

Pololu announces the release of

the Baby Orangutan B-168 robot

controller, the latest addition to Pololu’s line of

Orangutan robot controllers The compact module

has dimensions of 1.2” x 0.7”, and it can be configured

to fit in a solderless breadboard or a 24-pin dual in-line

package (DIP) socket For applications with low I/O usage,

the Baby Orangutan B-168 board can also be configured

with pins on just one side of the module for use as a single

in-line package (SIP) The diminutive size of the Baby

Orangutan B-168 makes it well suited for primary control of

miniature robots or for auxiliary control on larger robots

The Baby Orangutan B-168 is based on an Atmel

ATmega168 microcontroller running at 20 MHz with 16

KB of Flash program memory and 1 KB data memory

The use of the ATmega168 microcontroller makes the

Baby Orangutan B-168 compatible with the popular

Arduino development platform Free C and C++

development tools and libraries are also available

Integrated motor control sets the Orangutan family

of controllers apart from other small microcontroller

boards, and the Baby Orangutan B-168 features dual

high-performance, MOSFET-based H-bridges to deliver up

to 1A per channel over the 5-13.5V operating range

With hardware-based ultrasonic PWM generation, two

independent, bidirectional DC motors can be controlled

symmetrically and without any processor overhead

The unit price is $29.95

For further information, please contact:

PC Windows USB Inter face for

OWI-535 Robotic Arm

The robotic arm interface kit available from Images

connects OWI’s 535 Robotic Arm Edge™ to a personal

computer (IBM PC or compatible) The interface connects

to the PC’s USB port The software for the interfacepermits real time control and contains a built-in interactive script writer A user may write a script thatcontains up to 99 individual robotic arm functions(including pauses) into a single script file Script filesmay be replayed automatically for demonstratingcomputer controlled automation and animatronics.The Robotic Arm PC Interface creates a fun way oflearning and experimenting with computer automationand animatronics

The USB OWI-535 Interface (assembled and tested)costs $99.95; the USB OWI-535 Interface Kit (requiressoldering and assembly) is $84.95

The Interface Kit Includes:

• Windows 2000/XP/Vista program

• Printed circuit board for easy construction

• All components

For further information, please contact:

Motor Mount and Wheel Kit

It’s time to give your robot the mobility and style itdeserves with the new Motor Mount and Wheel Kitwith Position Controller from Parallax Powerful 12 VDC,

150 RPM motors are combined with precision machined

6061 aluminum hardware to provide enough power,strength, and beauty to make other robots jealous

Fax: 702•262•6894 Email: www@pololu.com Website: www.pololu.com

Pololu

Corporation

Trang 27

Conveniently positionedscrew holes in the bearingblock make mounting thiskit a breeze, and theincluded six inch pneumatic rubber tiresperform well on a variety ofsmooth or rugged terrains Thekit includes two position controllers whichuse a quadrature encoder system to reliablytrack the speed and position of each wheel with

36 positions/rotation resolution and report the data

on command via a 19.2 kbps serial bus The position

controllers can also be interfaced with HB-25 motor

controllers (sold separately) to automatically provide

user-definable smooth speed ramping and accurate position

control, which frees up the main processor to handle

more important tasks

The entire Motor Mount and Wheel Kit (#27971) is

a value at $279.95

For further information, please contact:

Flowcode for ARM

Microcontrollers

Matrix has recently

launched a newversion of their popular

graphical programming

language for

micro-controllers — ‘Flowcode

for ARM

micro-controllers.’ Now, 32-bit

ARM microcontrollers

are available for the

same price as eight

bit micros but offer

massive advantages to

developers: low power,

more I/O lines, several

times more ROM and

RAM than a typical eight bit micro, full floating point

and maths libraries, and a massive increase in processing

speed and power

This new version of Flowcode provides

engineers and developers access to all of these features

of the ARM based on Atmel’s popular range of AT91

microcontrollers Flowcode for ARM is also backwards

compatible with Flowcode for PICmicro®

microcontrollers and AVR® microcontrollers which

provides an easy migration route to 32 bit power

ARM hardware development tools, based on the Matrix’s E-blocks range, are also available A fully functional demonstration version is available on theMatrix website

For further information, please contact:

Four-Channel Digital Battery Test System

The Cadex C8000 is an advanced battery test systemcapable of performing complex lifecycle tests Thesetests may include discharging a battery with GSM,CDMA, or other pulses of choice Replicate battery runtime of a power tool, digital camera, or computingdevice by first capturing the current profile and thenapplying the load on the test battery for verification TheC8000 can also test the function of a Li-ion charger, verify battery safety circuits, and read SMBus registers.Automated programs assure safe charging and correctdischarge terminations; custom programs provide foruser-defined settings Each of four independent channelsdelivers up to 10A and 36V, with 0.1% FSR Total power

is 400W on charge and 320W on discharge The CadexC8000 runs as a stand-alone unit or with PC-BatteryLab™software

For further information, please contact:

SOFTWARE

TOOLS & TEST EQUIPMENT

Is your product innovative, less expensive, more functional,

or just plain cool? If you have a new product that you

would like us to run in our New Products section, please

email a short description (300-500 words) and a photo ofyour product to:

Website: www.cadex.com/c8

CadexElectronics

Trang 28

Featured This Month:

ROBOT PROFILE – Top

Ranked Robot This Month:

31 Billy Bob by Kevin Berry

Idesigned this 12 pound,Hobbyweight, robot with thepurpose of cramming as muchpower as I could into the smallest

box possible It’s low, it’s fast,and it’ll just keep slamming itsface into your spinning weaponuntil something breaks The horns

on top allow it totake its opponentsinto the wall of the arena whilewearing them like apolished metal hat.Apollyon won itsdebut tournamentthis past Decemberwith a 3-1 record

Trang 29

Figure 1 is a CAD model of

Apollyon drawn in SolidWorks Some

minor changes were made, but most

of the parts are almost identical to

the ones in the drawing, for example,

the side and internal frame rails, as

well as the horns that are used to

catch other robots The horns are

steel and the rest of the frame is

6061 aluminum (see Figure 2)

The sides of the frame are

lined up to show the profile of the

chassis, shown in Figure 3 The

holes in the top, front, and bottom

of the plate are tapped to allow

armor to be bolted directly to them

Figure 4 shows the chassis with

the baseplate, drive motors, and

inter-nal portion of the front armor installed

Figure 5 shows a mostly

assembled Apollyon sitting inside

the chassis of my 60 lb robot Ruiner

In Figure 6, Apollyon has been

painted, the front steel wedge has

been mounted, and the robot isalmost ready for competition

Take a look at Figure 7

You’ll see an internal shot ofApollyon the night before the

“Wreck the Halls” event inGreensboro, NC This showsthe layout of the electricalcomponents and the batterymount The piece of PVC pipe

in the back left of the robotserves as a power distributionblock and battery mount

Figure 8 shows Apollyon atWreck the Halls being prepared forthe first match of the competition

Apollyon won the 12 lb classwith a 3-1 record (see Figure 9)

The damage was all cosmeticand easily repaired The shaft collar

on the visible axel was torn off incombat A new front wedge isbeing designed to reduceimpacts on the chassis SV

• Victor 883 speed controller 2 $139.99

• GB42 gearbox 2 $85.99 (no longer available)

• Mini-EV motor 2 $8.99

• 18V 1,650 mAh NiMH battery pack $62.50

• Spektrum DX6 with BR6000 receiver 1 $199.99

• 3” Colson wheel with hub 2 $25.00 (no longer available)

(www.cncbotparts.com)

• GWS elevator/aileron mixer $14.99

• S-BEC Super BEC 5V $46.99 (being replaced

with receiver battery pack for future events)

• MS-05 power switch 1 $48.00

• 45A powerpole connectors $1.39/set

• #8 ring terminals $1.99/25

• Deans Wet Noodle Wire — 12 awg $1.25/ft

• Raw metal, nuts, and bolts (www.mcmaster.com and www.onlinemetals.com)

• Wood and plumbing pieces for battery mount — local hardware store

• Machining (www.teamwhyachi.com) NOTE: All items are from www.robotmarketplace.com unless noted otherwise.

Trang 30

from inexpensive, low-power imports(such as the drills offered by HarborFreight) to high-performance, name-brand drills (particularly DeWalt).

Unfortunately, drill gearmotors lackconvenient mounting features, oftenhave minimal bearings, and generallycontain slip clutches, making themunsuitable for use in robotic driveand mobility applications withoutmodifications While there are commercially-available modificationkits for some drill gearmotors, mostmotors must be modified by thehobbyist The photos here document

my modifications toDeWalt’s newest andmost powerful drillmotor SV

PHOTO 1 The 36-volt DeWalt DC900KL

contains a gearmotor rated for 750 watts of

power and is capable of a bone-crushing 200

foot-pounds of stall torque in low gear On

paper, this gearmotor should be capable of

moving a 200 pound robot at about five miles

per hour in a 4WD configuration.

PHOTO 3 The original motor mount is shown on the left, with

its twist-lock interface to the gearbox An alternative mount

— machined from a block of magnesium — is shown on the

right Long bolts will be used to assemble the motor mount

to the new gearcase.

PHOTO 2 The gearbox (part #629059-00) and motor (part

#639521-00) are shown without the drill casing They are

mounted to the casing by the small plastic tabs where the

motor and gearbox meet To mount the gearmotor securely

to a robot, we will make a metal replacement for the stock

plastic gearcase.

PHOTO 4 As shown on the left, the first-stage ring gear (part #628002-00) is free to rotate as part of the clutch mechanism To lock this gear

in place, we can machine it down to a square,

as shown on the right The modified gear presses into the motor mount.

PHOTO 5 The DC900KL uses a three-speed gearbox — for robotics use, the gears must

be secured for a single speed The stock gearcase (top) uses a set of molded-in teeth

to keep the ring gears (bottom) from turning when engaged; these have been duplicated in the magnesium gearcase (right).

PHOTO 8 The completed gearcase, with motor and gearbox parts installed, is shown here The parts were made on CNC equipment, but could be made manually with slight simplifications The motor, gearbox, and spare ring gears are available from www.dewaltservicenet com, at a total cost of $110 per motor-gearbox assembly.

PHOTO 7 The gearbox output (left) uses a simple double-D shaft geometry, duplicated in the hardened, high-strength shaft on the right The step in shaft diameter allows the new gearbox to be protected from impacts

by a bronze thrust bearing.

PHOTO 6 The stock gearbox output has an overrunning clutch to prevent back-driving.

Since this can cause the output to lock up under some conditions,

it must be disabled by removing the five pins and outer ring.

Trang 31

ROBOT PR FILE

TOP RANKED ROBOT THIS MONTH

Billy Bob has competed in House

of NERC 2006, Motorama 2007,

RoboGames 2007, Franklin Institute

2007, and Motorama 2008

Details are:

● Configuration — Vertical Spinner

● Frame — 2024 milled aluminum

frame

● Drive — Two AstroFlight 940s

with Team Whyachi gearboxes

● Wheels — Two 3.5” Colsons

● Configuration — Two wheel drive

in the rear

● Drive ESC — Two IFI Victor 883s

● Drive batteries — 21.6 volt A123

battery pack

● Weapon — Custom S7 tool

steel single tooth blade at 8,000RPM

● Weapon power — 24 volt NiCd

2,400 mAh battery pack

● Weapon motor — Axi 5330 at

24 volts

● Weapon ESC — Castle Creations

Phoenix HV85

● Armor — Rubber shock mounted

steel and a rear titanium wedge,along with various attachments forother types of bots

● Radio system — Spectrum DX6

● Future plans — Four-wheel drive

version

● Design philosophy — Balance the

drive, weapon, and armor withsmart engineering, and make itcool!

● Builders bragging opportunity —

Billy Bob went undefeated at its firstevent SV

All fight statistics are courtesy of BotRank

(www.botrank.com) as of May 10,

2008 Event attendance data is courtesy

of BotRank and The Builder’s Database

(www.buildersdb.com) as of May 10,

2008.

● by Kevin Berry

Weight

Class Bot Win/Loss Weight Class Bot Win/Loss

150 grams VD 26/7 150 grams Micro Drive 7/1

1 pound Dark Pounder 44/5 1 pound Dark Pounder 23/3

1 kg Roadbug 27/10 1 kg Roadbug 11/4

3 pounds 3pd 48/21 3 pounds Limblifter 12/1

6 pounds G.I.R 17/2 6 pounds G.I.R 11/2

12 pounds Solaris 42/12 12 pounds Surgical Strike 17/7

15 pounds Humdinger 26/4 15 pounds Humdinger 26/4

30 pounds Totally

Offensive 43/13 30 pounds Billy Bob 12/4

30 (sport) Bounty Hunter 9/1 30 (sport) Bounty Hunter 9/1

60 pounds Wedge of

Doom 43/5 60 pounds Texas Heat 11/4

120 pounds Devil's Plunger 53/15 120 pounds Touro 10/0

220 pounds Sewer Snake 43/12 220 pounds Sewer Snake 11/5

340 pounds SHOVELHEAD 39/15 340 pounds Ziggy 3/0

390 pounds MidEvil 28/9 390 pounds MidEvil 3/0

Top Ranked Combat Bots

Rankings as of May 10, 2008

History Score is calculated by

perfomance perfomance at all

events known to BotRank

Current Ranking is calculated by performance at all known events, using data from the last 18 months

History Score Ranking

Billy Bob – Currently Ranked #1

Historical Ranking: #9 Weight Class: 30 lb Featherweight Team: Benson Labs

Builder: Brian Benson Location: Winchendon, Massachusetts

BotRank Data Total Fights Wins Losses

Trang 32

If you rely on the Satcom Mag

motor to deliver a deadly blow or

a punishing push to your robotic

opponent, you will want to follow

these simple tips for improved

reliability and performance

Tape the Brush Covers

The quickest tip is to place trical tape over the four brush covers

elec-(see Figure 1) They don’t often come

loose, but losing a brush cap during

an event can be a disaster If yourmotors get particularly hot, youmight need high temperature Kaptontape, however I have never had aproblem with the cheap stuff

Replace the Case Screws (beginner)

The long case screws supplied

by Satcom are 10-32 by 3” stainless

steel with Phillipsdrive heads Theseare not up to heavycombat duty andshould be replaced

I use hex socket cap screws,McMaster-Carr part

#91251A360, forthis; they can betightened morethan the originalscrews and thethreads are betterformed (Figure 2)

The heads of thesescrews aresometimes

too tall for the counterbores in thefront of the motor, so start by grinding the heads down by about.016” (0.4 mm)

Grinding the heads down alsoputs a small burr around the 5/32”hex socket, making it a tight fit onyour hex driver Use the driver towiggle the screw about until itstarts to thread into a hole; afterthe first screw, this becomes veryeasy I like to apply a slathering ofLoctite 243 to make sure the screwsstay put NOTE: If you have a newerfour screw motor, one screw is only2-3/4” long; you will need to cutone of the replacement screwsdown to fit

Replace the Mounting Screws

The tiny 8-32 face mountingscrews always look inadequate tome; if you are mounting the motorwith these screws, I recommend up-sizing them to around 12-24 orM6 Start by removing the frontplate of the Mag motor, leaving thearmature and case in place

brush covers and

the timing marks.

FIGURE 2 Replacement case screws (Insert:

The replacements have far stronger heads.)

FIGURE 3 Aligning the mounting holes with a tapered pin (Insert: Which screw size would YOU trust?)

Trang 33

Position the front plate

accurately on your drill press using

a tapered pin made from the shank

of an old drill (see Figure 3) This

will keep the new mounting points

accurately in position; important if

you have a pre-made gearbox to fit

Drill each of the mounting holes

out to fit your preferred screw size;

5 mm for M6 and #17 for 12-24

sized screws

Tap the new holes and take

extra care to remove all the swarf

from the inside of the plate; you

don’t want conductive chips falling

into the armature! Re-assemble the

motor as detailed in previous tips; if

you combine this tip with thestronger case screw tip, you willhave one tough motor

Neutral Timing

Mag motors can be timedslightly advanced or retarded,although I find that advanced timingmakes hardly any difference so Ileave them neutrally timed If youwant to experiment or need toadjust the timing after repairs, here

is the quick way to do it Mark a linealong the case so it meets the rearend bell, as shown in Figure 1 Withthe case screws loosened enough to

just turn the case, rotate the case left until the screws touch amagnet Mark the position of yourline on the rim of the endbell Nowrotate the case left until the screwsstop on the opposite magnets

Mark this position on the endbellrim The marks on the rim representthe extremes of forward andreverse timing; draw a line midwaybetween your two extremes and this is your neutral timing position — too easy! SV

You can contact Nick via his build thread at

Twenty-five bots were registered

Rotunda Rumble was held at the

Mall Of America in Minneapolis,

MN on April 25th Twenty-seven

bots were registered

Smackdown in Sactown IV

was held on April 27th in

Sacramento, CA Eight bots were

registered

BotsIQ: The Competition 2008

was held April 30th–May 4th in

Miami Beach, FL One hundred thirty-five bots were entered

Roaming Robots presentedFenton Manor 2008 on May 4th

in Stoke On Trent, England

Upcoming Events for July-August 2008

D W Robots willpresent

Pennsylvania BotBlast 2008 in Bloomsburg, PA onJuly 12th Seventeen bots wereregistered at press time For more

details, go to www.dwrobots.

com/tournament.html.

War-BotsXtremewill presentWBX-V

“Taking theFifth” on July26th inSaskatoon,Saskatchewan, Canada Nineteenbots were registered at press time

For more details, go to www.war botsxtreme.com.

The North East Robotics Club willpresent House of Benson –Barnyard Brawl in Winchendon, MA,

on July 26th Thirty-six bots wereregistered at press time For more

details, go to www.nerc.us.

RoamingRobotswill presentGuildford

2008 inGuildford, England on June 15th,and UK Champs 08/RAF Fairford inGloucestershire, England on July12th and 13th For more details, go

to www.roamingrobots.co.uk.

RoboCore will present WinterChallenge 4th Edition inAmparo, Sao Paulo, Brazil on July 26th–27th SV

EVENTS

Results and Upcoming Events

Trang 34

including car audio, home audio, cellphones, DVD players, laptops, digitalphoto and video recorders, and ofcourse, massive plasma TVs There was

a 150 inch TV at the Panasonic booththat was the talk of the show Imagine

a TV as big as a queen-sized bed hanging on your wall? Seems like onewould have to reinforce the wall just tokeep it from falling down

Most of the booths that are technology specific are organized intoTech Zones, such as USB, ZigBee, Blu-ray Disc, HDMI, Mobile Internet andWiMAX, IPv6, Sustainable Technologies,

CES 2008

Robot Roundup

CES 2008

A s usual, in the second week of January,

over 100,000 technology lovers converged

on Las Vegas for the 41st annual Consumer Electronics Show (CES) The show

went on from 8 AM to 5 PM for four days, and

even then, it was almost impossible to see

everything CES took up 1.8 million square

feet of trade show space, spanning all of the

major convention centers in Las Vegas So much

walking is involved for the attendees, you can

easily blow out your feet unless you are wearing

running shoes.

Just about every electronic consumer device is represented at CES

by Ted Larson

Photo 1

Trang 35

and finally, Robotics With just a few exceptions, most of

the robots and robot companies were on display in the

Robotics Tech Zone or in nearby booths

The biggest exhibitor in the Robotics Tech Zone this

year was WowWee (www.wowwee.com) Every year they

seem to have many innovative, new designs and this year

was no exception The most noteworthy items they brought

out were the Femisapien, Tribot, and Rovio Femisapien is a

female counterpart for the popular Robosapien She

dances, poses, and with her 9x degrees of freedom is

capable of 56 interactive functions They had a nice demo

with three of them line-dancing together (Photo 1) She has

a learning mode where you can pose her and learn

sequences for playback, which seems like endless fun I

thought the best feature of all was that when brought in

proximity to a Robosapien, she is the boss and tells him

what to do (sounds a bit like my wife) With an MSRP of

$99, I can see her ending up on the desks of many geeks

for fun and show

Tribot (Photo 2) certainly received the most attention

out of the WowWee group, with the TV crews swarming

around him to get some footage of him doing his demo I

was really surprised at how much personality he had when

they switched him on He has animated ears and a pop-top

head, and goes on and on about how great it is that you

rescued him from his packaging when you turn him on for

the first time He has an omni-wheel, holonomic drive

system which seems to be a new theme for WowWee

robots Several of their newer products are now sporting

omni-wheels, including Rovio Rovio is a WiFi capable,

omni-wheeled robot with a camera and navigation

capabilities It can be tele-operated over the Internet,

with live streaming video of what it sees The notable and

unique technology item here is that it is using the Evolution

Robotics (www.evolution.com) Northstar 2.0 system

for navigation

Northstar is like “micro-GPS” for a robot It uses

constellations of infrared energy beamed onto the ceiling

from a fixed point to determine its location within a room

Rovio is the first use of this technology in a low-cost,

consumer package and it is quite impressive One can set

waypoints for Rovio within the room and it easily find its

way back to them, even if the robot is picked up and

moved At a retail price of $299, it is quite amazing that all

this technology can be packed into such a cool little robot

Since I am on the topic of indoor positioning, it is a

good time to mention the booth of Hagisonic Hagisonic

(www.hagisonic.com) is a Korean sensor manufacturer

which makes an indoor positioning system for robots called

StarGazer StarGazer analyzes the image of an infrared ray

which is reflected from a passive landmark with a unique

ID, mounted on the ceiling From this landmark, it is able to

determine its repetitive position down to 2 cm of accuracy.The technology is similar to that of Evolution RoboticsNorthstar system, although Evolution does not need to stickanything to the ceiling StarGazer is currently manufactured

as a module you can simply mount in a robot, place the IRprojectors in the room with the passive landmarks, and youare ready to go They had a nice demonstration of two littlerobots navigating around on the floor, avoiding obstacles,and mapping their positions (Photo 3)

Meccano showed their new additions to the Spykee

(www.spykeeworld.com) robot line-up All the Meccano

robots are being sold under the ERECTOR brand as a robot

CES 2008 Robot Roundup

“The biggest exhibitor in the Robotics

Tech Zone this year was WowWee ”

Photo 2

Photo 3

Trang 36

Erector set The four new robots in the line are Spykee Cell,

Spykee Miss, Spykee Vox, and Spykee Micro, three of which

are designed to cradle your iPod, and allow it to be voice

controlled and give it some personality (Photo 4) Spykee

Cell can be controlled from your cell phone via Bluetooth

and is targeted at both boys and girls Spykee Miss is an

emotional electronic friend targeted at girls that gives you

advice when you ask her a question Spykee Vox is also

voice controlled, an interactive friend, and can be either a

hero or a villain It is targeted primarily at boys Spykee

Micro is a small, little remote controlled robot that looks

similar to its larger counterparts, but is primarily just for

driving around and making noise Again, all the robots are

kits — in the spirit of the ERECTOR brand — and some are

easier to assemble than others When we were packing up

at the end of the show, the Meccano people were looking

to lighten their load for their trip home, so they gave us

several Spykee Micro kits I brought two assembled unitshome, and had great fun with my four-year-old daughterhaving robot races in the hallway with them

About 50 feet from the Robotics Tech Zone was theiRobot booth Among all the consumer robots they showed,the two that were the standouts were the iRobot LoojGutter Cleaning Robot (Photo 5) and the new iRobotRoomba 500 The Looj has piqued my interest ever since itwas announced, although I had never seen one in personbefore I have heard many rumors about what it can andcannot do, so I thought I would ask the tough questions,and see if I could clear some things up In a nutshell, theLooj is a remote controlled robot that is designed to be put

in a gutter and run up and down to dislodge any debris,using a rotating rubber agitator The idea is you climb upthe ladder to the corner of your house, put the Looj in thegutter, and run it up each gutter section, thus minimizing

CES 2008 Robot Roundup

Photo 4

Photo 5

Photo 6

Trang 37

the number of times you have to go up and down the

ladder They had a nice little demo with a piece of roof, a

gutter mounted to it that was filled with plastic leaves, and

they would drive the Looj down it and it would throw the

leaves all over the aisle in front of their booth (Photo 6) It

was great fun to see it go

After asking many questions, I came up with all the

things it cannot do, to dispel any myth making It cannot

climb the downspout, you cannot just throw it up on the

roof and let it do the rest, it cannot go around the corners

in your gutters, it is not compatible with ancient gutters,

with weird dimensions, and it won’t do the job all by itself

while you sit down on your lawn with a glass of lemonade

What it can do, it does very well, and is quite amazing Its

paddle is capable of blasting through all kinds of gunk in

your gutters, like pine needles, twigs, and sludge If your

gutters are a standard 2-1/4” size, it is capable of driving

underneath the straps that hold the gutter to the house,

so you can clean long sections If your house was perfectly

square, you would only need to ascend and descend the

ladder four times, thus minimizing your risk of life and limb

by falling off the ladder It is certainly better than the old

method of climbing up there with gloves and a plastic

scoop, and at only $99 for the base unit, it is cheaper than

most lawn and garden appliances

The iRobot Roomba 500 Series (Photo 7) was there,

driving around a little test carpet that anyone could scatter

all kinds of debris on, and it would happily slurp it up Every

time I see a new version of the Roomba I think, okay, what

now, seen this before, ho-hum However, this one has some

great new features that would make me upgrade It has

anti-tangle technology, that detects if it sucked up a carpet

fringe or an electrical cord, and automatically backs it out

of the beater-brush before continuing along and restarting

to clean It has upgraded bump switches which give it a

lighter touch to keep it from scuffing the baseboard or

furniture Also, it has a new Virtual Wall, called a

Lighthouse, that the robot interacts

with to allow it to be contained to one

room until the room is clean, which

then allows it to move onto the next

room and so on, and so on So, one

robot can clean multiple rooms, and

know when it has completed them all

At $349, it is in line with the pricing

of previous Roomba robots, and offers

significant improvements without a

significant increase in price

Roboware (www.roboware.

com.hk) had a booth in the Tech

Zone, where they were showing off an

impressive looking, three wheeled,

holonomic drive humanoid named E3

(Photo 8) I guess 2008 is the year for

holonomic drive humanoids?!?!

Roboware was founded by Mike Kim,

who previously was one of the

researches on the Canadarm space

robot on ISS, was part of the Hubble Telescope rescue project, and has contributed to some WowWee projectssuch as RSMedia, Elvis, and RSG products According toMike, E3 stands for Education, Entertainment, and Emotion,which makes it a platform much like a video game E3 canexpress its emotion through motions (head, arms, body,wheel), light, and multi-media with customized content

It has five login modes: Baby, Teen, House-Keeper, Single,and Silver Each mode has its own unique and updatablepersonality according to the user’s age E3 has WiFi built in

so it can be controlled remotely via the Internet andthrough its ad-hoc networking capabilities can be voice controlled, or stream or playback live video It is even capable of doing Sykpe teleconferencing E3 has a big 5”LCD mounted in his chest and runs Windows Mobile edition, so he can do many PDA-type functions, as well The retail price range of E3 will be between $1,500-$2,000and he will be available in the US around November of this year

Robotis (www.robotis.com) returned to CES again this

Photo 7

Photo 8 Photo 9

CES 2008 Robot Roundup

Trang 38

year with yet more newthings to show, such as

The Dynamixel servos are designed specificallyfor robotic actuator applications, and are networked together using

a communications bussuch as RS-485 or TTL signaling They are powerful, metal gearedservos with torque up to

64 kg-cm The Bioloid kits(as well as URIA) are constructed from theseservos URIA stands forUbiquitous RoboticInformation Assistant, and

is designed as a researchplatform for working with humanoids It has

a fully embedded PC onboard, running Windows XP, with

peripherals such as USB, LAN, Camera, VGA, WiFi, and a

microphone He has a nice big LCD in his chest so you can

see what is going on with the PC

Other interesting peripherals include a Passive

Infrared Sensor (PIR) and a six-axis gyro for measuring

motion The robot stands 22 inches tall and weighs about

12 pounds In comparison to most of the Robo-One type

humanoids, he is really, really big They didn’t give exact

pricing on this monster humanoid, but they were quite

specific that it is designed for researchers and not the

hobbyist With all the servos and PC onboard, I don’t

imagine he is going to be inexpensive

OLogic was there with plenty of interesting robots to

show (www.ologicinc.com) OLogic is an outsourced

research and development company with a focus on robotics, that I co-founded with Bob Allen Of course, webrought out some balancing robots to demonstrate ourcapabilities to design difficult control systems On the firstday of the show, we realized we could place one on top ofanother and do a Las Vegas acrobatic act, in true Vegasstyle (Photo 10) Needless to say, it always attracted acrowd and the TV people whenever we stacked them up.Dean Kamen, the inventor of many things including theSegway, came by and we were able to snap some photos

of us with Dean and the balancing robots (Photo 11)

NPC Robotics (www.npcrobotics.com) commissioned

OLogic to build a robot to demonstrate a device they havebeen reselling, called a Ribbon Lift (Photo 12) A Ribbon Lift

is a device that takes three stainless steel ribbons rolled up

on a spool like a tape measure, uses a motor to unwindthem, and stitchs them together into a self-assembling, triangular shaped pole Since we just finished the robotbefore CES, we brought it out to show off The robot isappropriately named “Giraffe” due to its long neck it canextend The lift mounted in the robot is capable of raising a

50 pound load to 15 feet, and can collapse down into aspool eight inches high by 20 inches in diameter It is quiteamazing to see it unfurl, and some people commented that

it seemed like magic, like Ali-Babba’s magic rope trick Wemounted a WiFi camera on the top and had it feeding a big plasma display to demonstrate its use for surveillanceapplications We are looking forward to building somerobots using the larger version of the Ribbon Lift that canlift a 500 pound load 25 feet in the air

Two robots I missed at CES this year, but heard about,were robots that showed up for just one day to make a

cameo appearance in the Robotics Trends booth (www roboticstrends.com) They were Pleo (www.pleoworld com) the Camarasaurus, made by UGOBE, and Zeno the Revolutionary Robotic Friend by Hanson Robotics (www zenosworld.com) It was a bummer I missed them both,

but there was so much

to see, and certainlyone couldn’t see it all.Hopefully, I will be able

to catch up with themboth next year SV

Ted Larson is the CEO of OLogic, Inc., and an active member in the Home Brew Robotics Club of Silicon Valley OLogic is an embedded systems research and development company with

a focus on robotics OLogic

is currently working with clients across a wide spectrum

of application domains such

as consumer electronics, toys, medical products, and

Trang 39

Getting a valid lower resolution reading of the encoder

is certainly better than high resolution unreliable data

I needed to know how far the motors really moved but

didn’t necessarily need to know with as high of a resolution

as the encoders were providing So, the task at hand was

a way to lower the resolution enough so that I could

always count on the readings and help keep track of the

movements A bit of encoder resolution may be lost but it

should still be close enough for this particular project

Symptoms o of tthe P Problem

I was watching the

encoder counts while running

some robot drive motors

During the initial bench

testing, the motors weren’t

running at full speed and the

results I saw were just as

expected However, things

got interesting when I started

ramping up the speed A very

peculiar thing started

happen-ing The encoder counts

started to rise as expected

but as the motors sped up,

the counts started going backwards! As it sped up somemore, it counted forward again This cycle continued backand forth a bit and then the counts were completely erratic

It appears that the encoder was exceeding the polling timeused by the controller and would start missing pulses athigher speeds Obviously, the encoders were not matched

up well with the controller I’ve seen this happen with myrobot drive base and also when I was using some salvagedencoders from HP scanners and DeskJet printers Wheneveryou see symptoms like this, it should throw up a red flag andmake you take a look to see if this may be the problem

Encoder processor board (component side) Encoder processor board (solder side).

One of the joys of the robotics hobby is mastering the art of interfacing A lot of the parts are already in front of us and we just need to make them work together Recently, I ran into a problem with a pair of quadrature encoders for the drive train on one of my robots The encoders themselves worked fine and were generating perfect quadrature outputs However, they were sending out data faster than the controller could handle at higher speeds As a result, the encoder readings were worthless and could not be trusted.

by Robert Doerr

Trang 40

Methods o of C Correcting tthe

Problem

There are a few different methods of fixing this

problem In a nutshell, we just need to reduce the number

of transitions of the encoder per revolution Obviously, the

encoder itself could be swapped out with a lower resolution

one If feasible, it may be possible to replace just the

encoder disk with one that has a fewer number of holes

Another option is to change the physical placement of the

encoder within the drive train When it is directly attached

to the armature of the motor, it may send out too many

pulses If it was moved downstream to the wheels

themselves or at some point in the gear train, it would slow

down the speed of the encoder It will still be sending the

same number of pulses per revolution of the encoder but it

will be rotating slower so we get the effect we’re after

Instead of altering the mechanics of the encoder — which

isn’t always an option — we can look into electronic means

of scaling the values At first, I considered making something

up with standard logic ICs but an easier and more flexible

option was to use a small microcontroller to help match up

the readings Whatever method is selected, we would like

the encoder to supply as high as possible a resolution

without sending them too fast for the controller to handle

Microcontroller tto tthe R Rescue

I’ve run into this problem with other encoders and

controllers I wanted a flexible solution so I decided to use a

small microcontroller to scale the encoder readings This

ended up being an excellent solution that can be used for

other projects, as well I selected an SX28 processor and

wrote all the code in SX/B This development went really

fast I started building the encoder scaling board one

evening and had a working prototype with the core

functionality the next day Everything is handled by the

single SX28 chip which costs under $3 A few more features were then added and the code cleaned up tomake it more presentable

One of the extra features is the ability to allow a hostcontroller to change the scaling factor on the fly Since itwill accept commands via serial connection, I also added aresonator for the clock timing to ensure the serial communi-cations would be reliable This microcontroller will beinstalled in between the existing encoder and the controller.This way, it can monitor the encoder, condition the signal,and output a virtual encoder signal to the host controller.Quadrature E Encoders 1 101

Many of you already know how tachometers and quadrature encoders work, but for those who don’t, here is

a quick review A tachometer signal only has one signal andtherefore only provides speed information and not anydirectional information It will only tell how fast somethingmoved but not any details about the direction of travel Itcan be used to get an estimate on distance traveled as long

as the direction is known ahead of time However, a problemcan come up if the tachometer stops at a transition (edge)

If there are any vibrations, it can toggle state back and forthand mistakenly give the impression that it is moving It ismore suited to regulate the speed of a motor where distance/direction are not important; just the speed is critical.Adding a second channel makes all the difference! Aquadrature sensor will have two signals 90 degrees out ofphase These two signals are commonly referred to asChannel A and Channel B With these, both position anddirection can be determined The direction is determined bycomparing if Channel A is leading Channel B or if Channel

B leads Channel A The distance can be computed by keeping track of the counts and factoring in the direction

of movement Velocity can be calculated by counting thenumber of pulses per second I’ve seen some references toencoders that state: “If A leads B, for example, the disk isrotating in a clockwise direction If B leads A, then the disk

is rotating in a counter-clockwise direction.”

That may be true for some encoders, but needs to beverified for the particular encoder being used I prefer tojust think of them as two channels of information and look

at which one is leading, then match that to the actual direction of how it is installed in the robot The clockwise/counter-clockwise description isn’t something that lendsitself well to straight quadrature encoders, so keeping itgeneric works out well After all, it shouldn’t matter if it isclockwise/counter-clockwise, right/left, up/down,

forward/back, etc The important part is that two distinctdirections can be determined; the rest is relative

Look at the examples of output from quadratureencoder, through a few cycles

Chan A 0 1 1 0 0 1 1 0 0 1 1 0Chan B 0 0 1 1 0 0 1 1 0 0 1 1

Encoder assembly on robot base.

Ngày đăng: 02/01/2014, 01:26

TỪ KHÓA LIÊN QUAN

TÀI LIỆU CÙNG NGƯỜI DÙNG

  • Đang cập nhật ...

TÀI LIỆU LIÊN QUAN