Robotics in theory and practice

Nemcovej 32, 042 00 Kosic, Slovak Republic 2 Manex, s.r.o., Alvinczyho 12, 040 01 Kosice, Slovak Republic 3ZŤS VVÚ, 040 01 Kosice, Slovak Republic a mikulas.hajduk@tuke.sk, bpeter.jencik

Tai ngay!!! Ban co the xoa dong chu nay!!!

Robotics in Theory and Practice

Edited by Lucia Pachnikova Mikulas Hajduk

Robotics in Theory and Practice

Selected, peer reviewed papers from the

11th International Conference Industrial, Service and Humanoid Robotics

ROBTEP 2012, November 14th - 16th 2012, Strbske Pleso, High Tatras, Slovakia

Edited by

Lucia Pachnikova and Mikulas Hajduk

Copyright 2013 Trans Tech Publications Ltd, Switzerland

All rights reserved No part of the contents of this publication may be reproduced or transmitted in any form or by any means without the written permission of the

Full text available online at http://www.scientific.net

Trans Tech Publications Ltd Trans Tech Publications Inc

Kreuzstrasse 10 PO Box 699, May Street

CH-8635 Durnten-Zurich Enfield, NH 03748

Switzerland USA

Phone: +1 (603) 632-7377 Fax: +41 (44) 922 10 33 Fax: +1 (603) 632-5611

e-mail: sales@ttp.net e-mail: sales-usa@ttp.net

PREFACE

Dear Distinguished Authors and Guests

It is our pleasure to warmly welcome you to 11th International Conference Robtep 2012, held on 14th –

16th November 2012, Strbske pleso, High Tatras, Slovakia

The aim of the Robtep 2012 Conference is to present the latest research and results of scientist such as professors, students, PhD students and engineers related to robotics and manufacturing systems This conference provides opportunities for the different areas delegates to exchange new ideas and application experiences face to face, to establish business or research relations and to find partners for future collaboration in research and projects

All papers published in this volume have been peer reviewed through processes administrated by the proceedings editors Reviews were conducted by experts referees to the professional and scientific standards The conference program is rich, we would like to thank to all who presented their papers, and our special thanks go to Dr Cecile Huet from European Commission for her interesting presentation of European Framework Programme and Horizon 2020 Programme as well as to Mr Elmo Shreder from euRobotics for his interesting presentation of European Robotics Platforms and possibilities to join them We’d like to extend our thanks to members of scientific committee for their effort to the conference; especially, we’d like to thank

to members of organizing committee for their hard working; finally, we would like to express our appreciations to the participants of this conference

The Robtep conference is organized within the project “Research of modules for intelligent robotic systems”, ITMS No 2622022141, OPVaV 2009/2.2/05 SORO

With our warmest regards

prof Ing Mikulas Hajduk, PhD

Conference Organizing Chair

Guarantee of conference and scientific committee

Dr.h.c prof Ing Anton IŽMÁR, CSc

Dr.h.c mult prof Ing František TREBU A, CSc

prof Ing Mikuláš HAJDUK, PhD

prof Ing Juraj SMR EK, PhD

prof Ing Vladimír OP, DrSc

Ing Jaromír JEZNÝ, PhD

Ing Ladislav VARGOV ÍK, PhD

Ing Peter JEN ÍK

Trends in industrial robotics development

Mikuláš Hajduk1, a , Peter Jenčík2, b , Jaromír Jezný3, c, Ladislav Vargovčík4, c

1

Technical University of Kosice, Faculty of Mechanical Engineering, Department of Production

Systems and Robotics, B Nemcovej 32, 042 00 Kosic, Slovak Republic 2

Manex, s.r.o., Alvinczyho 12, 040 01 Kosice, Slovak Republic

3ZŤS VVÚ, 040 01 Kosice, Slovak Republic a

mikulas.hajduk@tuke.sk, bpeter.jencik@manex.sk,cjaromir.jezny@ztsvvu.eu,

cladislav.vargovcik@ztsvvu.eu

Keywords: industrial robotics, multirobotic cells, duo robots, industrial mobile robots, co-worker

robots

Abstract The article describes the development and defines the change of approach in the

development of today's industrial robotics, provides an overview of the latest trends in the field of industrial robotics Until now, the industrial robots have been deployed to less demanding work environments to perform "only" handling operations and to synchronize the operations of individual facilities Now they are undergoing a major innovation process, the bulk of which is focused on increasing their intelligence and multi-functionality

Wide application possibilities of robots require managing their design based on a modular principle allowing the construction of a variety of kinematic configurations of robots, as well as of effectors and flexible and intelligent control The times when a robot was only suitable for repetitive handling operations are gone Today's range of robots includes nanorobots which are capable of handling molecules, large robots with capacity of more than 1 000 kg and robots for virtually every manufacturing and non-manufacturing sector, but also in radioactive environments, sea and space, and there are less and less areas without the use of robots So far we have been looking for new areas to use robots but we have reached the point of asking a question: “Is there an area where the use of robot has not been possible yet?”

Robotics development in the world

At the forefront of the global development of robotics are Japan, The USA, Europe and South Korea (Fig 1) The USA dominates in service robotics for military use deploying mobile robots type off-road They are unique in the field of space robots and in the development of interplanetary robots Interestingly, the USA does not dominate in industrial robotics, even though robots were first manufactured in the U.S (General Motors, Cincinnati Milacron, Westinghouse and General Electric) Well-known manufacturers of industrial robots in the U.S today are only Adept and San Jose-based Company

In Japan and South Korea research activities and production of service robotics are widely developed, humanoid robots includes These robots are primarily focused for household jobs, entertainment or rescue work It turns out to be one of the most traded goods in the next 10 years Japan and South Korea see great potential in the development of robots for elderly care Japan has traditionally been strong in industrial robotics

Industrial robots and service robotics dominate in Europe These are focused on mobile robotics, transport and logistics, and especially in the external environment (in urban environment) The second area is represented by robots to work with humans Europe is the leader in manufacturing and deploying industrial robots (with 33% representation) There are about 15 major companies producing industrial robots in Europe (KUKA, ABB, Reis, SCHUNK, STAUBLI, PROMOT, COMAU, CLOOS, FATRONIC)

Fig 1 The distribution of the global development of robotics

There are a few dominant national and international programs for robotics research in these areas The USA adopted the document Robotics and Automation Research Priorites for U.S Manufacturing in 2009, which emphasizes that robotics is the key to the transformation of production to achieve high competitiveness In Japan, large companies have their own programs to develop new solutions and applications of robots Also in Europe there are more and more research programs focused on robotics

The year 2010 is known as a strong comeback towards industrial robotics This stems from the fact that in 2009 the annual installation of robots fell to 60,000, in 2010 it was already 118,000 in

2011 about the 130000 The upward trend is expected in the next five years and in the 2017, the annual number of installed robots should exceed the value of 200000 This trend is based mainly on dynamic growth of markets and deploying robots, especially in China, South Korea and ASEAN countries

New areas of development and application of industrial robots

Development of industrial robotics abandons the individual and "isolated" deployment of robots and moves to a group building and deployment of workstation of type robot - human Changes in the approach to the development of today's industrial robotics are shown in tab 1

stabile industrial robot mobile reallocation

periodic or repeated cycles with little changes frequently changed tasks, rarely repeated cycles individual activity of robots robots’ cooperation

on-line/off-line programming on-line task assigning

no human-robot cooperation in the robot zone mutual human-robot cooperation on tasks

efficiency at middle and higher series higher efficiency a lower series

It has been shown that the preferred way of deploying the robots of type one robot - one action has been inefficient However, at the workplace there are much more activities identified as auxiliary such as the exact position of the object, withdrawal and many others, for which they were designed as a single-purpose device At the current time of innovation, these workplaces do not meet the requirements The solution is represented by the robots with automatically replaceable effectors, respectively technology heads or reconfigurable grippers as well as the use of multiple robots as a group with a common purpose and robots with multiple arms and increasing the autonomy of robots in more and more unstructured environment Typical applications of multirobotic cells include welding systems in which one or more robots carry out welding and positioning and handling of weldments is executed by the other robot, fig 2 The benefit is obvious The advantage of such sites is that they can perform more of various activities

Fig 2 Welding – typical application of multirobotic systems

From duo robots (Fig 3) can be expected in the near future to go beyond the human ability, even

in sensitivity, not only in strength, speed and accuracy The basic idea of duo robotic development are human activities carried out with both hands for everyday handling and in collaboration with several workers

Fig 3 Dual – arm robot Fig 4 IMR - Industrial Mobile Robot)

In multirobotic systems and duo robots the establishment of activities, handling paths and synchronization of their movements and speed are new challenges Key aspects of multirobotic systems are parallel control and synchronization and cooperation of their activities

Industrial mobile robots

Industrial robots on a mobile platform (Fig 4), also known as Industrial mobile robots (IMR - Industrial Mobile Robot) is presented as a new category of robots

Integration of industrial robot and mobile platform/gear as a fundamental part of the service robot is a logical consequence of their development (Fig 5)

Fig 5 Integration of industrial robot and mobile platform

Classical industrial robot is taking on a new technical characteristic which is mobility Presence

of synergies from such integration enables further increase in the degree of automation of production IMR application is mainly in the logistics chain in manufacturing, but also in non-manufacturing operations IMR with the equipment for installation and service robot effectors and tools include in the category of service robots Integration of industrial robots and mobile platform/chassis provides further extension of the use of industrial robots

Robot Co-worker

Logical result of further development and of application of industrial robots will be robot worker-human worker systems- based on high symbiosis (Fig 6) Since these robot systems are expected to be able not only to help a person in many different applications, but to be able to communicate in natural language Highly level of autonomy and intelligence and of course safety is expected too

co-Fig 6 Robot Co-worker Designers and engineers had a dominant role in the design of industrial robots up to now, and their job was to make the best robot integrated into the manufacturing process, the task of design now passes to raise intelligence "human" behavior of the robots Psychologists take particularly important role here The existing approach was based on the conditions of appropriate integration with other machinery In terms of the position of man, it is important to design the appropriate interface for programming the robot and ergonomic aspects of its maintenance

Robot Co-worker is equipped with 3D systems for sensing the environment, even outside its work area, especially taking into account the movements of humans and safety identification One area of research of robot -human common workplace is one method of finding the optimal allocation of tasks between robots and humans

Robot arm and human hands manipulate a single object together or both are involved in the same technology operation Such robotic arms are very sensitive and robot can respond to human commands It is believed that this type of robots will be of a great use, not only in industrial applications, but it will become our assistant, helper in many activities Furthermore, it is expected

to have wide use in medical rehabilitation, healthcare and support to immobile people

Conclusion

The basic requirement for the development of industrial robots is to improve their design in order

to achieve greater mobility and maneuverability and to increase their level of intelligence so that their autonomy is increased The main trend in the design of industrial robots is an intelligent mechanotronic drive solution module with direct integration in the robot joint The key technologies

of industrial robots are the use of new material components, which are stronger and lighter, and which achieves design improvements, better dynamic properties at a higher capacity and improved actuators and arms equipped with different sensors with sensitivity and visual homing to desired positions

[2] Olaru, S; Olaru, A, Assisted optimization of the dynamic behavior of the industrial robots with rheological dampers, WSEAS, ROCOM’08, Hangshou, China, 6-8aprilie, 2008

[3] Nieszporek T., Szymański W., Rygallo A.: A new control system of the industrial manipulator RIMP-401, In: Acta Mechanica Slovaca 2-A/2008, ISSN 1335-2393

[4] Arai, T., Pagello, E., Parker, L.E.: Advances in Multi-Robot Systems In: IEEE Transactions on robotics and automation, vol 18, no 5, 2002, p.655 -661 [5] Lynne E Parker: Current research in multirobot systems In: ISAROB 2003, Artif Life Robotics, 2003, DOI 10.1007/s10015-003-0229-9

[6] Farinelli, A., Iocchi, L., Nardi, D.: Multi-Robot Systems: A classification focused on coordination In: IEEE Transactions on System Man and Cybernetics, part B, pp 2015-2028,

2004

[7] Vesselenyi, T et all.: Augmented Reality Used to Control a Robot System via Internet In: CISSE International Joint Conferences on Computer, Information and Systems Sciences and Engineering, 2008, pp 539-544

[8] Bungău, C., Rus, A., Bratu, I., Robot-Assembly Task Change Position of The Center of Compliance, Proceedings, International Conferences TCMR – 2005, Machine Manufacturing, Chişinău, 2005

1, ,2, 2,

1

2

3kg 0.635m

0.5m

0.5

RobotControllerPC

2× 2

w B

w h

x, i w

1.2mm 4deg

Right robot position

Real time Non real time

Signal processing, Motion control

1

(1 + T F s)2

ˆ

s 1

y

s 1 s

s

DYk

PYk

Pk

Angle measured by rotary potentiometer Angle measured by camera

Filtered camera value

0s

1.5deg

Angle from potentiometer Angle from camera Disturbance

The analysis and design study of high speed robotic devices

Štefan Havlík and Jaroslav HrickoInstitute of Informatics, Slovak academy of Sciences Ďumbierska 1, 974 11 Banská Bystrica, Slovakia havlik@savbb.sk, hricko@savbb.sk

Keywords: compliant mechanisms, positioning accuracy, dynamics, modeling, simulation

Abstract The problem of multi d.o.f positioning devices based on compact compliant kinematic

mechanisms is to guarantee the desired positional accuracy in static and especially in dynamic modes of operation The study of accuracy and performance analysis of high speed devices is made

in this paper The influence of differences between stiffness and damping coefficients in actuated directions as well as mutual cross couplings between them are discussed in details and performance characteristics of such complex systems are simulated As proposed the problem of improving accuracy can be solved by insertion of the compensation member into control system that could be integrated in parallel or serial way The presented approach enables to verify the dynamical range of operation for small / micro positioning devices performing precise trajectory following tasks

Introduction

New applications of robotic systems in microelectronics, optoelectronics, medicine, biology, etc require to develop new task oriented robotic structures that exhibit specific performance together with development of sophisticated manufacturing technologies, sensing and control methods There are several specific requirements that follow from particular application tasks For instance:

- Extremely high positioning accuracy, which is usually better than 1 µm It is obvious that this accuracy can be achieved within a limited range of motions only

- Small or limited volume of the operation space, frequently some few cm3, or less than 1 mm3

In designing any mechanisms, at the beginning, there is always a first intuitive proposal As to compliant mechanisms the first proposal of the structure usually goes out from the similarity with some known rigid-body mechanisms Naturally, designing more complex compliant structures that include several elastic segments suppose calculations of possible errors that deteriorate positioning accuracy of mechanisms in static and dynamic modes of operations It can be said that in the design phase of compliant mechanisms much more attention and effort should be devoted than in cases of classic mechanisms made from rigid building elements

Problem of compliant mechanical structures for small, or micro, robotic devices are widely studied in many robotic laboratories Several papers deal with description, design and analysis of elastic joints [1, 2, 3], arms and the whole elastically compliant kinematic mechanisms [4, 5] On

the other hand, there is relatively rare literature that looks at these devices from the mechatronic point of view; i.e including control possibilities and complex performance characteristics

This paper analyzes problem of accuracy in designing compact compliant structures and building multi degrees of freedom (d.o.f) positioning devices for performing high speed operations The analysis is based on comparing characteristics that describe static and dynamic performance of the positioning mechanisms in particular axes considering possible positional errors of the end part These errors arise as result of different mechanical parameters and their mutual cross deflections in the compliant joints and whole structure of the mechanisms

The analysis and an approach to the design and control possibilities are given below

Design of kinematics and error analysis

The procedure of designing these kinds of devices usually consists of principal tasks, as follows:

- The design of kinematic mechanism i.e its topology and parameters Here several principal criteria related to geometry and motion specifications are verified and parameters of mechanism are optimized

- The stiffness / compliance analysis of previously designed mechanism where joints of rigid links were replaced by flexural segments Further criteria related to forces, displacement and dynamical performance are evaluated and the multi-criteria optimization procedure is applied

to choose the best solution

- Modeling characteristics and behavior of mechanisms with following simulation of the whole mechatronic system that include positioning mechanisms, sensors, actuators and control

- Evaluation of the expected performance and design of control

According to specifications of positioning problems the kinematics of motion mechanisms for rotation and translation end motions are solved as parallel kinematic structures Some configurations are shown in Fig 1, for rectangular / Cartesian motions of the end element and in Fig 2, combining rotational motions

Fig 1 Some configurations of motion mechanisms for Cartesian motions of end element

Fig 2 Some configurations of motion mechanisms for rotational motions of end element a) [13], b) [14], c) [15]

The accuracy of any positioning system is influenced by many factors that can be divided into four categories of errors:

a) Parametric / geometric errors (tolerances, i.e differences of actual dimensions of links from their nominal values, misalignments of axes, etc.)

b) Non-geometric errors (structural compliance of all elastic parts, thermal changes of dimensions, etc.)

c) Numerical errors (computer round-off, transformation / control resolution, etc.)

d) Environmental errors (uncertainty of coordinate systems and positions of parts in the working space, etc.)

The errors (a), (b) are systematic that could be identified, predicted and feed-forwardly corrected The (c), (d) are accidental errors and could be corrected only by sensory feedback way

In cases of compact compliant mechanisms there are further possible sources of errors that should be taken into account in static and dynamic modes of operation The total end positioning accuracy is especially influenced by not exactly defined deflections of particular joints / links and the flexural characteristics of the mechanical structure as whole It can be said that connections of links by elastic joint / parts are not ideal revolute or prismatic joints There are always some cross flexural effects that deteriorate final accuracy of the multi degrees of freedom mechanisms This is one of crucial problems of precise manipulation and task for design of compliant mechanisms with minimal cross-compliance effects As joint positions can not be directly sensed it is need to build the kinematic model and careful compliance analysis of the whole flexure Such mechanism behaves as the spring body and any change of its position corresponds to a level of elastic energy / work has to be done or remove Thus, the compliance / stiffness characteristics of the kinematic flexure have dominant influence on kinematic accuracy of the mechanisms

The static errors can be expressed and calculated when consider assumptions:

- The flexure behaves in an ideal linear force – deflection system

- All errors are relatively small (second order terms) with respect to the range of motions

Then, the flexural characteristics of a joint /segment in the structure can be expressed (see Fig 3)

C.L

where, in general, d is the six-component deflection vector that consists of lateral and angular components, L is the load vector that consists of force and torque components, C and K = C -1 are the (6x6) compliance and stiffness matrices respectively; all related to the joint reference system

Oj(x,y,z) Description of flexible joints (hinges) is given in [1, 2]

Fig.3 The elastic segment

Considering superposition principle the end positional error will respect serial / parallel / or compound configuration of the kinematic structure [10, 11] Thus, the end error for serial connected segments will be:

H j

T jH j jH j

j jH

e =∑ =∑

)

(2) and the error for parallel arrangement of elastic segments

H 1 jH j T jH

here, T jH is the transformation matrix that relate the joint, j - reference system, into the end, H -

references

Expression for errors in compound structure combines calculations of errors for both serial and parallel arranged parts of the mechanisms related to the end reference system Under the above assumptions, applying SVD (Singular Value Decomposition) method it is possible to evaluate and compare “the kinematic quality” of joints, particular elastic segments and the mechanisms, as whole [8, 10]

All these errors due to cross compliance effects represent static errors and, for slowly moved mechanisms can be easily compensated by feed-forward ways using the compliance model of the mechanisms [6, 7, 9]

More complex situation arises in cases of the high speed multi - d.o.f positioning mechanisms where beside the stiffness and damping parameters their mutual relations in particular d.o.f / control axes should be considered To show this dependence the brief and very simple experimental simulation with the 2 d.o.f mechanism, depicted scheme in Fig 4, has been done The mechanically

coupled system represent following parameters: m- mass, k 1 and k 2 – the stiffness coefficients and

b 1 , b 2 - the damping coefficients in equations for particular directions of motion y 1 ,y 2 and k 12 , k 21

and b 12 , b 21 are coefficients that represent mutual couplings

t x k k

k k t x

t x b b

b b t y

t y k k

k k t y

t y b b

b b t

12 1 2

1 2 21

12 1 2

1 2 21

12 1 2

1 2 21

12 1 2

Performing simulation, the accuracy of end trajectory of A point is observed and errors from the

desired input in rectangular form are calculated For control systems in both directions the same parameters of transmit functions were appropriately chosen for given frequency / time period

Fig 4 The scheme of the 2 d.o.f mechanical system

As can be seen from simulation experiment in Fig 5, the change of the operation speed towards higher values (smaller time period T) results in increase errors and the system collapses at T=0,08s Another situation becomes when parameters of the mechanism in x and y directions are different, as shown in Fig 6 It is observed that differences between stiffness and damping coefficients in particular directions results in not negligible errors too Higher rates of corresponding parameters deteriorate the end accuracy in higher speed modes, as well This fact, especially the difference of damping parameters is not satisfactory solved in literature and its influence is frequently neglected

0 0.2 0.4 0.6 0.8 1 0

0.2 0.4 0.6 0.8 1

Fig 5 Behavior of the system under increasing speed of motion

0 0.2 0.4 0.6 0.8 1 0

0.2 0.4 0.6 0.8 1

Fig 6 Effect of different stiffness and damping coefficients a) k 1 /k 2 =0.0125, b) b 1 /b 2 =5

In reality, due to cross flexural effects of elastic segments any compliant mechanisms should be

considered as mechanically coupled system, i.e where coupling coefficients k 12 , k 21 ≠ 0 and b 12 , b 21

≠ 0 This practically means that any desired displacement of the mechanisms is corrupted by

superimposed static and dynamic errors In order to minimize this influence, the complex compliant mechanisms are usually designed symmetrically, i.e with the same configuration of elastic structures in particular directions of motions / axes, if possible Anyway, especially in majority of parallel mechanisms these influences and performance characteristics, as discussed below, should

be taken into account in design and compensated in operation

When evaluate dynamic behavior of mechatronic devices methods using frequency transfer functions are commonly used The general, the frequency transfer function in form

( ) ω ( ) ω jϕ( )ω

e A j

Consists of the amplitude part A(ω) and the phase φ(ω) part, both depending on input frequency ω

Then, for the purpose of this paper, the error in the frequency domain represents the difference

between amplitudes of desired ideal (input) and real (output) of the harmonic motion: e A = A o (ω) –

A i (ω) with the difference in phases ϕ(ω), as for the planar mechanisms is shown in Fig 7

Fig 7 Representation of the amplitude and phase errors

The ideal system represent the device where stiffness coefficients and damping coefficients in

both moving axes have the same values; i.e it is for both ratios: k p = k 1 /k 2 = 1 and b p = b 1 /b 2 = 1,

and k 12 , k 21 = 0 and b 12 , b 21 = 0

The influence of unequal values of stiffness and damping coefficients on the output amplitude and phase errors for two axis positioning system is analyzed in details in [12] Here, the amplitude error is expressed as the difference of amplitudes between two moving axes

where Re C and Im C are differences of real and imaginary components of frequency transfer functions

(6) for particular moving axes Thus, in case of a “real” positioning system, where k 12 and b 12 or k 21 and b 21 have non- zero values, solving (7) this amplitude error will be

6 2 8

2 6 2 2 2 2 6 2 2 2

2 2

2 2

1 1

i gi c g

fc ei eg

f e

k k m b

b m

+

− + +

− +

− +

− +

−

=

ω ω

ω ω

ω ω

(8)

where indexes 1, 2, 12 and 21 denote mutual stiffness and damping between corresponding axes, ω

is the operating frequency and c, e, f, g, i are auxiliary variables defined as follows

2 12 2 1 2

12 2 1 2

1

2 1 2

12 12 1 2

k k k i b mk mk

b

b

g

b b m f m e b k k

=

(9)

As can be seen in Fig 8, the amplitude error for diverse stiffness in coefficients, i.e k p ratios depends on motion frequency, but the values of this error are relatively small (less than x 10-4) and can be neglected On the other hand; as depicted simulation in Fig 9, under some frequencies of motions the amplitude errors due to diverse damping in particular directions have non-negligible values These errors can reach for some frequencies the values about 1000 times greater [12]

0 1 2 3 4

x 10-40.5

1 1.5 2 2.5 3 3.5 4 4.5 5

Fig 8 The dependence of amplitude errors on different stiffness for various motion frequencies

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.5

1 1.5 2 2.5 3 3.5 4 4.5 5

Some design considerations

The natural frequency of oscillations is the parameter that relates mass, stiffness and damping of a mechanical system Knowledge of natural frequency of mechanical oscillations should naturally correspond to the frequency of the control system In general, the natural frequency of the system should be adequately out / far of the motion / operation frequency In cases of compliant mechanisms there are two contradictory requirements that should be compromised:

- The compliant (mass-spring) mechanisms should be designed with minimal stiffness (in joints and the structure, as whole) in order to minimize dead / energy of elasticity need for flexural motions This requirement of efficiency naturally leads to design of mechanics with lower natural frequencies

- High operation speeds and motion frequency leads to more stiff mechanical structures

Both these facts confirm the need of careful application of techniques for modeling and simulation, yet in design phases with verification of the expected performance

Following the above discussion both static and dynamic errors should be compensated in precise positioning devices For both kinds of errors two mathematical models can be built The static - model includes compliance characteristics of the mechanical structure and the desired control variables are feed-forwardly corrected In principle it can include models for elimination of thermal distortions Compensation of errors due to system dynamics can be made by additional correction block, as depicted control scheme in Fig 10 for two axis system

Micro-robotic system

Reference

input

System output

fS(s)

Static error compensation2

a(s) b(s) Reference axis Output y

Output x

eD(s) fD(s) Dynamic error compensation

Fig 10 One possible control scheme of the positioning device

To simplify the design process, modeling and simulation of compliant devices some available

SW tools can be applied Own approach using MATLAB and his SimMechanics toolbox with development of specific mathematical models of joints and mechanisms is presented in [16, 17] Beside possibility to optimize the geometry of elastic joints it enables to achieve the output kinematic and dynamics characteristics of the whole mechanisms Some examples of modeled mechanisms are on the Fig 11

Acknowledgments

This paper presents the research work supported by the national scientific grant agency VEGA under project No.:2/0006/10 “Construction and control of micro-electro-mechanical elements and devices”

[4] L L Howell, Compliant Mechanisms, Wiley-IEEE, 2001,ISBN 047138478X, 459 p

[5] T.S Smith, Flexures: elements of elastic mechanisms, Gordon and Breach Science Publishers,

[8] Š Havlík, Passive compliant mechanisms for robotic (micro) devices in: 13th World Congress

in Mechanism and Machine Science, Guanajuato, Mexico, 2011, p 1-7

[9] Š Havík, Smart mechanisms for robotic devices, in: 21th International Workshop on Robotics

in Alpe-Adria-Danube Region, Napoli, Italia, 2012, ISBN 978-88-95430-45-4, p 216-222

[10] Š Havlík, Analysis and modeling flexible robotic (micro) mechanisms in: Proceedings of the 11th World Congress in Mechanism and Machine Science Tianjin, April 1-4, 2004, Vol 3, PR of China, pp.1390-1395

[11] Š Havlík, G Carbone, Design of compliant Robotic Micro-Devices, in: 15th International Workshop on Rotobics in Alpe-Adria-Danube Region, RAAD 2006, June 15 – 17, 2006, Balatonfured, Hungary, ISBN: 963-7154-48-5

[12] J, Hricko, Analysis, modeling and optimization of compliant mechanisms design, in: PhD dissertation, FEI STU Bratislava, 2010, (In slovak)

[13] Y K Yong, T-F Lu, Kinetostatic modeling of 3-RRR compliant micro-motion stages with flexure hinges, Mechanism and Machine Theory, vol 44 (2009), p 1156–1175

[14] B.-J Yi, et al., Design and Experiment of a 3-DOF Parallel Micromechanism Utilizing Flexure Hinges, IEEE Transactions on robotics and automation, vol 19, no 4, august 2003

[15] P.R Ouyang, A spatial hybrid motion compliant mechanism: Design and optimization, Mechatronics, vol 21, (2011), p 479–489

[16] J Hricko, R Harťanský, Š Havlík, Modeling Compliant Mechanical Joints for Micro-Robotic Devices, in: RAAD 2009: 18th International Workshop on Robotics in Alpe-Adria-Danube Region, Brasov, Romania, Bucuresti: Printech, ISSN 2066-4745

[17] J Hricko, Modelling Compliant Mechanisms – Comparison of Models in MATLAB / SimMechanics vs FEM, in: 21th International Workshop on Robotics in Alpe-Adria-Danube Region, Napoli, Italia, 2012, ISBN 978-88-95430-45-4, p 57-62

Special Gripping Elements for Handling with Flat Objects

Marcel Horák1, a, František Novotný2, b

1, 2 Technical University of Liberec Institute for Nanomaterials, Advanced Technologies and Innovation

Studentská 2, 461 17 Liberec 1, Czech Republic

a marcel.horak@tul.cz, b frantisek.novotny@tul.cz

Keywords: Handling, vacuum, gripping element, radial loading, adhesion

Abstract The paper analyses possibilities to use new materials having a high degree of adhesion for

designing vacuum gripping elements with a view to increase the radial capacity and at the same time

to preserve the compressed air consumption or the vacuum level Structural modification of the suction cup is presented with a bearing supporting plate having a material with an adhesion layer [1,

2, 3, 4] on the contact boundary and allowing the down-pressure to be regulated depending on mechanical properties of the object kept In conclusion the results achieved are summarized illustrating a marked increase of the radial load capacity with regard to the defined vacuum level and the degree of shifting in the contact profile compared with the standard suction cups with a similar geometrical diameter in relation to the contact surface

Introduction

In connection with the new technological processes there is an unprecedented problem, namely the demand to realize the gripping process on one side of the plate only without any possibility to make use of e.g its edges at the most unfavorable handling modes in the vertical direction and the contact plane when the suction cups are stressed solely in a radial way, eventually in a combined one So distributed loading brings marked changes in the contact area geometry (contact profile shape) owing to the elastomer friction, adhesive and material properties and the vacuum degree These changes affect very much the contact stability, and finally they can lead up to the collapse of gripping [5, 6, 7]

Authors’ aim was to design and laboratory-verify new gripping element facilities This element combines principles of vacuum and adhesion and so it gives a guarantee that a higher level of the safe load of the element will be achieved in the radial direction, i.e in the contact plain, when keeping energy demands (the pressure air consumption)

Concept of combined gripping element

The designed solution (Fig 1) combines the vacuum gripping element (GE) with a rigid flange, an elastic sealing border and a move-out adjustable plate coated with an adhesive layer

The proposed element is composed of the basic body 1 fitted with the flexible sealing rim 2 along its periphery When contacting with the handled object, this rim produces a vacuum-tight cavity in which – when the vacuum source is connected through the hole 6 – it is possible to control the vacuum level In the element body, the bearing plate 4 is inserted treated by the adhesive layer 5 The plate together with the layer can be positioned arbitrarily owing to the contact plane by means

of a locking screw 3 At the same time, the plate 4 together with the layer 5 constitutes a plane contact surface 7

Taking into consideration that the adhesive layer positioning using a threaded connection with the element body is insufficient, complicated, and because of the element different geometry it is necessary to define always its optimal position (down-pressure), the initial solution was replaced by

a variant having automatically adjustable position of the bearing plate [8, 9]

Principle is as follows: after the gripping element 1 is contacted with the handled object 7, a produced space 10,

11 sealed by means of the sealing rim 2 is evacuated through the hole 8, and simultaneously the bearing plate 4 is put optimally into contact with the object 7 through a piston

3 which is connected with the bearing plate 4, and in the same time it seals the space 10 An integral part of the plate

4 is the adhesive layer 5 forming a contact boundary 12 together with the object 7 A down-pressure level between the plate 4 or layer 5 and the object 7 can be control by the pressure value in a control space 9 above the piston 3 In this way an optimal level of the down-pressure can be achieved depending on a deformation of the rim 2, mechanical properties and the surface profile of the object 7 and so a maximum grow of friction forces during handling can be provided

The technical solution concept is illustrated in Fig 2 which comprises a detailed section of the gripping element configuration with a rigid threaded connection 6 of the bearing plate and the piston Fig 3 shows the solution enabling to adjust automatically the adhesive layer or the bearing plate orientation depending on an orientation of the object contact surface (the plate and the piston are connected through a ball joint 13) in position when the adhesive insert is out of the contact with the object handled

Fig 2 Rigid connection of bearing plate Fig 3 Flexible connection of bearing plate

Laboratory verification of gripping element function

The designed gripping element functioning was tested under laboratory conditions, and results obtained were evaluated in view of the usual suction cups having similar geometric diameter (Fig 4)

The tests were carried out using various vacuum levels, and subsequently the rigid flange displacement was evaluated for two states of a contact surface It is obvious that any contamination

of the contact surface of an object handled reduces substantially the friction factor, which logically results in a loss in the level of friction (radial) forces transferred by both the vacuum and combined gripping elements

Fig 1 Vacuum gripping element

Fig 4 Tested gripping elements

Based on the extensive series of experimental data and their detailed analysis, it is possible, as far

as the given gripping elements are concerned, to state that the tested suction cup FCF75P with the geometric diameter 75 mm, produced by the PIAB, is only very little sensible to the contact surface contaminated with water, and it shows stable values of carrying capacity in the radial direction changing only minimally (Fig 5, DS – dry surface, WS – surface contaminated with water) On the other side, at low FRAD values, the values of the rigid flange displacement are relatively high owing

to an external loading force, which is especially connected with distinctive deformation behavior of the patented material on a polyurethane basis marked DURAFLEX®

Rigid flange displacement of vacuum GE [mm]

20 kPa 30 kPa 40 kPa 50 kPa 60 kPa 70 kPa 80 kPa

Fig 5 The level of friction force depending on the condition of the contact surface (PIAB FCF75P)

Unlike two other selected representatives being tested, the SCHMALZ suction cup PFYN80 (geometric diameter 80 mm) is very sensible to the surface contaminated with water From the graphic output in Figure 6 (DS – dry surface, WS – surface contaminated with water) it is obvious that the carrying capacity goes down up to a third of initial values being measured in the regime of the dry and clean contact surface, which is caused by very flat character of the suction cup contact surface

In view of the geometric diameter mere 80 mm, the designed vacuum gripping element, having the legal protection according to UV 22075, has compared with the PIAB and SCHMALZ suction cups very interesting displacement values depending on an external radial loading force both at dry and wet surfaces

Rigid flange displacement of vacuum GE [mm]

20 kPa 30 kPa 40 kPa 50 kPa 60 kPa 70 kPa 80 kPa

Fig 6 The level of friction force depending on the condition of the contact surface (SCHMALZ

Rigid flange displacement of vacuum GE [mm]

20 kPa 30 kPa 40 kPa 50 kPa 60 kPa 70 kPa 80 kPa

Fig 7 The level of friction force depending on the condition of the contact surface (UV 22075)

Moreover, at displacement values in the range 0.5 – 1.0 mm, the sensibility to a contamination is only minimal, and in the event of the displacement value increasing up to the contact collapse level, the carrying capacity is reduced about one third Further it is possible to claim that the given gripping element reaches approximately the same level of the carrying capacity in the radial direction as the suction cup FCF75P in spite of that the geometric diameter is smaller by 15 mm (Fig 7, DS – dry surface, WS – surface contaminated with water)

The gripping element according to the given technical solution is usable within the wide range of handling and gripping processes having higher demands on safety of the object gripping and keeping It is advantageous e.g to handling flexible objects (minimization of elastic deformations of objects having a low lateral rigidity), handling with the gripping plane vertically oriented, etc It is suitable for uneven and rough surfaces, and generally for applications where external forces are acting in parallel with the gripping plane This solution also minimizes the compress air consumption, and simultaneously it maintains a force response level Fig 8 illustrates the proposed gripping element 2 in and assembly designed for mounting to a frame of a multi-element gripping head in combination with an orientation compensator 3 and a position compensator 4 (the adhesive layer is in contact with the object handled 1)

Fig 8 Gripping element with position compensator

Summary

The paper submitted describes possibilities how to take advantage of adhesive layers when designing gripping elements Its main part was concentrated on the problems of an increase carrying capacity of gripping elements in the radial direction for the reason of completely new demands for vacuum gripping heads relating to new production technologies and processes The vacuum-adhesive gripping element was designed and laboratory-tested During tests it was shown that the use of adhesive layers results in the radial carrying capacity rise in order of tens per cent in comparison with standard solutions when the vacuum level is kept Logically it follows from this that the designed system can be operated in so-called passive regime, in which the gripping force is defined by only an adhesion degree In conclusion the possibility to use the designed element in combination with a standard position compensator was presented

Presented gripping element is mainly designed for handling with flat objects without contact polluted surfaces The function element has not been tested in a dusty or dirty environment and it can be assumed that the contaminated surface will negatively affect the level of the gripping forces

http://onlinelibrary.wiley.com/doi/10.1002/adma.201104191/pdf

[2] Sethi, S., Ge, L., Ci, L., Ajayan, P M and Dhinojwala, A Gecko-Inspired Carbon Based Self-Cleaning Adhesives Nano Letters, 2008, Vol 8, No 3, pp 822-825, http://www2.uakron.edu/cpspe/dhinojwala/pdfs/SeS-2008.pdf

Nanotube-[3] Ge, L., Sethi, S., Ci, L., Ajayan, P M., and Dhinojwala, A Carbon nanotube-based synthetic gecko tapes PNAS, June 26, 2007, vol 104, no 26, pp 10792–10795, http://www2.uakron.edu/cpspe/dhinojwala/pdfs/GeL-2007.pdf

[4] Giannakopoulos, A E., Venkatesh, T A., Lindley, T C and Suresh, S The Role of Adhesion

in Contact Fatigue Acta mater Vol 47, No 18, 1999, pp 4653-4664, http://nanomechanics.mit.edu/papers/Role_Adhesion_99.pdf

[5] Horák, M., Novotný, F Deformation Analysis of Suction Cup under Combined Load In proceedings: 10th International Carpathian Control Conference Zakopane: University of Science and Technology Poland, 2009, p 403 - 406, ISBN 83-89772-51-5

[6] Novotný, F., Horák, M Problems of Loading Capacity of Suction Cups in the Radial Direction In: Proceedings of the Fifth International Conference on Optimisation of the Robots and Manipulators, Calimanesti, Romania, 2010, p 70-74, ISBN 978-981-08-5840-7, doi: 10.3850/978-981-08-5840-7_S2-4

[7] Horák, M., Novotný, F Increases in the Radial Capacity of Vacuum Gripping Elements MM Science Journal, October 2011, Special Edition - 20th International Workshop on Robotics in Alpe-Adria-Danube Region (RAAD), October 5-7 2011, MKČR E 7645, www.mmscience.eu, p 122-

127 ISSN 1803-1269 (Print), ISSN 1805-0476 (On-line)

[8] Horák, M., Novotný, F Podtlakový úchopný prvek Užitný vzor, číslo zápisu 22075, 18 4

2011, MPT B 65 G 47/91, Úřad průmyslového vlastnictví ČR (www.upv.cz)

[9] Horák, M., Novotný, F Podtlakový úchopný prvek Patent, číslo 302959, 14 12 2011, MPT B

65 G 47/91, B 65 H 5/14, B 60 R 9/058, Úřad průmyslového vlastnictví ČR (www.upv.cz)