o to thong minh

In the hurry driving situation both headway distance and velocity errors are smaller Human accelerator pedal operations are wavier 0 50 100 Time [s] Driver model... Flat and straight ro

Ô TÔ THÔNG MINH

TS ĐÀM HOÀNG PHÚC 2012.4.14 　 SƠN TÂY

Bộ môn Ô tô và xe Chuyên dụng

Đại học Bách khoa Hà nội

INTELLIGENT VEHICLE

• Investigate the motivation of adding Intelligence to a car.

• Explore problems and solutions.

• Survey the current state of research.

• Identify future research trends.

• Definitions / Motivation

• Design Goals

• Problems / Solutions - Theory

• Current Industry Solutions

• Future Trend

Integrated Vehicle Dynamics Control

Active safety devices (driver assistance systems) for reducing road accidents

Controller

Driver-Vehicle Cooperation

Background

Motivation : Integrated Vehicle Technology and Robotics

Yaw moment generated

by Lane keeping system

Skidding avoidance is needed to secure vehicle stability during transient steering manoeuvre.

Driver steering behaviour adaptation

・　 lane keeping task : lane keeping control activates

・ lane changing : stability control activates

 Machine Intelligence : to provide service with respect to the operation task

 Sensor information is used to recognize the operation modes /behaviour

■ 　 Environment adaptation

■ 　 Driver adaptation

■ 　 Conventional driver assistance systems use average characteristics of drivers.　　　　 　 　→　 They do not fit the driver preference and cause sense of discomfort

　　　 　→　 Accident prevention effect cannot fully be obtained as expected

Traffic environment (map, vicinity vehicles) Driver state (drowsy, tired, hurry)

Motivation and Objectives

(Dangerous)

(Cautious)

0 0.5 1 1.5 2 2.5 3 0

Driving Database Storage &

Driver Behavior Analysis

Deceleration point at intersections

Following or Stopping etc.

+ _

0 0.5 1 1.5 2 2.5 3 0

Driving Database Storage &

Driver Behavior Analysis

Deceleration point at intersections

Following or Stopping etc.

+ _

Based on long-term naturalistic driving data of a driver

in actual traffic condition, the following algorithms for synthesizing DA are designed

(1)”Feature of usual driving behavior”, (2)“Unusual behavior detection and prediction” ,(3)“Individual adaptation of driver assistance system”

Schematic diagram of driver assistance system

Driver Assistance Systems

• Night Vision

• Adaptive Cruise Control

• Collision Warning

• Collision Avoidance

• Driver Impairment Monitoring

• Advanced Driver Assistance

• Cooperative Infrastructure

• Automated Driving

TTLC 　：　 Time 　 To 　 Line 　 Crossing 　 [s]

Forward collision warning system

　 warni ng

brake on

• Provide the info./warning/maneuvers to the driver at proper timing and intensity

• Adaptive in real-time to traffic environment & individual drivers' characteristics

Today your headway distance

is shorter than usual

It is not your normal behavior.

Adaptive region

Collision Warning Braking Intervention Safe Distance Advisory

Driver assistance system

10 s

Normal driving Hazardous

Post-acc.

Critical (pre-crash)

Time-to-Collision

0 s

1 s

Collision

Continuous Sensing and Individual User-Adaptive Mechatronics

 Advanced Driver Assistance Systems (Car-Robotics Technology)

In the hurry driving situation both headway distance and velocity errors are smaller

Human accelerator pedal operations are wavier

0 50 100

Time [s]

Driver model

Normal Hurry

Brake pedal Vehicle

Vehicle velocity

Acceleration

To propose the methodology of driving maneuver (state) recognition

algorithm based on statistical machine learning technique from database

State of the Art of Driver Model/Driver Manoeuvre Recognition

Following Braking

Cruising

Decelerating

Stopping

Preceding vehicle Host vehicle

Following a leading vehicle

Stopping

Braking for forward vehicle collision avoidance

Independently free driving

Deceleration at stop line or red signal

Classification of driving behavior (longitudinal direction)

ACC, FSRA

FCWS, PCS

ISA

Stop Assistance

①Camera/Image Synthesizer

②Global Positioning System

Rear

6-image synthesizer

③ CAN Signal Probe

Steering wheel angle, Accelerator pedal,

Brake pedal, Winker,

Speed, Accelerations, Yaw rate, …

Headway distance, Relative speed

CAN-probe OBD connector

NTSC/DV conversion

RS/USB conversion

RS/USB conversion

USB/HUB

Laptop-based Logger 　Captured VDO image

Front Front left Front right Face Foot

Sensor Network of Continuous Sensing Drive Recorder

Digital data from on-board vehicle

sensors

Image data

Vehicle position data(GPS)

Ground TruthEstimated

50 100 150 200 250 300 350 400

50 100 150 200 250 300 350 400

0

0 50 100 150 200 250 300 350 400 0

50 100

0 50 100 150 200 250 300 350 400 0

5

0 50 100 150 200 250 300 350 400 -10

0 10

0 50 100 150 200 250 300 350 400 -2

0 2

0 50 100 150 200 250 300 350 400 -0.5

0 0.5

Inverse of time to collision

Time headway Relative velocity

Longitudinal acceleration

Time [s]

V R

 　 Low computational cost

with expectable high accuracy

 　 Easy to interprete the

physical meanings of the model

Driver-Vehicle-Environment modeling based on Boosting Sequential Labeling

Comparison between the ground truth (reference) and the estimated state

Road sections ： R2

Comparison between the ground truth (reference) and the estimated state

Flat and straight road, =0.8

Hurry Driving ： 5 times

Driver model

uBe　

　 DS　Vehicle　 Model　 (C P ++ P )　

Engine　 Model　

Motor　 Model　

+　 +　

Human longitudinal control driver-vehicle system in closed-loop

Five drivers conducted experiments

In the hurry driving situation both headway distance and velocity errors are smaller

Human accelerator pedal operations are wavier

0 50 100

Time [s]

Driver model

The relationship between velocity tracking performance with fuel economy

Ｖf following vehicle velocity

Ｖp Preceding vehicle velocity

Better in velocity tracking performance will give better fuel economy

Velocity Tracking Performance

The average velocity error

There are strong relationship between accelerator pedal operation characteristics and velocity tracking performance with fuel economy Therefore the driver model need to describe driver’s accelerator pedal operation and represent of vehicle control process.

Accelerator pedal variance P hm_var [%]

Vehicle　

L ongitudinal　 Model　

The driver attempt to diminish distance error

between actual headway distance and his

desired headway distance

R hw T

： Relative velocity

： Headway distance error

： Accelerator pedal stroke

： Preceding vehicle velocity

a P

Vehicle　 Longitudinal　 Model　

C Preceding velocity feed-forward gain

Relative velocity feedback gain

0 20 40 60 80 100 120 0

20 40 60 80

In normal driving situation, the

time headway, T hw are large and

driver control gain H x and H V are small

In hurry driving situation T hw are

small and driver control gain H x and H V are large.

Verify the accuracy of the driver model

NAGAI　LAB

Driver model

Identified Driver Model Parameters

15

0 50 100

0 20 40 60 80 100 120 0

The comparison of hurry driving simulation results and experimental data (D4)

In hurry driving situation, the headway distance and velocity errors of simulation results are well-fitted with the experimental data

Average velocity errors of simulation results are well-fitted with the experimental

results

Confirms correctness of the identified parameter values and ability of

representations of vehicle control process of proposed driver model

Comparison of average velocity error between the simulation results and experimental data of five drivers

The proposed driver model can be used to study of driver behavior

in a systematical driver-vehicle system.

Average error between simulation results and experimental data: 5.1%

NAGAI　LAB

Driver model

Comparison of fuel economy between the simulation results and experimental data of five drivers

Vehicle　

L ongitudinal　 Model　

Vehicle　 Longitudinal　 Model　

The headway distance gain, Hx

contributes stronger effect on fuel economy than relative

velocity gain, H V

Drivers who are good sensitive

in headway distance tend to give good fuel economy.