Intelligent Driving Agents Intelligent Driving Agents Microscopic - - PowerPoint PPT Presentation

▶

Mar 03, 2024 322 likes •474 views

Intelligent Driving Agents Intelligent Driving Agents Microscopic traffic simulation with reactive Microscopic traffic simulation with reactive driving agents driving agents ITS 2001 Conference, Oakland ITS 2001 Conference,

SLIDE 1

Delft University of Technology

Intelligent Driving Agents Intelligent Driving Agents

“Microscopic traffic simulation with reactive “Microscopic traffic simulation with reactive driving agents” driving agents”

ITS 2001 Conference, Oakland ITS 2001 Conference, Oakland Patrick Ehlert and Leon Rothkrantz Patrick Ehlert and Leon Rothkrantz August 28th, 2001 August 28th, 2001

SLIDE 2

TUDelft

1 1

Overview of presentation Overview of presentation

Project

Project

Design of driving agent

Design of driving agent

Implementation in prototype simulator

Implementation in prototype simulator

Results, conclusions and future work

Results, conclusions and future work

SLIDE 3

TUDelft

2 2

Goals:

Goals:

1. Perform reasoning from local viewpoint
1. Perform reasoning from local viewpoint
2. Investigate interaction between drivers
2. Investigate interaction between drivers
3. Create flexible and realistic traffic simulator
3. Create flexible and realistic traffic simulator
Used

Used tactical tactical-

level

level reasoning reasoning

Project Project

Create intelligent agent with human

Create intelligent agent with human-

like driving behavior driving behavior

SLIDE 4

TUDelft

3 3

Design: driving agent Design: driving agent

Perform human

Perform human-

like tactical driving

like tactical driving

Real

Real-

time control of vehicle

time control of vehicle

Expandable and flexible functionality

Expandable and flexible functionality

SLIDE 5

TUDelft

4 4

Behavior rules Sensors Communication

Vehicle Supervisor /

ther agents

Arbiter Controller & Memory

Environment

Design: driving agent Design: driving agent (continued)

(continued)

Parameters Behavior rules Collision avoidance Road following Overtaking Traffic lights Car following

SLIDE 6

TUDelft

5 5

Decided to create new prototype

Decided to create new prototype simulation program simulation program

Implementation Implementation: simulator

: simulator

Time

Time-

oriented simulator
riented simulator
Kinematic

Kinematic 2D motion model 2D motion model

SLIDE 7

TUDelft

6 6

Implementation: simulator Implementation: simulator

Simulation controller

User interface

1: update

Environment

Simulated objects

Traffic light controllers Traffic lights Vehicles Roads Intersections

Picture of environment

2: visual feedback

Timer

SLIDE 8

TUDelft

7 7

Implementation: agent Implementation: agent

Agents

b: send

rders

c: sleep

Reasoning Sensors

Environment

Simulated objects

Traffic light controllers Traffic lights Vehicles Roads Intersections a: get information

SLIDE 9

TUDelft

8 8

Implementation: agent’s rules Implementation: agent’s rules

Behavior

Behavior rules are directly coded into the rules are directly coded into the program for fast performance program for fast performance

example: example: If If (agent speed < preferred speed) (agent speed < preferred speed) then then Accelerate (normal) Accelerate (normal)

SLIDE 10

TUDelft

9 9

Implementation: example Implementation: example

SLIDE 11

TUDelft

10 10

Results Results

Simulation prototype with

Simulation prototype with

Up to 40 vehicles (agents) in real

Up to 40 vehicles (agents) in real-

time

time

Human

Human-

like driving behavior

like driving behavior

Interaction between drivers

Interaction between drivers

SLIDE 12

TUDelft

11 11

Conclusions Conclusions

Advantages agent

Advantages agent-

based simulation

based simulation

increased realism

increased realism

allows more flexibility

allows more flexibility

distributed processing possible

distributed processing possible

Disadvantages

Disadvantages

increase computational load

increase computational load

many parameters, more difficult validation

many parameters, more difficult validation

SLIDE 13

TUDelft

12 12

Future w ork Future w ork

Expand simulator and agent functionality

Expand simulator and agent functionality

Use distributed approach (more agents)

Use distributed approach (more agents)

Nanoscopic

Nanoscopic simulation simulation

Use agent model to control real vehicles

Use agent model to control real vehicles

More info at More info at http://www. http://www.kbs kbs. .twi twi. .tudelft tudelft. .nl nl/People/Staff/P.A.M.Ehlert/ /People/Staff/P.A.M.Ehlert/ai ai/project_IDA.html /project_IDA.html