CERMA 2007 / Tutorial Autonomous Vehiclesosorio.wait4.org/oldsite/palestras/CERMA07-Vehicles-2pp.pdf · CERMA 2007 / Tutorial Control: Intelligent Autonomous Vehicles Low Level Autonomous

1

Grupo de Pesquisas em Veículos Autônomos

Autonomous Vehicles Research Group - Unisinos

CERMA 2007 / TutorialCERMA 2007 / Tutorial

11

Autonomous Vehicles:Autonomous Vehicles:Research, Design and Implementation of Research, Design and Implementation of

Intelligent Autonomous VehiclesIntelligent Autonomous Vehicles

Autonomous Vehicles Research Group Autonomous Vehicles Research Group -- GPVAGPVAhttp://http:// www.eletrica.unisinos.br/~autonomwww.eletrica.unisinos.br/~autonom

Tutorial page: http://inf.unisinos.br/~osorio/pales tras/cerma07.Tutorial page: http://inf.unisinos.br/~osorio/pales tras/cerma07. htmlhtml

Dr. Fernando S. Dr. Fernando S. OsórioOsório -- Applied Computing Applied Computing PostPost --Grad. ProgramGrad. Program PIPCAPIPCADr. Christian R. Dr. Christian R. KelberKelber -- Electrical Engineering / Computer Eng.Electrical Engineering / Computer Eng.Dr. Dr. CláudioCláudio R. Jung R. Jung -- Applied Computing Applied Computing M.ScM.Sc. Program PIPCA. Program PIPCAM.ScM.Sc. . FarleiFarlei HeinenHeinen -- Computer Engineering B.Sc. (Director)Computer Engineering B.Sc. (Director)


Intelligent Autonomous Vehicles

Intelligent Autonomous VehiclesIntelligent Autonomous Vehicles

• Introduction• Robotic: Automatons, Mobile Robots and Autonomous R obots� Perception, Action, Locomotion e Communication� Control and Intelligence• Intelligent Vehicles� Technologies for Vehicle Automation� Control pyramid• Intelligent Control of Autonomous Vehicles� Control : Computational Architectures � Simulation of Autonomous Vehicles • Computer Vision

• Practical Applications

Agenda22

2



Control LayersControl Layers

ControlPyramid 33

[[KelberKelber etet al., IEEE ISIE 2005]al., IEEE ISIE 2005]




Control of Electrical and Mechanical Systems:• ABS Braking System• Traction Control • Active Suspension• Stability Control• Steering Systems• Drive-by-Wire Technology

44

ControlPyramid

3


Intelligent Autonomous Vehicles55

ControlPyramid


Videos-Robotica\atuador.mpg



ControlPyramid


4




CONTROL:CONTROL:ComputationalComputationalArchitecturesArchitectures

66

Low Level

Set Point

Control Loop

High Level

Task Control

Hierarchical ControlAdaptive ControlIntelligent Control

ControlPyramid



Control Layers Control Layers -- Low LevelLow Level

66

Low Level

ACC with Stop and Go

Joystick control interface

Distance Controller

Steering Controller / Lane Follow

Set Point

Control:Low Level

Control Loop

[[KelberKelber etet al., IEEE ISIE 2005]al., IEEE ISIE 2005]

5


Intelligent Autonomous VehiclesControl:Low Level

Joystick control interfaceSteering Controller / Lane Follow




CONTROL:CONTROL:ComputationalComputationalArchitecturesArchitectures

66

Low Level

Set Point

Control Loop

High Level

Task Control

Hierarchical ControlAdaptive ControlIntelligent Control

ControlPyramid

6


Intelligent Autonomous VehiclesControl:Low Level

Autonomous Parking

Darpa Challenge 2005 - Desert

Darpa Challenge 2007 - UrbanGo back and Enter Final Adjust

Search for a Parking Space Align with next car



• Sensorial Models• Kinematics Models

• Robotic Control: * Reactive* Deliberative* Hierarchical* Hybrid

• Environment Maps* Building Maps* Path Planning * SMPA - Sense Model Plan Act

• Problems: * Complex tasks* Avoid Obstacles: Static / Mobile - Unexpected obstacles* Robot actual position estimation - Where am I ?

CONTROLCONTROL: Computational Architectures: Computational Architectures

77

Control Architectures

7



Autonomous RobotsAutonomous RobotsSensors e ActuatorsSensors e Actuators

* Sensors: Environment Perception

* Actuators: Motors - Control actions and robot disp lacement

MIT - OCW88

Autonomous:Perception & Action




* Sensors:

Distance: Light, Sound, Touch Other Components

• Infrared > Battery charge meter

• Sonar (ultrasound) e Radar > Temperature, Pressure

• Laser > Humidity

• Video Cameras - Linear / Matrix (CCD), Mono or Binoc ular > Smoke.

• Contact Sensor (bumpers, "cat whiskers") > Smells, etc

Positioning and Orientation

• GPS

• Compass

• Gyroscope

• Odometer

• Beacons (ex. radio beacons) or Video Cameras

* Actuators...

99

Autonomous:Perception

8



* Sensors:

Distance : Light, Sound, Touch Other Components




• Video Cameras - Linear / Matrix (CCD), Mono or Binoc ular > Smoke.


1010





* Sensors:

Distance : Light, Sound, Touch Other Components




• Video Cameras - Linear / Matrix (CCD) > Smoke.



9




* Sensors:


• GPS

• Compass

• Gyroscope

• Odometer

• Beacons (ex. radio beacons)

• Video Cameras





* Sensors:


• GPS

• Compass

• Gyroscope

• Odometer

• Beacons (ex. radio)

• Video Cameras


10



* Actuators:

Locomotion :

• Step motor / DC Motors: Direction (Steering), Veloc ity (Speed)

Wheels, Tank Tracks, Propulsion, Legs

Tricycle steering, Differential drive, Skid steer,

Ackermann steering, Synchro drive, ...

• Walking: Equilibrium problem / Gait control

• Propulsion: Aquatic, Underwater, Aerial

Robotic Arms : Grippers, Lifters, Fingers, ...

1414


Autonomous:Action



* Actuators:

Locomotion :

• Step motor / DC Motors: Direction (Steering), Veloc ity (Speed)

Wheels, Tank Tracks, Propulsion, Legs

Tricycle steering, Differential drive, Skid steer,

Ackermann steering, Synchro drive, ...

1414


Autonomous:Action

Kinematics Model

Velocity vSteering wheel angle ΦPosition (X,Y)Orientation θ

11



Intelligent Autonomous RobotsIntelligent Autonomous Robots<< Intelligence >><< Intelligence >>

* Action Planning

* Ability to Perceive the Environment

* Ability to Decide

* Ability to Act

* High Level Tasks Planning

* Reaction: Sensorial-Motor Integration

* Estimate Actual and Future States: Environment + Behavior = Interaction

* Adaptation and Learning

* Robustness: Unexpected Situations

=> From where do I start ???

1515

AutonomousRobots ?


Intelligent Autonomous VehiclesAutonomousRobots ?

Robótica Autônoma InteligenteRobótica Autônoma InteligenteInteligência e RobóticaInteligência e Robótica

1616

* Action Planning

* Ability to Perceive the Environment

* Ability to Decide

* Ability to Act

* High Level Tasks Planning

* Reaction: Sensorial-Motor Integration

* Estimate Actual and Future States: Environment + Behavior = Interaction

* Adaptation and Learning

* Robustness: Unexpected Situations

=> From where do I start ???

12


Intelligent Autonomous VehiclesControl: ComputationalArchitectures 1717

• Sensorial Models• Kinematics Models

• Robotic Control: * Reactive* Deliberative* Hierarchical* Hybrid

• Environment Maps* Building Maps* Path Planning * SMPA - Sense Model Plan Act

• Problems: * Complex tasks* Avoid Obstacles: Static / Mobile - Unexpected obstacles* Robot actual position estimation - Where am I ?

CONTROLCONTROL: Computational Architectures: Computational Architectures



Documents

CERMA 2007 / Tutorial Autonomous Vehiclesosorio.wait4.org/oldsite/palestras/CERMA07-Vehicles-2pp.pdf · CERMA 2007 / Tutorial Control: Intelligent Autonomous Vehicles Low Level Autonomous