Behaviour-Based%20Control%20in%20Mobile%20Robotics

1
Behaviour-Based Controlin Mobile Robotics

• Paul Fitzpatrick

2
Definition of Robotics
Sensors
Perception
Robotics is the Intelligent Connection of
Perception to Action
Reality
Robot
Action
Actuators
3
Classical Robotic Control

• Pipelined approach

Planning
Motor Control
Perception
Modelling
Task Execution
Actuators
Sensors
4
Practical Problems with Classical Control
Planning
Modelling
Task Execution
Delayed reaction to stop robot
Motor Control
Perception
Sensors
Actuators
Slow to respond
5
Research Problems with Classical Control

• Temptation to focus on model rather than reality

Other influences
Perception
Modelling
Reality
Planning
Task Execution
Motor Control
6
Classical Control Summary

• Assumes a complete internal world model can be
  built, then manipulated
• Slow. Modelling must occur before robot can
  react to changes in environment
• Emphasis on using model is misleading-
• Makes complex tasks seem solvable by directing
  attention away from perception
• Makes potentially simple tasks complicated
• Requires crippling simplications

7
Alternatives to Classical Control
Traditional AI
complex cognition simple environment
simpler cognition complex environment
Cognitive complexity
behaviour- based systems
reactive systems
Environmental complexity
8
Behaviour-Based Control

• Rather than functional decomposition, use
  Behaviours
• As much as possible, behaviours interact with
  each other through the environment, not the
  system
• The world is its own best model, so consult it
  directly whenever practical
• Use distributed representations tailored to the
  particular behaviours using them

9
Behaviour-Based ControlSubsumption Architecture
Modules act in parallel, in layers of increasing
priority
Higher layers can view and modify the data flow
in lower layers
Work usefully
Use maps
Sensors
Actuators
Explore
All modules have access to sensors
All modules have access to actuators
Wander
Avoid obstacles
10
Advantages of Subsumption
Explore
Wander
Use maps
Reacts immediately to stop robot
Work usefully
Avoidance
Fast!
11
Disadvantages of Subsumption

• Rigid priority scheme and strict layering are
  limiting -
• Priorities must be evaluated and hardwired at
  design time
• Requires behaviours to fit into a simple
  single-inheritance hierarchy
• Behaviours cannot be combined, only enhanced
  linearly

12
Alternatives?
Extend scope of behaviour-based systems by
improving behaviour glue
Traditional AI
Cognitive complexity
behaviour- based systems
reactive systems
Environmental complexity
13
Lateral Architecture
All modules have access to sensors
All modules have access to actuators
14
Features of Lateral

• Lateral has a dynamic priority system designed to
  make it easy to build behaviours using other
  behaviours
• A behaviour is given the priority its highest
  priority user at a given time thinks it should
  have (sponsorship)
• Behaviours expose a limited public interface to
  users, rather than allowing access to their
  internal data flows.

15
Example Behaviour Edge Following
EdgeMan
oNudge
iSide
oMotor
Logical Sensors
16
Prowling Behaviour
Prowling uses edge following in its implementation
ProwlMan
oEdge
iSelect
oPatrol
Logical Sensors
oExplore
17
Prowling - Behaviours used
Prowling behaviour is built from a hierarchy of
simpler behaviours.
18
Prowling - Edge Following
ProwlMan active, and sponsoring edge following
(EdgeMan)
NudgeMan beats SeekMan because, even though it is
lower level, it has more sponsorship
19
Prowling - Exploring
ProwlMan active, and sponsoring exploration
(ExploreMan)
20
More Behaviours
New behaviours fit in transparently to the ones
already present
21
Competition
Competition between behaviours is resolved by the
sponsor
22
Lateral - Implementation
Extended syntax for expressing parallel processes
(as state machines) and data flow between them
Simulator
PROCESS ModuleA int localVariable INPUT(int,
boredom) INPUT(Point, distance) OUTPUT(MotorCt
rl, motor) OUTPUT(Message, msg) _at_CONTROL /
/ Priority control _at_state1 DoSomething()
NEXT state2 _at_state2 DoSomethingElse() N
EXT _at_
class ModuleA public ZACProcess private int
localVariable public ZACInputltintgt
boredom ZACInputltPointgt distance ZACOutputltMot
orCtrlgt motor ZACOutputltMessagegt msg int
ZAC_Run ( int ZAC_state ) ... ...
Runtime Support
C
Lateral
Robot
23
Hardware
Khepera

• Khepera miniature robot
• Processor 68332
• Proximity sensors
• Light sensors
• Stepper motors

24
Khepera Simulator
Robot Control Panel
Robot path tracer
25
(No Transcript)
26
Priority Implementation Connections