Introduction to Artificial Intelligence CS 438 Spring 2008

Download Report

Transcript Introduction to Artificial Intelligence CS 438 Spring 2008

Introduction to Artificial Intelligence
CS 438 Spring 2008
• Today
– AIMA, Ch. 25
– Robotics
• Thursday
– Robotics continued
• Home Work due next Tuesday
– Ch. 13: 13.6 a – d
– Ch. 14: 14.1 a – d
BigDog Robot
Robotics Logistics
• We have enough robots for every 2 people
• Pair programming
• Find a partner by Thursday
– Preferably one that has your has a schedule
similar to yours
Why Study Robotics?
Where are the Robots?
• Exploration
Where are the Robots?
• Industrial Robots
Where are the Robots?
• Medicine
Where are the Robots?
• Service Robots
Where are the Robots?
• Consumer Robots
Where are the Robots?
• Cultural Robots
Advances in AI and in Robotics are one and the same.
And It’s Fun 
What is a Robot?
• An autonomous system which exists in
the physical world, can sense its
environment, and can act on it to
achieve some goals.
Living Autonomously
• An autonomous robot acts on its own
decisions
• Robots are not directly controlled by
humans
– Can take input and advice from humans
• Robots are not teleoperated
– Making them much more difficult and
interesting than Battlebots
The Physical World is Harsh
•
•
•
•
•
Partially observable
Stochastic
Continuous
Dynamic
Multi-agent (typically)
Dealing with the Physical World
A robot needs to be able to handle its environment
or the environment must be altered and
controlled.
• Close World Assumption
– The robot knows everything relevant to performing
“Complete World Model”
– no surprises
• Open World Assumption
– The robot does not assume complete knowledge
– The robot must be able to handle unexpected events.
Sensing the Environment
• Sensors allow the robot to perceive its
environment to get information that
allows it to make decisions
– Humans have 6 senses what are they?
– What sensors does a robot need?
Movie Clip
Acting on the Environment
• Robots have effectors that allow it to
change the state of the world
– What are human effectors?
– What effectors can robots have?
Movie Clip
Movie Clip
Achieving a Goal
• Achieving a goal requires intelligent
decision making
– Artificial Intelligence
Movie
Spectrum of Self-Control
Teleoperation: Human Control
Autonomous (AI) Control
Shared Human – Robot Control
Spectrum of Robot Control
Autonomous Mobile Robots have to Solve
Difficult Problems
• Where am I?
– Localization Problem
• How do I get there?
– Path Finding Problem
• How do I find the door?
– Object Recognition Problem
• What are you asking me to do?
– Language Understanding Problem
• How can I tell you the answer to your question?
– Speech Generation Problem
What is a Robot Control Architecture?
how much of the world do we need to represent internally ?
Robot Architecture
how should we internalize the world ?
what inputs do we have ?
what outputs can we effect ?
what algorithms connect the two ?
how do we use this “internal world” effectively ?
Robot Control Architecture
Deliberative/Hierarchical Robot Control
• Emphasizes Planning
• Robot senses the world, constructs a model
representation of the world, “shuts its eyes”,
creates a plan of action, makes the action,
then senses the results of the action.
sense
...
Stanford Cart
1968
1976
plan
ACTING
motor control
task execution
planning
world modeling
perception
Shakey
SENSING
Sense - Plan - Act
act
MERs
2003-
...
Deliberative: Good & Bad
• Goal Oriented
– Solve problems that need cognitive abilities
– Ability to optimize solution
• Predictable
• Dependence on a world model
– Requires a closed world assumption
• Frame Problem
• Symbol Grounding Problem
Reactive/Behavior-Based Control
• Ignores world models
• “The world is its own best model”
Sense
• Tightly couples perceptions to actions
– No intervening abstract representations
• Primitive Behaviors are used as building
blocks
– Individual behaviors can be made up of
primitive behaviors
• Reactive: no memory
• Behavior-Based: Short Term Memory (STM)
Act
Sensing and Acting
• Reactive Paradigm tightly couples
perceptions to actions
– No intervening abstract representations or time
history
• Individual Behaviors are used as building
blocks
Stimulus
Behavior
SR Diagram
Response
Behavior Coordination
• If multiple behaviors are possible which one does the
robot do?
– Competitive coordination: winner-take-all
– Cooperative coordination: behavioral fusion
– Combination
Where does the overall robot
behavior come from?
• No overall goal, no planning
• Emergent Behavior
– Emergence is the appearance of a novel property of a
whole system that cannot be explained by examining
the individual components, for example the wetness
of water.
– Overall behavior is a result of robots interaction with
its surroundings and the coordination between the
individual behaviors.
Reactive: Good & Bad
• Works with the Open World Assumption
– Provides a timely response in a dynamic
environment where the environment is difficult
to characterize and contains a lot of
uncertainty.
• Unpredictable
• Low level intelligence
– Cannot manage tasks that require memory
and higher level cognition
• Tasks requiring localization and order dependent
steps
Hybrid Paradigm
• Lyons 1992
• Combines Reactive and Deliberative Control
Planner
Sense
Act
Planning – Reactive Interaction
• Reactive is primary control and Planner
provides advise
– Planner configures the Reactive system
• Planner is primary and Reactive provides
actions to avoid uncertain situations
– Layered approach
– Requires re-planning
• Planner and Reactive work concurrently