AI Principles, Lecture on Planning Planning Jeremy Wyatt.

Slides:



Advertisements
Similar presentations
Language for planning problems
Advertisements

CSE391 – 2005 NLP 1 Planning The Planning problem Planning with State-space search.
Planning II: Partial Order Planning
Planning Module THREE: Planning, Production Systems,Expert Systems, Uncertainty Dr M M Awais.
Planning Module THREE: Planning, Production Systems,Expert Systems, Uncertainty Dr M M Awais.
CLASSICAL PLANNING What is planning ?  Planning is an AI approach to control  It is deliberation about actions  Key ideas  We have a model of the.
Situation Calculus for Action Descriptions We talked about STRIPS representations for actions. Another common representation is called the Situation Calculus.
Plan Generation & Causal-Link Planning 1 José Luis Ambite.
Planning Planning is fundamental to “intelligent” behaviour. E.g.
All rights reserved ©L. Manevitz Lecture 61 Artificial Intelligence Planning System L. Manevitz.
Sussman anomaly - analysis The start state is given by: ON(C, A) ONTABLE(A) ONTABLE(B) ARMEMPTY The goal by: ON(A,B) ON(B,C) This immediately leads to.
Planning Russell and Norvig: Chapter 11 Slides adapted from: robotics.stanford.edu/~latombe/cs121/2003/ home.htm.
Planning CSE 473 Chapters 10.3 and 11. © D. Weld, D. Fox 2 Planning Given a logical description of the initial situation, a logical description of the.
Artificial Intelligence II S. Russell and P. Norvig Artificial Intelligence: A Modern Approach Chapter 11: Planning.
Planning Some material adapted from slides by Tim Finin,Jean-Claude Latombe, Lise Getoor, and Marie desJardins.
Ch. 10 – Planning Supplemental slides for CSE 327 Prof. Jeff Heflin.
1 Classical STRIPS Planning Alan Fern * * Based in part on slides by Daniel Weld.
Artificial Intelligence 2005/06
Planning Copyright, 1996 © Dale Carnegie & Associates, Inc. Chapter 11.
Artificial Intelligence Chapter 11: Planning
Planning Russell and Norvig: Chapter 11. Planning Agent environment agent ? sensors actuators A1A2A3.
Planning Problem Solving  Planning –Action Centred –More flexible decision strategies A Representation for Planning –Add & Delete Lists Planning Techniques.
1 Lecture 12 example (from slides prepared by Prof. J. Rosenchein)
Artificial Intelligence 2005/06 Planning: STRIPS.
Planning: Part 1 Representation and State-space Search COMP151 March 30, 2007.
Planning Copyright, 1996 © Dale Carnegie & Associates, Inc. Chapter 11.
Intro to AI Fall 2002 © L. Joskowicz 1 Introduction to Artificial Intelligence LECTURE 12: Planning Motivation Search, theorem proving, and planning Situation.
1 Action Planning (Where logic-based representation of knowledge makes search problems more interesting) R&N: Chap. 11, Sect. 11.1–4.
PLANNING Partial order regression planning Temporal representation 1 Deductive planning in Logic Temporal representation 2.
PLANNING Partial order regression planning Temporal representation 1 Deductive planning in Logic Temporal representation 2.
Classical Planning Chapter 10.
Feng Zhiyong Tianjin University Fall planning.
1 07. The planning problem 2  Inputs: 1. A description of the world state 2. The goal state description 3. A set of actions  Output: A sequence of actions.
April 3, 2006AI: Chapter 11: Planning1 Artificial Intelligence Chapter 11: Planning Michael Scherger Department of Computer Science Kent State University.
22/11/04 AIPP Lecture 16: More Planning and Operators1 More Planning Artificial Intelligence Programming in Prolog.
CS.462 Artificial Intelligence SOMCHAI THANGSATHITYANGKUL Lecture 07 : Planning.
Constraint Satisfaction Problems (CSPs) CPSC 322 – CSP 1 Poole & Mackworth textbook: Sections § Lecturer: Alan Mackworth September 28, 2012.
Planning (Chapter 10)
CPS 270: Artificial Intelligence Planning Instructor: Vincent Conitzer.
CPS 570: Artificial Intelligence Planning Instructor: Vincent Conitzer.
Planning (Chapter 10)
Automated Planning and Decision Making Prof. Ronen Brafman Automated Planning and Decision Making Partial Order Planning Based on slides by: Carmel.
1 Chapter 16 Planning Methods. 2 Chapter 16 Contents (1) l STRIPS l STRIPS Implementation l Partial Order Planning l The Principle of Least Commitment.
Automated Planning Dr. Héctor Muñoz-Avila. What is Planning? Classical Definition Domain Independent: symbolic descriptions of the problems and the domain.
Introduction to Planning Dr. Shazzad Hosain Department of EECS North South Universtiy
AI Lecture 17 Planning Noémie Elhadad (substituting for Prof. McKeown)
1/16 Planning Chapter 11- Part1 Author: Vali Derhami.
Intro to Planning Or, how to represent the planning problem in logic.
Classical Planning Chapter 10 Mausam / Andrey Kolobov (Based on slides of Dan Weld, Marie desJardins)
Intro to CS 541 (AI Planning) Jim Blythe Jose Luis Ambite Yolanda Gil.
Computing & Information Sciences Kansas State University Monday, 23 Oct 2006CIS 490 / 730: Artificial Intelligence Lecture 25 of 42 Monday, 23 October.
Computing & Information Sciences Kansas State University Friday, 20 Oct 2006CIS 490 / 730: Artificial Intelligence Lecture 24 of 42 Friday, 20 October.
Planning I: Total Order Planners Sections
Planning in FOL Systems sequences of actions to achieve goals.
Artificial Intelligence 2004 Planning: Situation Calculus Review STRIPS POP Hierarchical Planning Situation Calculus (John McCarthy) situations.
Causal-link planning 1 Jim Blythe. 2 USC INFORMATION SCIENCES INSTITUTE Causal Link Planning The planning problem Inputs: 1. A description of the world.
Ch. 11 – Planning Supplemental slides for CSE 327 Prof. Jeff Heflin.
CLASSICAL PLANNING. Outline  The challenges in planning with standard search algorithm  Representing Plans – the PDDL language  Planning as state -
Planning (Chapter 10) Slides by Svetlana Lazebnik, 9/2016 with modifications by Mark Hasegawa-Johnson, 9/2017
Planning (Chapter 10)
Introduction Contents Sungwook Yoon, Postdoctoral Research Associate
Planning (Chapter 10)
AI Planning.
Planning José Luis Ambite.
Planning Chapter
Planning Problems On(C, A)‏ On(A, Table)‏ On(B, Table)‏ Clear(C)‏
Supplemental slides for CSE 327 Prof. Jeff Heflin
CS344 : Introduction to Artificial Intelligence
Russell and Norvig: Chapter 11 CS121 – Winter 2003
Prof. Pushpak Bhattacharyya, IIT Bombay
Presentation transcript:

AI Principles, Lecture on Planning Planning Jeremy Wyatt

AI Principles, Lecture on Planning Plan  Situation calculus  The frame problem  The STRIPS representation for planning  State space planning: Forward chaining search (progression planning) Backward chaining search (regression planning)  Reading: Russell and Norvig pp

AI Principles, Lecture on Planning Situation Calculus  The situation calculus is a general system for allowing you to reason automatically about the effects of actions, and to search for plans that achieve goals.  However, it has some flaws that took a long time to iron out…  The situation calculus is composed of Situations: S0 is the initial situation Result(a,S0) is the situation that results from applying action a in situation S0 Fluents: functions and predicates that may be true of some situation e.g. ¬Holding(G1,S0) says that the agent is not holding G1 in situation S0

AI Principles, Lecture on Planning Situation Calculus  Actions are described in the situation calculus using two kinds of axioms: Possibility Axioms: say when you can apply an action Effect Axioms: that say what happens when you apply them  The essential problem with situation calculus however, is that Effect Axioms only say what changes, not what stays the same, but reasoning in the situation calculus requires the explicit representation of all the things that don’t change. These are handled by Frame Axioms, and hence this is called the Representational Frame Problem. Reasoning about all the effects and non-effects of a sequence of actions is thus very inefficient. This is known as the Inferential Frame Problem.  Eventually solutions to both the Frame Problems were found. But in the meantime researchers began to look for representations to support more efficient planning.

AI Principles, Lecture on Planning The STRIPS representation  The Grandad of planning representations  Avoids the difficulties of more general representations (e.g. situation calculus) for reasoning about action effects and change  Devised for, and used in the Shakey project  The key contribution of the representation of action effects is to assume that anything that isn’t said to change as the result of an action, doesn’t change  It is this assumption (sometimes called the STRIPS assumption) that avoids the representational frame problem

AI Principles, Lecture on Planning STRIPS operators  All effects are modelled, anything that isn’t in the effects list stays the same  The effect P^¬Q adds P to the world description and removes Q  Variables in the operator (e.g. x) must be bound (unified) with entities in the description of the world to apply the operator, so Move(C,A,B) requires {b/C,x/A,y/B}, i.e. b is substituted by C etc. BA C BA C Action(Move(b,x,y))- Preconditions: On(b,x) ^ Clear(b) ^ Clear(y) Effects: On(b,y) ^ Clear(x) ^ ¬On(b,x) ^ ¬Clear(y)

AI Principles, Lecture on Planning STRIPS operators  STRIPS doesn’t allow us to use full first order logic  So we have to create a Clear(b) predicate, rather than use to indicate that there is no block x on block b  Also we actions like Move(C,A,A) create inconsistent effects  And we may need additional special operators e.g. Move_to_table(b,x,Table) whenever the effects are slightly different BA C BA C Action(Move(b,x,y))- Preconditions: On(b,x) ^ Clear(b) ^ Clear(y) Effects: On(b,y) ^ Clear(x) ^ ¬On(b,x) ^ ¬Clear(y)

AI Principles, Lecture on Planning The STRIPS state and goal descriptions  In the state description anything that isn’t stated is assumed to be false (Closed World Assumption)  So only +ve literals are allowed in the state description  Only ground literals are allowed in the goals. So not On(Next_Alpha(x),C) or On(x,C)  The goal description specifies a set of states BA C On(A,Table)On(B,Table) On(C,A) Clear(C) Clear(B) B A C On(A,B) On(B,C) State Goal

AI Principles, Lecture on Planning Application of an action  Apply Move(C,A,B) BA C On(A,Table)On(B,Table) On(C,A) Clear(C) Clear(B) BA State State Action(Move(b,x,y))- Preconditions: On(b,x) ^ Clear(b) ^ Clear(y) Effects: On(b,y) ^ Clear(x) ^ ¬On(b,x) ^ ¬Clear(y) On(A,Table)On(B,Table) On(C,A) Clear(C) Clear(B) Action(Move(C,x,y))- Preconditions: On(C,x) ^ Clear(C) ^ Clear(y) Effects: On(C,y) ^ Clear(x) ^ ¬On(C,x) ^ ¬Clear(y) Action(Move(C,A,y))- Preconditions: On(C,A) ^ Clear(C) ^ Clear(y) Effects: On(C,y) ^ Clear(A) ^ ¬On(C,A) ^ ¬Clear(y) Action(Move(C,A,B))- Preconditions: On(C,A) ^ Clear(C) ^ Clear(B) Effects: On(C,B) ^ Clear(A) ^ ¬On(C,A) ^ ¬Clear(B) On(A,Table)On(B,Table) On(C,A) Clear(C) Clear(B) Action(Move(C,A,B))- Preconditions: On(C,A) ^ Clear(C) ^ Clear(B) Effects: On(C,B) ^ Clear(A) ^ ¬On(C,A) ^ ¬Clear(B) On(A,Table)On(B,Table) Clear(C) On(A,Table)On(B,Table) On(C,B)Clear(C) Clear(A) C

AI Principles, Lecture on Planning Forward Chaining Search On(A,Table) On(B,Table) On(C,A) Clear(C) Clear(B) On(A,Table) On(B,Table) On(C,Table) Clear(A) Clear(C) Clear(B) On(A,Table) On(B,Table) On(C,B) Clear(C) Clear(A) Move(C,A, Table) Move(C,A,B)

AI Principles, Lecture on Planning Forward Chaining Search  Because of the restrictions on STRIPS representation there can only be a finite number of states  You need a graph search algorithm (e.g. A*)  You need a decent heuristic to control search behaviour  Forward search is inefficient in the number of irrelevant actions there are. BA C FED

AI Principles, Lecture on Planning Backward Chaining Search On(B,C) On(A,B) Move(A,x,B) Action(Move(b,x,y))- Preconditions: On(b,x) ^ Clear(b) ^ Clear(y) Effects: On(b,y) ^ Clear(x) ^ ¬On(b,x) ^ ¬Clear(y) {x/Table} Initial State: On(A,Table) On(B,Table) On(C,A) Clear(C) Clear(B) C BA A C B

AI Principles, Lecture on Planning Backward Chaining Search On(B,C) On(A,B)On(A,Table) Clear(A) Clear(B) On(B,C) Move(A,x,B) Action(Move(b,x,y))- Preconditions: On(b,x) ^ Clear(b) ^ Clear(y) Effects: On(b,y) ^ Clear(x) ^ ¬On(b,x) ^ ¬Clear(y) {x/Table} Initial State: On(A,Table) On(B,Table) On(C,A) Clear(C) Clear(B) C BA B AC A C B

AI Principles, Lecture on Planning Backward Chaining Search On(B,C) On(A,B)On(A,Table) Clear(A) Clear(B) On(B,C) Move(A,x,B) Action(Move(b,x,y))- Preconditions: On(b,x) ^ Clear(b) ^ Clear(y) Effects: On(b,y) ^ Clear(x) ^ ¬On(b,x) ^ ¬Clear(y) {x/Table} Initial State: On(A,Table) On(B,Table) On(C,A) Clear(C) Clear(B) C BA Move(b,A,x) On(A,Table) Clear(B) On(B,A) Clear(C) B AC A C B {x/C, b/B} B AC

AI Principles, Lecture on Planning Backward Chaining Search On(B,C) On(A,B)On(A,Table) Clear(A) Clear(B) On(B,C) Move(A,x,B) Action(Move(b,x,y))- Preconditions: On(b,x) ^ Clear(b) ^ Clear(y) Effects: On(b,y) ^ Clear(x) ^ ¬On(b,x) ^ ¬Clear(y) {x/Table} Move(B,x,C) Initial State: On(A,Table) On(B,Table) On(C,A) Clear(C) Clear(B) C BA Move(b,A,x) On(A,Table) Clear(B) On(B,A) Clear(C) {x/Table} B AC A C B {x/C, b/B} B AC BAC On(A,Table) On(B,Table) Clear(B) Clear(C) Clear(A)

AI Principles, Lecture on Planning Backward Chaining Search  An operator is chosen that achieves one of the unachieved parts of the goal state  Variables in the action are bound to entities in the world  When an operator is applied Its positive effects are deleted to create the predecessor Each precondition is added to the list of sub-goals  A sub-goal that is not true in the initial state is then chosen to be achieved

AI Principles, Lecture on Planning Linear vs non-linear planning  In early planners it was assumed that sub-goals could be solved completely independently  And the plans produced for these sub-goals could then be sequenced  This approach is called Linear Planning (Sacerdoti, 1975), but it is incomplete (not guaranteed to find solutions when they exist) because it doesn’t interleave the solutions.  Non-interleaved planners can’t solve some problems (e.g. the Sussman Anomaly)  Non-linear planners attempt to solve this problem BA C C B A

AI Principles, Lecture on Planning Conclusion  Planning is an important area of AI. State of the art planners are now beyond human level performance.  Specialist representations of action effects have been used to overcome issues of representational and inferential inefficiency  The STRIPS representation has been influential, succeeded by ADL and the overarching framework of PDDL  State space planning still requires search, either forward or backward chaining  This is the ancient history of planning, later on came partial order planning ( ), satisfiability planners (1992 on), graph planning (1995 on), and then the resurgence of state space planning (1996 on), and also planning under uncertainty (decision theoretic planning) (1997 on).