SlideShare ist ein Scribd-Unternehmen logo
1 von 4
Downloaden Sie, um offline zu lesen
Outline
Artificial Intelligence
Planning

•
•
•
•

Definition of AI planning
Key ideas: representation and search
STRIPS representation
partial-order planning and other techniques
for reducing complexity of plan search
• limiting assumptions of classical planning
• Recent trends on planning research

Blocks world example

A planning problem is specified by:
• an initial state
• a goal
• a model of the actions that can be performed
A solution to a planning problem is a sequence of
actions (a plan) that transforms the initial state
into a state that satisfies the goal description,
and possibly optimizes some measure of
performance.

B

A
initial state

goal state

Plan: Move C from A to table
Move B from table to C
Move A from table to B

Distinctions
• Planning is closely-related to problemsolving and search, but there are important
differences
• Planning is also related to, but different
from, scheduling, theorem-proving,
automatic programming, and work in
control theory, decision theory and
operations research, etc.

A
B
C

C

Applications
•
•
•
•
•

Robot navigation and control
Spacecraft control (Deep Space One)
factory and manufacturing operations
intelligent tutoring
natural language generation and
understanding (plan recognition)
• many others

1
Beginnings of planning research

Key issues: Representation and search
• A domain-independent planner consists of a
language for expressing planning problems and
an associated search algorithm for finding plans
• Examples:

Situation calculus (1959-)
QA3 (1969)
frame problem
Resolution theorem-proving (1965)
STRIPS (1972)

GPS (1969)

Epistemological and heuristic adequacy
The two key issues in planning research are:
– modeling actions and change
– organizing the search for plans
“There is a spectrum of more and more expressive languages
for representing the world, an agent’s goals, and possible actions.
The task of writing a planning algorithm is harder for more
expressive representation languages, and the speed of the resulting
algorithm decreases as well.” (Dan Weld)

– situation calculus and resolution theorem-proving
– STRIPS language and planning algorithm
– many others

STRIPS Language (1)
• States and goals described by conjunctions of
predicates applied to constant symbols
• Example:
– constants: blocks A, B, C, table
– predicates:
• (clear A) = “block A has nothing on it”
• (on A, B) = “block A is on block B”
• (on B, table) = “block B is on the table”

– state: (clear A) ∧ (on A B) ∧ (on B table)

STRIPS Language (2)
• Actions represented as follows:
– name:
– precondition list:
– delete list:
– add list:

(move x from y to z)
(on x y) ∧ (clear x) ∧ (clear z)
(on x y) ∧ (clear z)
(on x z)

• An action (operator) with variables is called
an action (operator) schema

Partial descriptions
• Only relevant aspects of the world need to be
included in state description
• STRIPS Assumption: Everything about the
world that is not mentioned by the action
schema remains unchanged by action – this is
“solution” to frame problem

2
Influence
• The STRIPS planning algorithm had many
deficiencies and was never used again.
Many improved planning algorithms have
been developed since.
• The STRIPS language, and variations of it,
continues to be used by most planners

Search in situation space

Searching for a plan
• Besides modeling actions and change, the
second important issue in planning is how
to make the search for a plan efficient
• Distinction between searching in situation
space and plan space

Situation space for the blocks world

• States represent situations
• Edges represent actions that transform a
situation into a new situation
• Search strategies:
– forward from initial state (progression) or
backward from goal (regression or backward
chaining)
– total-order planning = plan is build action-byaction in forward or backward direction

Search in plan space

Plan space for the blocks world

• First used by planner called NOAH [1975].
• States represent partially specified plans.
• Edges represent plan-refinement operations (e.g.
adding actions or ordering constraints).
• Search strategies
– least-commitment = planner avoids making decisions
until a good reason to make a choice
– partial-order planning = decide what actions to
perform before deciding ordering of actions

3
Example of nonlinear plan
(move A from table to B)

Finish

Start

(move B from table to C)

Example of partial-order planning

Advantages of partial-order planning
• By postponing choice about how to order actions,
can avoid search of exponential possible orderings
• It may take only a few steps to construct a plan that
would take thousands of steps using a standard
total-order approach
• The least-commitment approach means the planner
only needs to search in places where subplans
interact with each other

The null plan for the Sussman anomaly contains two actions:
*start* specifies the initial state and *end* specifies the goal.

• The following simple blocks-world planning
problem is called the “Sussman anomaly”

C
A

B

A
B
C

• The next few slides show some of the steps involved
in finding a plan using a partial-order planner

The plan after adding a causal link to support (on b c)
The agenda contains [(clear b) (clear c) (on b table) (on a b)]

The plan after adding a causal link to support (clear b)
The agenda is set to [(clear c) (on b table) (on a b)]

4

Weitere ähnliche Inhalte

Ähnlich wie Slides15

Goal stack planning.ppt
Goal stack planning.pptGoal stack planning.ppt
Goal stack planning.pptSadagopanS
 
21CSC206T_UNIT 5.pptx artificial intelligence
21CSC206T_UNIT 5.pptx artificial intelligence21CSC206T_UNIT 5.pptx artificial intelligence
21CSC206T_UNIT 5.pptx artificial intelligenceANTOARA2211003040050
 
CS1017-unitIV.pdf
CS1017-unitIV.pdfCS1017-unitIV.pdf
CS1017-unitIV.pdfSaswatSeth
 
Cs344 lect15-robotic-knowledge-inferencing-prolog-11feb08
Cs344 lect15-robotic-knowledge-inferencing-prolog-11feb08Cs344 lect15-robotic-knowledge-inferencing-prolog-11feb08
Cs344 lect15-robotic-knowledge-inferencing-prolog-11feb08Praveen Kumar
 
1.1 1.4-introduction
1.1 1.4-introduction1.1 1.4-introduction
1.1 1.4-introductionatulg0213
 
cs344-lect17-robotic-planning-25feb08.ppt
cs344-lect17-robotic-planning-25feb08.pptcs344-lect17-robotic-planning-25feb08.ppt
cs344-lect17-robotic-planning-25feb08.pptRaviShankarPratapDeo1
 
ProblemSolving(L-2).pdf
ProblemSolving(L-2).pdfProblemSolving(L-2).pdf
ProblemSolving(L-2).pdfAQSA SHAHID
 
C03.05-Scheduling.key.pdf
C03.05-Scheduling.key.pdfC03.05-Scheduling.key.pdf
C03.05-Scheduling.key.pdfssuser8babb7
 
NEAL-2016 ARL Symposium Poster
NEAL-2016 ARL Symposium PosterNEAL-2016 ARL Symposium Poster
NEAL-2016 ARL Symposium PosterBarbara Jean Neal
 
Dsoop (co 221) 1
Dsoop (co 221) 1Dsoop (co 221) 1
Dsoop (co 221) 1Puja Koch
 
Cs221 logic-planning
Cs221 logic-planningCs221 logic-planning
Cs221 logic-planningdarwinrlo
 

Ähnlich wie Slides15 (20)

Planning
PlanningPlanning
Planning
 
CH4_AI_Lecture.ppt
CH4_AI_Lecture.pptCH4_AI_Lecture.ppt
CH4_AI_Lecture.ppt
 
AI_Planning.pdf
AI_Planning.pdfAI_Planning.pdf
AI_Planning.pdf
 
Goal stack planning.ppt
Goal stack planning.pptGoal stack planning.ppt
Goal stack planning.ppt
 
21CSC206T_UNIT 5.pptx artificial intelligence
21CSC206T_UNIT 5.pptx artificial intelligence21CSC206T_UNIT 5.pptx artificial intelligence
21CSC206T_UNIT 5.pptx artificial intelligence
 
CS1017-unitIV.pdf
CS1017-unitIV.pdfCS1017-unitIV.pdf
CS1017-unitIV.pdf
 
Unit-IV.pptx
Unit-IV.pptxUnit-IV.pptx
Unit-IV.pptx
 
RPT_AI-06_A_Planning Intro.ppt
RPT_AI-06_A_Planning Intro.pptRPT_AI-06_A_Planning Intro.ppt
RPT_AI-06_A_Planning Intro.ppt
 
Cs344 lect15-robotic-knowledge-inferencing-prolog-11feb08
Cs344 lect15-robotic-knowledge-inferencing-prolog-11feb08Cs344 lect15-robotic-knowledge-inferencing-prolog-11feb08
Cs344 lect15-robotic-knowledge-inferencing-prolog-11feb08
 
1.1 1.4-introduction
1.1 1.4-introduction1.1 1.4-introduction
1.1 1.4-introduction
 
16355694.ppt
16355694.ppt16355694.ppt
16355694.ppt
 
cs344-lect17-robotic-planning-25feb08.ppt
cs344-lect17-robotic-planning-25feb08.pptcs344-lect17-robotic-planning-25feb08.ppt
cs344-lect17-robotic-planning-25feb08.ppt
 
ProblemSolving(L-2).pdf
ProblemSolving(L-2).pdfProblemSolving(L-2).pdf
ProblemSolving(L-2).pdf
 
AI_unit IV Full Notes.pdf
AI_unit IV Full Notes.pdfAI_unit IV Full Notes.pdf
AI_unit IV Full Notes.pdf
 
C03.05-Scheduling.key.pdf
C03.05-Scheduling.key.pdfC03.05-Scheduling.key.pdf
C03.05-Scheduling.key.pdf
 
NEAL-2016 ARL Symposium Poster
NEAL-2016 ARL Symposium PosterNEAL-2016 ARL Symposium Poster
NEAL-2016 ARL Symposium Poster
 
Dsoop (co 221) 1
Dsoop (co 221) 1Dsoop (co 221) 1
Dsoop (co 221) 1
 
Scheduling And Htn
Scheduling And HtnScheduling And Htn
Scheduling And Htn
 
Cs221 logic-planning
Cs221 logic-planningCs221 logic-planning
Cs221 logic-planning
 
l2.pptx
l2.pptxl2.pptx
l2.pptx
 

Mehr von Praveen Kumar (20)

Summer2014 internship
Summer2014 internshipSummer2014 internship
Summer2014 internship
 
Summer+training 2
Summer+training 2Summer+training 2
Summer+training 2
 
Summer+training
Summer+trainingSummer+training
Summer+training
 
Solutions1.1
Solutions1.1Solutions1.1
Solutions1.1
 
Scribed lec8
Scribed lec8Scribed lec8
Scribed lec8
 
Scholarship sc st
Scholarship sc stScholarship sc st
Scholarship sc st
 
Networks 2
Networks 2Networks 2
Networks 2
 
Mithfh lecturenotes 9
Mithfh lecturenotes 9Mithfh lecturenotes 9
Mithfh lecturenotes 9
 
Mcs student
Mcs studentMcs student
Mcs student
 
Math350 hw2solutions
Math350 hw2solutionsMath350 hw2solutions
Math350 hw2solutions
 
Matching
MatchingMatching
Matching
 
Line circle draw
Line circle drawLine circle draw
Line circle draw
 
Lecture3
Lecture3Lecture3
Lecture3
 
Lec2 state space
Lec2 state spaceLec2 state space
Lec2 state space
 
Graphtheory
GraphtheoryGraphtheory
Graphtheory
 
Graphics display-devicesmod-1
Graphics display-devicesmod-1Graphics display-devicesmod-1
Graphics display-devicesmod-1
 
Games.4
Games.4Games.4
Games.4
 
Dda line-algorithm
Dda line-algorithmDda line-algorithm
Dda line-algorithm
 
Cse3461.c.signal encoding.09 04-2012
Cse3461.c.signal encoding.09 04-2012Cse3461.c.signal encoding.09 04-2012
Cse3461.c.signal encoding.09 04-2012
 
Comerv3 1-12(2)
Comerv3 1-12(2)Comerv3 1-12(2)
Comerv3 1-12(2)
 

Kürzlich hochgeladen

My Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 PresentationMy Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 PresentationRidwan Fadjar
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking MenDelhi Call girls
 
The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024Rafal Los
 
Breaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path MountBreaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path MountPuma Security, LLC
 
Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024The Digital Insurer
 
GenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationGenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationMichael W. Hawkins
 
Maximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptxMaximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptxOnBoard
 
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...Miguel Araújo
 
Kalyanpur ) Call Girls in Lucknow Finest Escorts Service 🍸 8923113531 🎰 Avail...
Kalyanpur ) Call Girls in Lucknow Finest Escorts Service 🍸 8923113531 🎰 Avail...Kalyanpur ) Call Girls in Lucknow Finest Escorts Service 🍸 8923113531 🎰 Avail...
Kalyanpur ) Call Girls in Lucknow Finest Escorts Service 🍸 8923113531 🎰 Avail...gurkirankumar98700
 
Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slidevu2urc
 
08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking MenDelhi Call girls
 
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking MenDelhi Call girls
 
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...HostedbyConfluent
 
Google AI Hackathon: LLM based Evaluator for RAG
Google AI Hackathon: LLM based Evaluator for RAGGoogle AI Hackathon: LLM based Evaluator for RAG
Google AI Hackathon: LLM based Evaluator for RAGSujit Pal
 
SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024Scott Keck-Warren
 
Salesforce Community Group Quito, Salesforce 101
Salesforce Community Group Quito, Salesforce 101Salesforce Community Group Quito, Salesforce 101
Salesforce Community Group Quito, Salesforce 101Paola De la Torre
 
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 3652toLead Limited
 
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptxHampshireHUG
 
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)Gabriella Davis
 
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...Drew Madelung
 

Kürzlich hochgeladen (20)

My Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 PresentationMy Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 Presentation
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
 
The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024
 
Breaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path MountBreaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path Mount
 
Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024
 
GenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationGenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day Presentation
 
Maximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptxMaximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptx
 
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
 
Kalyanpur ) Call Girls in Lucknow Finest Escorts Service 🍸 8923113531 🎰 Avail...
Kalyanpur ) Call Girls in Lucknow Finest Escorts Service 🍸 8923113531 🎰 Avail...Kalyanpur ) Call Girls in Lucknow Finest Escorts Service 🍸 8923113531 🎰 Avail...
Kalyanpur ) Call Girls in Lucknow Finest Escorts Service 🍸 8923113531 🎰 Avail...
 
Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slide
 
08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men
 
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
 
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
 
Google AI Hackathon: LLM based Evaluator for RAG
Google AI Hackathon: LLM based Evaluator for RAGGoogle AI Hackathon: LLM based Evaluator for RAG
Google AI Hackathon: LLM based Evaluator for RAG
 
SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024
 
Salesforce Community Group Quito, Salesforce 101
Salesforce Community Group Quito, Salesforce 101Salesforce Community Group Quito, Salesforce 101
Salesforce Community Group Quito, Salesforce 101
 
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
 
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
 
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)
 
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
 

Slides15

  • 1. Outline Artificial Intelligence Planning • • • • Definition of AI planning Key ideas: representation and search STRIPS representation partial-order planning and other techniques for reducing complexity of plan search • limiting assumptions of classical planning • Recent trends on planning research Blocks world example A planning problem is specified by: • an initial state • a goal • a model of the actions that can be performed A solution to a planning problem is a sequence of actions (a plan) that transforms the initial state into a state that satisfies the goal description, and possibly optimizes some measure of performance. B A initial state goal state Plan: Move C from A to table Move B from table to C Move A from table to B Distinctions • Planning is closely-related to problemsolving and search, but there are important differences • Planning is also related to, but different from, scheduling, theorem-proving, automatic programming, and work in control theory, decision theory and operations research, etc. A B C C Applications • • • • • Robot navigation and control Spacecraft control (Deep Space One) factory and manufacturing operations intelligent tutoring natural language generation and understanding (plan recognition) • many others 1
  • 2. Beginnings of planning research Key issues: Representation and search • A domain-independent planner consists of a language for expressing planning problems and an associated search algorithm for finding plans • Examples: Situation calculus (1959-) QA3 (1969) frame problem Resolution theorem-proving (1965) STRIPS (1972) GPS (1969) Epistemological and heuristic adequacy The two key issues in planning research are: – modeling actions and change – organizing the search for plans “There is a spectrum of more and more expressive languages for representing the world, an agent’s goals, and possible actions. The task of writing a planning algorithm is harder for more expressive representation languages, and the speed of the resulting algorithm decreases as well.” (Dan Weld) – situation calculus and resolution theorem-proving – STRIPS language and planning algorithm – many others STRIPS Language (1) • States and goals described by conjunctions of predicates applied to constant symbols • Example: – constants: blocks A, B, C, table – predicates: • (clear A) = “block A has nothing on it” • (on A, B) = “block A is on block B” • (on B, table) = “block B is on the table” – state: (clear A) ∧ (on A B) ∧ (on B table) STRIPS Language (2) • Actions represented as follows: – name: – precondition list: – delete list: – add list: (move x from y to z) (on x y) ∧ (clear x) ∧ (clear z) (on x y) ∧ (clear z) (on x z) • An action (operator) with variables is called an action (operator) schema Partial descriptions • Only relevant aspects of the world need to be included in state description • STRIPS Assumption: Everything about the world that is not mentioned by the action schema remains unchanged by action – this is “solution” to frame problem 2
  • 3. Influence • The STRIPS planning algorithm had many deficiencies and was never used again. Many improved planning algorithms have been developed since. • The STRIPS language, and variations of it, continues to be used by most planners Search in situation space Searching for a plan • Besides modeling actions and change, the second important issue in planning is how to make the search for a plan efficient • Distinction between searching in situation space and plan space Situation space for the blocks world • States represent situations • Edges represent actions that transform a situation into a new situation • Search strategies: – forward from initial state (progression) or backward from goal (regression or backward chaining) – total-order planning = plan is build action-byaction in forward or backward direction Search in plan space Plan space for the blocks world • First used by planner called NOAH [1975]. • States represent partially specified plans. • Edges represent plan-refinement operations (e.g. adding actions or ordering constraints). • Search strategies – least-commitment = planner avoids making decisions until a good reason to make a choice – partial-order planning = decide what actions to perform before deciding ordering of actions 3
  • 4. Example of nonlinear plan (move A from table to B) Finish Start (move B from table to C) Example of partial-order planning Advantages of partial-order planning • By postponing choice about how to order actions, can avoid search of exponential possible orderings • It may take only a few steps to construct a plan that would take thousands of steps using a standard total-order approach • The least-commitment approach means the planner only needs to search in places where subplans interact with each other The null plan for the Sussman anomaly contains two actions: *start* specifies the initial state and *end* specifies the goal. • The following simple blocks-world planning problem is called the “Sussman anomaly” C A B A B C • The next few slides show some of the steps involved in finding a plan using a partial-order planner The plan after adding a causal link to support (on b c) The agenda contains [(clear b) (clear c) (on b table) (on a b)] The plan after adding a causal link to support (clear b) The agenda is set to [(clear c) (on b table) (on a b)] 4