Automated Planning: Theory & Practice (The Morgan Kaufmann Series in Artificial Intelligence)

Automated Planning: Theory & Practice (The Morgan Kaufmann Series in Artificial Intelligence)

Malik Ghallab

Language: English

Pages: 635

ISBN: 1558608567

Format: PDF / Kindle (mobi) / ePub


Automated planning technology now plays a significant role in a variety of demanding applications, ranging from controlling space vehicles and robots to playing the game of bridge. These real-world applications create new opportunities for synergy between theory and practice: observing what works well in practice leads to better theories of planning, and better theories lead to better performance of practical applications.

Automated Planning mirrors this dialogue by offering a comprehensive, up-to-date resource on both the theory and practice of automated planning. The book goes well beyond classical planning, to include temporal planning, resource scheduling, planning under uncertainty, and modern techniques for plan generation, such as task decomposition, propositional satisfiability, constraint satisfaction, and model checking.

The authors combine over 30 years experience in planning research and development to offer an invaluable text to researchers, professionals, and graduate students.

*Comprehensively explains paradigms for automated planning.
*Provides a thorough understanding of theory and planning practice, and how they relate to each other.
*Presents case studies of applications in space, robotics, CAD/CAM, process control, emergency operations, and games.

*Provides a thorough understanding of AI planning theory and practice, and how they relate to each other.
*Covers all the contemporary topics of planning, as well as important practical applications of planning, such as model checking and game playing.
*Presents case studies and applications in planning engineering, space, robotics, CAD/CAM, process control, emergency operations, and games.
*Provides lecture notes, examples of programming assignments, pointers to downloadable planning systems and related information online.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

would be difficult to keep track of what each state meant. In a set-theoretic representation, we can make each state’s meaning more obvious by using propositions to represent the statuses of the containers, the 2.2 Set-Theoretic Representation crane1 25 c8 c7 c6 c3 c2 c1 c5 c4 pile2 robot pile1 loc1 loc2 Figure 2.1 .A DWR state for Example 2.4. crane, and the robot. For example, the state shown in Figure 2.1 might be represented as: {nothing-on-c3, c3-on-c2, c2-on-c1, c1-on-pile1,

containing a solution plan that correctly achieves the required goals. Plan-space planning differs from state-space planning not only in its search space but also in its definition of a solution plan. Plan-space planning uses a more general plan structure than a sequence of actions. Here planning is considered as two separate operations: (1) the choice of actions, and (2) the ordering of the chosen actions so as to achieve the goal. A plan is defined as a set of planning operators together with

shown in this section. Let us first introduce an example. Example 6.1 Consider a simplified DWR domain with no piles and no cranes where robots can load and unload autonomously containers and where locations may contain an unlimited number of robots. In this domain, let us define a problem (see Figure 6.1) with two locations (loc1 and loc2), two containers (conta and contb), and two robots (robr and robq). Initially, robr and conta are in location loc1, robq and contb are in loc2. The goal is to

to the nogood table at a level i only if the call to GP-Search fails to establish g at this level from mutex and other nogoods found or established at the previous level. Extract(G, g , i) if i = 0 then return ( ) if g ∈ ∇(i) then return(failure) πi ← GP-Search(G, g , ∅, i) if πi = failure then return(πi ) ∇(i) ← ∇(i) ∪ {g } return(failure) end Figure 6.7 Extraction of a plan for a goal g. 128 Chapter 6 Planning-Graph Techniques GP-Search(G, g , πi , i) if g = ∅ then do ← Extract(G,

attempts, general purpose logic-based planning was done by deduction, e.g., by first-order resolution theorem proving (see the seminal work by Green in 1969 [249]). Deduction, however, could not compete with algorithms designed specifically for planning, such as search-based algorithms in the state or plan space. In the first AIPS Planning Competition in 1998, the BlackBox planning system [315] was competitive with the most efficient planners, including Graphplan and planners that do heuristic

Download sample

Download