Planungsbasierte Kommunikationsmodelle

Size: px

Start display at page:

Download "Planungsbasierte Kommunikationsmodelle"

Nickolas Allen
6 years ago
Views:

1 Planungsbasierte Kommunikationsmodelle Einführung in die Planung Konstantina Garoufi 19. Oktober 2011

2 Some Dictionary Definitions of Plan plan n. 1. A scheme, program, or method worked out beforehand for the accomplishment of an objective: a plan of attack. 2. A proposed or tentative project or course of action: had no plans for the evening. 3. A systematic arrangement of elements or important parts; a configuration or outline: a seating plan; the plan of a story. 4. A drawing or diagram made to scale showing the structure or arrangement of something. 5. A program or policy stipulating a service or benefit: a pension plan. 2

3 plan n. 1. A scheme, program, or method worked out beforehand for the accomplishment of an objective: a plan of attack. 3

4 plan n. 2. A proposed or tentative project or course of action: had no plans for the evening. 4

5 plan n. 3. A systematic arrangement of elements or important parts; a configuration or outline: a seating plan; the plan of a story. 5

6 plan n. 4. A drawing or diagram made to scale showing the structure or arrangement of something. 6

7 plan n. 5. A program or policy stipulating a service or benefit: a pension plan. 7

8 Some Dictionary Definitions of Plan plan n. 1. A scheme, program, or method worked out beforehand for the accomplishment of an objective: a plan of attack. 2. A proposed or tentative project or course of action: had no plans for the evening. These two are closest to the meaning used in AI 3. A systematic arrangement of elements or important parts; a configuration or outline: a seating plan; the plan of a story. 4. A drawing or diagram made to scale showing the structure or arrangement of something. 5. A program or policy stipulating a service or benefit: a pension plan. 8

9 02 Clamp board 03 Establish datum point at bullseye (0.25, 1.00) 004 B VMC Install 0.15-diameter side-milling tool [a representation] of future behavior usually a set of actions, with temporal and other constraints on them, for execution by some agent or agents. - Austin Tate 02 Rough side-mill pocket at (-0.25, 1.25) length 0.40, width 0.30, depth Finish side-mill pocket at (-0.25, 1.25) length 0.40, width 0.30, depth Rough side-mill pocket at (-0.25, 3.00) length 0.40, width 0.30, depth Finish side-mill pocket at (-0.25, 3.00) length 0.40, width 0.30, depth C VMC Install 0.08-diameter end-milling tool [MIT Encyclopedia of the [...] 004 T VMC Total time on VMC1 Cognitive Sciences, 1999] 005 A EC Pre-clean board (scrub and wash) 02 Dry board in oven at 85 deg. F 005 B EC Setup 02 Spread photoresist from RPM spinner 005 C EC Setup 005 D EC Setup 02 Photolithography of photoresist using phototool in "real.iges" 02 Etching of copper 005 T EC Total time on EC1 006 A MC Setup 02 Prepare board for soldering 006 B MC Setup A portion of a manufacturing process plan 02 Screenprint solder stop on board 9

Generating Plans of Action Computer programs to aid human planners Project management (consumer software) Plan storage and retrieval»e.g., variant process planning in manufacturing Automatic schedule generation»various OR and AI techniques Processes: Opn A BC/WW Setup Runtime LN Description 001 A VMC1 2.

10 Generating Plans of Action Computer programs to aid human planners Project management (consumer software) Plan storage and retrieval»e.g., variant process planning in manufacturing Automatic schedule generation»various OR and AI techniques Processes: Opn A BC/WW Setup Runtime LN Description 001 A VMC Orient board 02 Clamp board 03 Establish datum point at bullseye (0.25, 1.00) 001 B VMC Install 0.30-diameter drill bit 02 Rough drill at (1.25, -0.50) to depth Finish drill at (1.25, -0.50) to depth C VMC Install 0.20-diameter drill bit 001 T VMC Total time on VMC1 [...] 004 A VMC Orient board 02 Rough drill at (0.00, 4.88) to depth Finish drill at (0.00, 4.88) to depth 1.00 [...] 02 Clamp board 03 Establish datum point at bullseye (0.25, 1.00) 004 B VMC Install 0.15-diameter side-milling tool 02 Rough side-mill pocket at (-0.25, 1.25) length 0.40, width 0.30, depth Finish side-mill pocket at (-0.25, 1.25) length 0.40, width 0.30, depth Rough side-mill pocket at (-0.25, 3.00) length 0.40, width 0.30, depth Finish side-mill pocket at (-0.25, 3.00) length 0.40, width 0.30, depth C VMC Install 0.08-diameter end-milling tool [...] 004 T VMC Total time on VMC1 005 A EC Pre-clean board (scrub and wash) 02 Dry board in oven at 85 deg. F 005 B EC Setup 02 Spread photoresist from RPM spinner For some problems, we would like to generate plans (or pieces of plans) automatically Much more difficult Automated-planning research is starting to pay off Here are some examples 005 C EC Setup 005 D EC Setup 02 Photolithography of photoresist using phototool in "real.iges" 02 Etching of copper 005 T EC Total time on EC1 006 A MC Setup 02 Prepare board for soldering 006 B MC Setup 02 Screenprint solder stop on board 006 C MC Setup 02 Deposit solder paste at (3.35,1.23) on board using nozzle [...] 31 Deposit solder paste at (3.52,4.00) on board using nozzle 006 D MC Dry board in oven at 85 deg. F to solidify solder paste 006 T MC Total time on MC1 [...] 011 A TC Perform post-cap testing on board 011 B TC Perform final inspection of board 011 T TC Total time on TC1 999 T Total time to manufacture 10

11 Space Exploration Autonomous planning, scheduling, control NASA: JPL and Ames Remote Agent Experiment (RAX) Deep Space 1 Mars Exploration Rover (MER) 11

12 Manufacturing Sheet-metal bending machines - Amada Corporation Software to plan the sequence of bends [Gupta and Bourne, J. Manufacturing Sci. and Engr., 1999] 12

13 Bridge Baron - Great Game Products 1997 world champion of computer bridge [Smith, Nau, and Throop, AI Magazine, 1998] 2004: 2nd place LeadLow(P 1 ; S) Finesse(P 1 ; S) Games FinesseTwo(P 2 ; S) Us:East declarer, West dummy Opponents:defenders, South & North Contract:East 3NT On lead:west at trick 3 East: KJ74 West: A2 Out: QT98653 PlayCard(P 1 ; S, R 1 ) West 2 EasyFinesse(P 2 ; S) StandardFinesse(P 2 ; S) BustedFinesse(P 2 ; S) (North Q) (North 3) StandardFinesseTwo(P 2 ; S) StandardFinesseThree(P 3 ; S) FinesseFour(P 4 ; S) PlayCard(P 2 ; S, R 2 ) PlayCard(P 3 ; S, R 3 ) PlayCard(P 4 ; S, R 4 ) PlayCard(P 4 ; S, R 4 ) North 3 East J South 5 South Q 13

14 Outline What is planning? Conceptual model Classical planning Basic notions

15 Conceptual Model 1. Environment System Σ State transition system Σ = (S,A,E,γ) S = {states} A = {actions} E = {exogenous events} γ = state-transition function 15

16 State Transition Σ = (S,A,E,γ) System S = {states} A = {actions} E = {exogenous events} State-transition function γ: S x (A E) 2 S S = {s 0,, s 5 } A = {move1, move2, put, take, load, unload} E = {} γ: see the arrows s 1 put take location 1 location 2 location 1 location 2 move2 move1 move2 move1 s 3 put take location 1 location 2 location 1 location 2 unload load s 4 move2 move1 location 1 location 2 location 1 location 2 The Dock Worker Robots (DWR) domain s 0 s 2 s 5 16

17 Conceptual Model 2. Controller Observation function h: S O Controller Given observation o in O, produces action a in A s 3 location 1 location 2 17

18 Conceptual Model 3. Planner s Input Planning problem Planner 18

19 s 1 s 0 Planning put Problem take Description of Σ Initial state or set of states Initial state = s 0 Objective Goal state, set of goal states, set of tasks, trajectory of states, objective function, Goal state = s 5 location 1 location 2 move2 move1 location 1 location 2 unload load s 3 s 4 put take move2 location 1 location 2 move2 move1 location 1 location 2 s 2 s 5 move1 location 1 location 2 location 1 location 2 The Dock Worker Robots (DWR) domain 19

20 Conceptual Model 4. Planner s Output Planner Instructions to the controller 20

21 s 1 s 0 Plans put Classical plan: a sequence of actions location 1 location 2 move2 move1 take take location 1 location 2 move2 move1 take, move1, load, move2 s 3 put s 2 Policy: partial function from S into A location 1 location 2 take location 1 location 2 unload load {(s 0, take), (s 1, move1), (s 3, load), (s 4, move2)} location 1 location 2 s 4 move2 move1 location 1 location 2 s 5 The Dock Worker Robots (DWR) domain 21

22 Outline What is planning? Conceptual model Classical planning Basic notions

23 Domains In practice, not feasible to develop domain-independent planners that work in every possible domain Make simplifying assumptions to restrict the set of domains Classical planning Historical focus of most automated-planning research 23

24 A0: Finite system: finitely many states, actions, events A1: Fully observable: the controller always knows Σ s current state A2: Deterministic: each action has only one outcome A3: Static (no exogenous events): no changes but the controller s actions A4: Attainment goals: a set of goal states S g A5: Sequential plans: a plan is a linearly ordered sequence of actions (a 1, a 2, a n ) Restrictive Assumptions A6: Implicit time: no time durations; linear sequence of instantaneous states A7: Off-line planning: planner doesn t know the execution status 24

25 Classical planning requires all eight restrictive assumptions Offline generation of action sequences for a deterministic, static, finite system, with complete knowledge, attainment goals, and implicit time Reduces to the following problem: Given (Σ, s 0, S g ) Classical Planning Find a sequence of actions (a 1, a 2, a n ) that produces a sequence of state transitions (s 1, s 2,, s n ) such that s n is in S g. This is just path-searching in a graph Nodes = states Edges = actions Is this trivial? 25

26 Complexity Generalize the earlier example: Five locations, three robot carts, 100 containers, three piles»then there are states Number of particles in the universe is only about The example is more than times as large! location 1 location 2 move2 move1 s 1 put take Automated-planning research has been heavily dominated by classical planning Dozens (hundreds?) of different algorithms 26

27 Outline What is planning? Conceptual model Classical planning Basic notions

28 A running example: Dock Worker Robots Generalization of the earlier example A harbor with several locations»e.g., docks, docked ships, storage areas, parking areas Containers»going to/from ships Robot carts»can move containers Cranes»can load and unload containers 28

29 A running example: Dock Worker Robots Locations: l1, l2, Containers: c1, c2, can be stacked in piles, loaded onto robots, or held by cranes Piles: p1, p2, fixed areas where containers are stacked pallet at the bottom of each pile Robot carts: r1, r2, can move to adjacent locations carry at most one container Cranes: k1, k2, each belongs to a single location move containers between piles and robots if there is a pile at a location, there must also be a crane there 29

30 A running example: Dock Worker Robots Fixed relations: same in all states adjacent(l,l ) attached(p,l) belong(k,l) Dynamic relations: differ from one state to another occupied(l) at(r,l) loaded(r,c) unloaded(r) holding(k,c) empty(k) in(c,p) on(c,c ) top(c,p) top(pallet,p) Actions: take(c,k,p) put(c,k,p) load(r,c,k) unload(r) move(r,l,l ) 30

31 Representations: Motivation In most problems, far too many states to try to represent all of them explicitly as s 0, s 1, s 2, Represent each state as a set of features e.g.,»a vector of values for a set of variables»a set of ground atoms in some first-order language L Define a set of operators that can be used to compute statetransitions Don t give all of the states explicitly Just give the initial state Use the operators to generate the other states as needed 31

32 Classical Representation Start with a first-order language»language of first-order logic Restrict it to be function-free»finitely many predicate symbols and constant symbols, but no function symbols Example: the DWR domain Locations: l1, l2, Containers: c1, c2, Piles: p1, p2, Robot carts: r1, r2, Cranes: k1, k2, 32

33 Atom: predicate symbol and args Classical Representation Use these to represent both fixed and dynamic relations adjacent(l,l ) attached(p,l) belong(k,l) occupied(l) loaded(r,c) holding(k,c) in(c,p) top(c,p) at(r,l) unloaded(r) empty(k) on(c,c ) top(pallet,p) Ground expression: contains no variable symbols - e.g., in(c1,p3) Unground expression: at least one variable symbol - e.g., in(c1,x) Substitution: θ = {x 1 v 1, x 2 v 2,, x n v n } Each x i is a variable symbol; each v i is a term Instance of e: result of applying a substitution θ to e Replace variables of e simultaneously, not sequentially 33

34 States State: a set s of ground atoms The atoms represent the things that are true in one of Σ s states Only finitely many ground atoms, so only finitely many possible states 34

35 Operators Operator: a triple o=(name(o), precond(o), effects(o)) precond(o): preconditions» literals that must be true in order to use the operator effects(o): effects» literals the operator will make true name(o): a syntactic expression of the form n(x 1,,x k )» n is an operator symbol - must be unique for each operator» (x 1,,x k ) is a list of every variable symbol (parameter) that appears in o Purpose of name(o) is so we can refer unambiguously to instances of o Rather than writing each operator as a triple, we ll usually write it in the following format: 35

36 Planning domain: language plus operators Corresponds to a set of state-transition systems Example: operators for the DWR domain 36

37 Actions take(crane1,loc1,c3,c1,p1) precond: belong(crane,loc1), attached(p1,loc1), empty(crane1), top(c3,p1), on(c3,c1) effects: holding(crane1,c3), empty(crane1), in(c3,p1), top(c3,p1), on(c1,c1), top(c1,p1) An action is a ground instance (via substitution) of an operator Note that an action s name identifies it unambiguously take(crane1,loc1,c3,c1,p1) 37

38 Notation Let S be a set of literals. Then S + = {atoms that appear positively in S} S = {atoms that appear negatively in S} Let a be an operator or action. Then precond + (a) = {atoms that appear positively in a s preconditions} precond (a) = {atoms that appear negatively in a s preconditions} effects + (a) = {atoms that appear positively in a s effects} effects (a) = {atoms that appear negatively in a s effects} effects + (take(k,l,c,d,p)) = {holding(k,c), top(d,p)} effects (take(k,l,c,d,p)) = {empty(k), in(c,p), top(c,p), on(c,d)} 38

39 Applicability An action a is applicable to a state s if s satisfies precond(a), i.e., if precond + (a) s and precond (a) s = An action: A state it s applicable to take(crane1,loc1,c3,c1,p1) precond: effects: belong(crane,loc1), attached(p1,loc1), empty(crane1), top(c3,p1), on(c3,c1) holding(crane1,c3), empty(crane1), in(c3,p1), top(c3,p1), on(c1,c1), top(c1,p1) s 1 = {attached(p1,loc1), in(c1,p1), in (c3,p1), top(c3,p1), on(c3,c1), on (c1,pallet), attached(p2,loc1), in(c2,p2), top(c2,p2), on(c2,palet), belong (crane1,loc1), empty(crane1), adjacent(loc1,loc2), adjacent(loc2,loc1), at(r1,loc2), occupied(loc2, unloaded (r1)} 39

40 Executing an Applicable Action Remove a s negative effects, and add a s positive effects γ(s,a) = (s effects (a)) effects + (a) 40

41 Planning Problems Given a planning domain (language L, operators O) Statement of a planning problem: a triple P=(O,s 0,g)» O is the collection of operators» s 0 is a state (the initial state)» g is a set of literals (the goal formula) Planning problem: P = (Σ,s 0,S g )» s 0 = initial state»s g = set of goal states» Σ = (S,A,γ) is a state-transition system» S = {all sets of ground atoms in L}» A = {all ground instances of operators in O}» γ = the state-transition function determined by the operators We ll often say planning problem when we mean the statement of the problem 41

42 Plans and Solutions Plan: any sequence of actions σ = a 1, a 2,, a n such that each a i is an instance of an operator in O The plan is a solution for P=(O,s 0,g) if it is executable and achieves g i.e., if there are states s 0, s 1,, s n such that» γ (s 0,a 1 ) = s 1» γ (s 1,a 2 ) = s 2»» γ (s n 1,a n ) = s n»s n satisfies g 42

43 Example problem Let P 1 = (O, s 1, g 1 ), where g 1 ={loaded(r1,c3), at(r1,loc2)} O = {the four DWR operators given earlier} 43

44 Solution P 1 has infinitely many solutions Here are three of them: take(crane1,loc1,c3,c1,p1), move(r1,loc2,loc1), move(r1,loc1,loc2), move(r1,loc2,loc1), load(crane1,loc1,c3,r1), move(r1,loc1,loc2) take(crane1,loc1,c3,c1,p1), move(r1,loc2,loc1), load(crane1,loc1,c3,r1), move (r1,loc1,loc2) move(r1,loc2,loc1), take(crane1,loc1,c3,c1,p1), load(crane1,loc1,c3,r1), move (r1,loc1,loc2) They each produce this state: 44

45 What does this all have to do with linguistics? In this seminar, we will see how automated planning principles can be applied to the modeling of human communication... and how linguistic knowledge can be integrated with non-linguistic one under a single powerful formalism We are mostly concerned with language generation, but we ll also have a look into language interpretation models (which are based on plan recognition)

46 Course syllabus and slides id=planung

Artificial Intelligence Planning

Artificial Intelligence Planning Gholamreza Ghassem-Sani Sharif University of Technology Lecture slides are mainly based on Dana Nau s AI Planning University of Maryland 1 For technical details: Related