74 419 Artificial Intelligence 200506 Partially Ordered Plans

  • Slides: 16
Download presentation
74. 419 Artificial Intelligence 2005/06 Partially Ordered Plans - or: "How Do You Put

74. 419 Artificial Intelligence 2005/06 Partially Ordered Plans - or: "How Do You Put Your Shoes On? "

Partially Ordered Plans - or: "How Do You Put Your Shoes On? " Partially

Partially Ordered Plans - or: "How Do You Put Your Shoes On? " Partially Ordered Plans: • no strict sequence • partly parallel • observe threats

Least Commitment Strategy Partially Instantiated Plans Least Commitment Strategy In general, make as little

Least Commitment Strategy Partially Instantiated Plans Least Commitment Strategy In general, make as little concrete as possible, i. e. leave things undetermined until you have to determine them and become concrete. Partially Instantiated Plans During planning, variables have not necessarily to be instantiated immediately. Instantiation can wait, until binding becomes necessary

Partial Order Planning 1 Start with a rough plan and refine iteratively. First plan

Partial Order Planning 1 Start with a rough plan and refine iteratively. First plan consists only of start and finish actions: • start - T as precondition, initial world state as effect • finish - goal as precondition, NIL as effect Select actions to achieve sub-goals separately, quasi in parallel partial-order plan Fulfill open preconditions (sub-goals), until no more unsatisfied preconditions are left (last one is T of start)

Partial Order Planning 2 Add causal links to connect effects from actions to matching

Partial Order Planning 2 Add causal links to connect effects from actions to matching preconditions for plan. Causal links specify a partial order. Recognize threats - the effect of an action A negates the precondition of another action B. Add threats as partial order to plan: B<A (do B before A).

Partial Order Planning - Threats partial order plan = set of action strings (partial

Partial Order Planning - Threats partial order plan = set of action strings (partial plans) Problem: Detect and resolve threats, i. e. conflicts between actions – where the precondition of one action is deleted by another action – by choosing an adequate ordering of actions: if action b is a threat to action a, then a<b, i. e. a has to occur before b. (see also Russell/Norvig textbook, The POP Planner)

Partial Order Planning - Overall Use plan transformation operators to refine the partial plan

Partial Order Planning - Overall Use plan transformation operators to refine the partial plan and construct a complete plan: • add an action (operator), • reorder actions (operators), • instantiate actions (operators). A partial order plan consists of a set of action sequences (partial plans; action strings) which together achieve the complete set of goal literals. Threats induce an additional partial order of these action sequences.

Hierarchical Planning / Plan Decomposition Plans are organized in a hierarchy. Links between nodes

Hierarchical Planning / Plan Decomposition Plans are organized in a hierarchy. Links between nodes at different levels in the hierarchy denote a decomposition of a “complex action” into more primitive actions (operator expansion). Example: move (x, y, z) operator expansion pickup (x, y) putdown (x, z) The lowest level corresponds to executable actions of the agent.

Hierarchical Planning / Plan Decomposition • hierarchical organisation of 'actions' • complex and less

Hierarchical Planning / Plan Decomposition • hierarchical organisation of 'actions' • complex and less complex (or: abstract) actions • lowest level reflects directly executable actions • planning starts with complex action on top • plan constructed through action decomposition • substitute complex action with plan of less complex actions (pre-defined plan schemata; or learning of plans/plan abstraction, cf. ABSTRIPS) • overall plan must generate effect of complex action

Abstract Planning ABSTRIPS Consider different criticality values of preconditions in planning. Start with global,

Abstract Planning ABSTRIPS Consider different criticality values of preconditions in planning. Start with global, abstract plan. Then refine plan by trying to fulfill preconditions of abstract plan: • Choose preconditions with highest criticality values first ( = most difficult to achieve). • Then lower criticality value and continue with planning.

Other Issues in Planning Disjunctive Preconditions Conditional Effects n change is due to specific

Other Issues in Planning Disjunctive Preconditions Conditional Effects n change is due to specific condition n integrate into partial planning with threats Disjunctive Effects n parallel future worlds to consider All-Quantified Variables (in preconditions and effects) n only for finite, static Universe of objects

Real World Agents 1 n Consider Sensors and Effectors n n n perception of

Real World Agents 1 n Consider Sensors and Effectors n n n perception of environment, e. g. vision ensure correspondence between internal map of robot and environment, e. g. locating robot low-level body control, e. g. Motion Control (behaviour routines, e. g. Fuzzy or Neural Network Controller) other sensor information for body control and environment mapping, e. g. bumpers, radar sensors for other information channels and cognitive processes, e. g. speech – language

Real World Agents 2 n Low-level Processing and Control n n Motion Control Audio

Real World Agents 2 n Low-level Processing and Control n n Motion Control Audio Recording and low-level analysis Medium-level Processing n Navigation / Route Planning n Robot Location Higher-level Processing n n n Speech Recognition, NLP, . . . Strategies, Planning BDI (Belief-Desire-Intention) - Architecture

Real World Agents 3 n Multi-Agents n n Language / Communication → communicating agents

Real World Agents 3 n Multi-Agents n n Language / Communication → communicating agents mental Models of other Agents cooperating agents n Strategies cooperating agents n n Deontic Systems Trust

Additional References Nils J. Nilsson: Artificial Intelligence – A New Synthesis. Morgan Kaufmann, San

Additional References Nils J. Nilsson: Artificial Intelligence – A New Synthesis. Morgan Kaufmann, San Francisco, 1998. Konolidge, K. and K. Myers: The Saphira Architecture for Autonomous Mobile Robots (Robot Soccer Class Project) Guzzoni, D. et al. : Many Robots Make Short Work. (AAAI’ 96 Robot Competition - Meeting Scheduling) Martina Veloso, MIT (Robo. Cup)

Web Links Robo. Cup official web page – www. robocup. org Active Media Robotics

Web Links Robo. Cup official web page – www. robocup. org Active Media Robotics (pioneer, Saphira) – www. activmedia. com SONY’s Robo. Dog AIBO – www. aibo. com PBS Videos – Robots Alive, 04 -09 -97 (AAAI’ 96, Maze/Meeting Scheduling Robot Competition) Games Machines Play, 05 -21 -2002 (Robo. Cup, Seattle 2001)