论文信息 - A Replanning Algorithm for Decision Theoretic Planning: Empirical Evaluation

A Replanning Algorithm for Decision Theoretic Planning: Empirical Evaluation

In this paper, we present a replanning algorithm for a utility-based, hierarchical planner, and we compare it with replanning from scratch in a resource-bounded, utility driven agent architecture. We illustrate the experimental methodology we designed and implemented to investigate the role of the two strategies and we discuss the results obtained by executing experiments according to this methodology in different domains.

Guido Boella | Rossana Damiano | G. Boella | R. Damiano

[1] Félix Ingrand,et al. Interleaving Temporal Planning and Execution in Robotics Domains , 2004, AAAI.

[2] Diane J. Cook,et al. Incorporating decision-theoretic planning in a robot architecture , 2003, Robotics Auton. Syst..

[3] Guido Boella,et al. An Architecture for Normative Reactive Agents , 2002, PRIMA.

[4] Karen L. Myers. CPEF: A Continuous Planning and Execution Framework , 1999, AI Mag..

[5] Peter Haddawy,et al. Utility Models for Goal‐Directed, Decision‐Theoretic Planners , 1998, Comput. Intell..

[6] Michael Wooldridge,et al. Intention Reconsideration Reconsidered , 1998, ATAL.

[7] Michael Wooldridge,et al. The Belief-Desire-Intention Model of Agency , 1998, ATAL.

[8] Peter Haddawy,et al. Abstracting Probabilistic Actions , 1994, UAI.

[9] Peter Haddawy,et al. Decision-theoretic Refinement Planning Using Inheritance Abstraction , 1994, AIPS.

[10] Bernhard Nebel,et al. Plan Modification versus Plan Generation: A Complexity-Theoretic Perspective , 1993, IJCAI.

[11] Michael E. Bratman,et al. Intention, Plans, and Practical Reason , 1991 .