This paper presents the complete integrated planning, executing and learning robotic agent Rogue. We describe Rogue's task planner that interleaves high-level task planning with real world robot execution. It supports multiple, asynchronous goals, suspends and interrupts tasks, and monitors and compensates for failure. We present a general approach for learning situation-dependent rules from ex...

We prove that determining if a valid proof plan exists is undecidable in general, even if the meta-theory in which pre-and post-conditions are checked is itself decidable. On a more positive note, we prove that an iterative deepening proof planner is sound (only nds valid proof plans) and complete (if it terminates signaling failure or fails to terminate then no valid proof plan exists). We als...

This paper presents the complete integrated planning, executing and learning robotic agent Rogue. We describe Rogue's task planner that interleaves high-level task planning with real world robot execution. It supports multiple, asynchronous goals, suspends and interrupts tasks, and monitors and compensates for failure. We present a general approach for learning situation-dependent rules from ex...

As any other problem solving task that employs search, AI Planning needs heuristics to efficiently guide the problem-space exploration. Machine learning (ML) provides several techniques for automatically acquiring those heuristics. Usually, a planner solves a problem, and a ML technique generates knowledge from the search episode in terms of complete plans (macro-operators or cases), or heurist...

Traditional approaches to the motion-planning problem can be classified into solutions for single-query and multiple-query problems with the tradeoffs on run-time computation cost and adaptability to environment changes. In this paper, we propose a novel approach to the problem that can learn incrementally on every planning query and effectively manage the learned road-map as the process goes o...

ION SWITCH Planning Done ? Resources Alloc ? NO YES NO YES NO YES PLANNER

We present a logic programming based conditional planner that is capable of generating both conditional plans and conformant plans in the presence of sensing actions and incomplete information. We prove the correctness of our implementation and show that our planner is complete with respect to the 0approximation of sensing actions and the class of conditional plans considered in this paper. Fin...

ley as a test bed for designing intelligent agents. The system consists of an overall agent architecture and five components within the architecture. The five components are (1) goaldirected learning (GDL), a decisiontheoretic method for selecting learning goals; (2) probabilistic bias evaluation (PBE), a technique for using probabilistic background knowledge to select learning biases for the l...