High-Level Planning in a Mobile Robot Domain (Preprint)

Abstract

An application of the SIPE planning system to high-level task planning for an autonomous indoor mobile robot is presented. The primary purpose was to evaluate the adequacy of SIPE for this domain, extending and improving the system in the process. The mobile robot domain as encoded in SIPE and the approach to interfacing the planner and the lower-level routines are described. The bulk of the paper presents both problems encountered during the process of encoding this domain, and extensions of the planning system that were made to solve them. The most significant addition was a redesign of the deductive capability of the planner, which is described in some detail. Efficiency considerations and the ability to intermingle planning and execution are discussed. The most important problem encountered involved hierarchical planning, an ambiguous term. We present a definition of it, and examine several of the reasons for this ambiguity. An explication of hierarchical-planning implementations entails two distinct notions: abstraction level and planning level A problem in currently implemented planners that is caused by mixing these two levels is presented and various remedies suggested. Three solutions that have been implemented in the current SIPE planning system are described.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jul 15, 1986
Accession Number
ADA461873

Entities

People

  • David E. Wilkins

Organizations

  • SRI International

Tags

Communities of Interest

  • Autonomy

DTIC Thesaurus Topics

  • Abstracts
  • Ambiguity
  • Availability
  • Classification
  • Coding
  • Contracts
  • Efficiency
  • Information Operations
  • Instructions
  • Monitoring
  • Security
  • Standards

Readers

  • Artificial Intelligence
  • Robotics and Automation.
  • Systems Analysis and Design

Technology Areas

  • AI & ML
  • AI & ML - Autonomous Systems
  • AI & ML - DoD AI Strategy
  • AI & ML - Machine Learning Algorithms
  • AI & ML - Neural Networks
  • Autonomy