Contingency-Tolerant Robot Motion Planning and Control.

Abstract

This dissertation describes a new approach and enabling techniques to build the navigation system of a mobile robot operating in a partially known environment. The main layout of this environment is known in advance, but the locations and shapes of some smaller objects may not be known. Environments of this type include shop-floors, clean rooms, offices, etc. The problem is to automatically determine how the robot should move from one position to another without colliding with any of the objects in the environment. In order to be both efficient and robust, the navigation system should interweave a planning component and a reaction component. The planning component should take advantage of available prior knowledge to produce globally efficient plans. However, it should be aware that knowledge may be incomplete and generate lesser-committed plans that leave some freedom of choice at execution time to deal with contingencies. The reaction component should organize the robot's behavior according to both the ongoing plan and the sensory inputs.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 01, 1993
Accession Number
ADA326069

Entities

People

  • Wonyun Choi

Organizations

  • Stanford University

Tags

Communities of Interest

  • Autonomy
  • Sensors

DTIC Thesaurus Topics

  • Artificial Intelligence
  • Autonomous Navigation
  • Collision Avoidance
  • Computer Programming
  • Computer Science
  • Computers
  • Control Systems
  • Guidance
  • Lisp Programming Language
  • Motion Planning
  • Navigation
  • Navigational Equipment
  • Range Finding
  • Robot Navigation
  • Robots
  • Three Dimensional
  • Two Dimensional

Readers

  • Joint Military Operations and Doctrine.
  • Robotics and Automation.

Technology Areas

  • AI & ML
  • AI & ML - Autonomous Systems
  • Autonomy