A distributed model for mobile robot environment-learning and navigation

Front Cover
MIT Artificial Intelligence Laboratory, 1990 - Artificial intelligence - 128 pages
0 Reviews
This thesis presents a method for robust mobile robot navigation, large space learning, and path planning, based on a totally distributed architecture. The described methods were implemented and tested on a physical robot. The robot, Toto, consists of an omnidirectional base supplied with a ring of twelve ultrasonic ranging sensors and a compass. It is fully autonomous with all power and processing onboard. All experimental data were gathered in unaltered office environments with static and dynamic obstacles. Toto is an example of incremental design methodology. The robot was programmed in the Behavior Language, based on the subsumption architecture. Its behavior consists of three real-time, reactive layers of competence: collision-free boundary tracing, landmark detection, and environment learning and path planning. Low-level navigation consists of a collection of simple reflex-like rules which, when acting in parallel, result in an emergency boundary-tracing behavior. This behavior is used by the landmark detector which dynamically extracts features from the environment using the way the robot is moving as it is moving. The landmarks are used to construct a distributed map of the environment. The map is represented as a graph of landmarks. The links in the graph are used to indicate topological adjacency, and are assigned dynamically. The structure of the environment is used to bound the outdegree of the graph nodes resulting in linear graph connectivity. (KR).

From inside the book

What people are saying - Write a review

We haven't found any reviews in the usual places.

Contents

Introduction
10
A Review of Related Work
18
Philosophy and Motivation
27

7 other sections not shown

Common terms and phrases

Bibliographic information