Publication detail

Case-Based Reasoning Approaches to Robot Navigation

DVOŘÁK, J., HODÁL, J.

Original Title

Case-Based Reasoning Approaches to Robot Navigation

Type

conference paper

Language

English

Original Abstract

In this paper we deal with possibilities of using case-based reasoning methods for improvement of autonomous robot navigation in a partially known dynamic environment. We describe CBR approaches to both levels of navigation, global and local, but the main attention is focused on a path planning problem. The aim is to help with the search of paths between two points without collisions with known and unknown obstacles. At the same time, minimizing the length, difficulty and risk of paths is needed. The environment can be modeled by a two-dimensional grid or as a continuous space in which known obstacles and unattainable areas are defined. Verified paths are stored in a case base along with information about its properties. When proposing a path, first of all, the most similar already used paths are found. Then they are adapted to the new problem. When no sufficiently resembling paths are found, we use other methods for searching for the path (local search, genetic algorithms, etc.).

Key words in English

case-based reasoning, robot navigation, path planning

Authors

DVOŘÁK, J., HODÁL, J.

RIV year

2003

Released

1. 6. 2003

Publisher

Brno University of Technology, Faculty of Mechanical Engineering

Location

Brno

ISBN

80-214-2411-7

Book

Proceedings of the 9th International Conference on Soft Computing MENDEL 2003

Pages from

326

Pages to

333

Pages count

8

BibTex

@inproceedings{BUT14079,
  author="Jiří {Dvořák} and Jaroslav {Hodál}",
  title="Case-Based Reasoning Approaches to Robot Navigation",
  booktitle="Proceedings of the 9th International Conference on Soft Computing MENDEL 2003",
  year="2003",
  pages="8",
  publisher="Brno University of Technology, Faculty of Mechanical Engineering",
  address="Brno",
  isbn="80-214-2411-7"
}