Mobile Robot Path Planning by Means of Case-Based Reasoning

Článek recenzovaný mimo WoS a Scopus

en

In this paper, we deal with mobile robot path planning in a partially known dynamic environment modelled by a two-dimensional grid in which known static obstacles and hazardous areas are defined. The aim of the path planning is searching for a path from a start to a goal position without collisions with known obstacles minimizing length and difficulty (risk) of the path. We propose path planning methods based on using a case graph, which is a structure composed of segments of already used paths. These methods try to construct a new path combining paths in the case graph with paths found by the A* algorithm. We present results of comparing proposed methods and methods based only on using Dijkstra's algorithm and the A* algorithm.

Mobile robot, path planning, case-based reasoning, graph searching algorithms

2005-09-20

1210-2717

12

A1

219–

8