Fuzzy Genetic Motion Planning for an Autonomous Mobile Robot

摘要

Optimal motion planning is critical for the successful operation of an autonomous mobile robot. Many approaches have been proposed using either fuzzy logic or genetic algorithms (GAs). Most approaches offer only path planning or trajectory planning, but not both. The GA-based approach presented in this paper provides both path and trajectory planning, and has the advantage of allowing terrain conditions to be included in determining the optimal path. The terrain conditions are modeled using fuzzy linguistic variables to allow for the imprecision and uncertainty of the terrain data. Although a number of methods have been proposed using GAs, few are appropriate for a dynamic environment or provide response in real-time. The method proposed in this paper is robust, allowing the robot to adapt to dynamic conditions in the environment.