Performance Metrics and Evaluation of a Path Planner based on Genetic Algorithms

摘要

This paper focuses on the analysis of the performance of an innovative genetic path planner designed for a single agent exploration. The proposed method is a generalization of the well-known Traveling Salesman Problem (TSP) that we call Subtour problem and it can be formulated as finding the shortest possible path for visiting a subset of n given targets over a known area. The algorithm is based on a Genetic Algorithm coupled with a heuristic local search method. To evaluate the proposed planner, an extensive performance evaluation has been done.