Area Allocation for Electric Vehicle Coverage Path Planning
Nikolaos Baras, Antonios Chatzisavvas, Dimitris Ziouzios, Ioannis Vanidis, Minas Dasygenis · 2023
Coverage path planning (CPP) plays a pivotal role in several application domains, such as agriculture, robotics, and surveillance. At its core, CPP aims to identify a route that ensures complete coverage of a specified area in the least amount of time. This study introduces a novel method for spatial allocation in CPP by leveraging affinity propagation clustering (APC). By employing APC, the proposed technique groups the target area into clusters based on shared attributes. Subsequently, a robot is designated to each of these clusters, and a unique route is charted for each to ensure comprehensive coverage of its assigned region. The primary objective of the proposed method is to enhance cluster distribution among robots, thus minimizing both communication overhead and path length. The efficacy of the approach is evaluated through simulation experiments, where it is benchmarked against other prevailing methods. The results indicate that the proposed methodology surpasses its counterparts and yields sub-areas of higher quality to the robots. Due to its effectiveness, the suggested approach may effectively handle a wide range of CPP area division problems in various environment designs and sizes.