Research on Path Planning for Industrial Cleaning Robots in Complex Environments: An Enhanced Immunogenetic Algorithm Approach

Abstract

As industrial automation advances, cleaning robots in complex environments have demonstrated significant value in enhancing efficiency and reducing labor costs. This paper presents a path planning method based on an improved Immune Genetic Algorithm (IGA) for comprehensive coverage path planning in complex industrial settings. Initially, the Boustrophedon Cellular Decomposition (BCD) method is employed to effectively segment the complex scene, identifying key areas for cleaning tasks. Subsequently, the path planning efficiency and adaptability are enhanced through an optimized genetic algorithm that integrates immune memory, Inver-Over operators, and 3-opt operators to refine the traversal sequence between regions. Moreover, this study incorporates the A* algorithm and the Biologically Inspired Neural Network (BINN) algorithm for efficient inter-regional path transitions and precise intra-regional traversal, respectively. A series of simulation experiments validate the superiority of the proposed algorithm in terms of path length, coverage rate, and computational time. The results demonstrate that this method not only effectively addresses the challenges of path planning in complex industrial environments but also significantly improves cleaning efficiency, offering high practical value and potential for broader application.