Multi-Strategy Enhanced Secret Bird Optimization Algorithm for Solving Obstacle Avoidance Path Planning for Mobile Robots

Abstract

Mobile robots play a pivotal role in advancing smart manufacturing technologies. However, existing Obstacle avoidance path Planning (OP) algorithms for mobile robots suffer from low stability and applicability. Therefore, this paper proposes an enhanced Secret Bird Optimization Algorithm (SBOA)-based OP algorithm for mobile robots to address these challenges, termed AGMSBOA. Firstly, an adaptive learning strategy is introduced, where individuals enhance the diversity of the algorithm’s population by summarizing relationships among candidates of varying quality, thereby strengthening the algorithm’s ability to locate globally optimal obstacle avoidance path regions. Secondly, a group learning strategy is incorporated by dividing the population into learning and teaching groups, enhancing the algorithm’s exploitation capabilities, improving the accuracy of obstacle avoidance path planning, and reducing actual runtime. Lastly, a multiple population evolution strategy is proposed, which balances the exploration/exploitation phases of the algorithm by analyzing the nature of different individuals, improving the algorithm’s ability to escape suboptimal obstacle avoidance path traps. Subsequently, AGMSBOA was used to solve the OP problem on five maps and two OP problems in real-world environments. The experiments illustrate that AGMSBOA achieves more than 5% performance improvement in path length and a 100–win rate in runtime metrics, as well as faster convergence and stability of the solution. Therefore, AGMSBOA proposed in this paper is an efficient, robust, and robust OP method for mobile robots.