An improved sand cat swarm optimization for moving target search by UAV

Moving target search is a challenging dynamic path planning problem. In this scenario, unmanned aerial vehicles endeavor to locate a moving entity based on sensor information, utilizing the optimal path generated by a search algorithm. Based on the Bayesian principle, the task can be transformed into an optimization issue of the fitness function with the maximum probability of capturing the objective. In this study, an improved version of the sand cat swarm optimization algorithm, called the ISCSO search algorithm, is designed to tackle the moving target search issue effectively. Firstly, the presented ISCSO algorithm enhances the planning efficiency of the algorithm by encoding the unmanned aerial vehicle search path information into a set of motion paths through the motion-encoded mechanism. Secondly, the elite pooling strategy and the adaptive T-distribution are constructed to effectively improve the algorithm's ability to escape local optima and enhance its variability. Finally, ISCSO proposes a main architecture that seamlessly merges the search and attack methods of the sand cat swarm optimization algorithm, striking a balance between global and local search capabilities. To evaluate the superiority of the proposed algorithm, nine diverse search scenarios are constructed to verify its performance. The simulation results demonstrate that ISCSO achieves higher detection accuracy and offers more effective search paths for locating dynamic targets in comparison to other well-established metaheuristic algorithms. Code has been available at https://github.com/yb-niu1/ISCSO.