Scholarlink

<p>The advancement of drone technology has made it increasingly common for drone swarm communication to be affected by natural and human factors. This study will use an improved A* algorithm for path search. Aiming at the problem of communication interference, a drone swarm anti-communication interference algorithm is designed, and the network structure and related functions of the reinforcement learning algorithm are improved. This study conducted simulation tests on the improved algorithm and compared the test results with other similar algorithms. When six drones flew together along the path planned by the research method, each drone could safely reach the target location along the shortest path. Under noisy conditions, research algorithms could actively shorten the relative distance with other individuals to eliminate the influence of noise. The shortest planned path and flight time in the discrete plot were 489.23&#xa0;m and 16.65&#xa0;s, which are 17.30 and 32.45% shorter than the suboptimal Artificial Potential Field (APF); In the maze map, they are 786.47&#xa0;m and 88.06&#xa0;s, (shorter by 18.35 and 16.92% compared to APF). When the number of drones was 18, the average number of collisions and incidents beyond communication range was 112.71 and 253.97 (APF is 136.84 and 337.46). The research method can plan the shortest path of unmanned aerial vehicles in communication interference environments and enable them to safely reach the target location in a short period, which is of great significance for enhancing the application value of unmanned aerial vehicles in military reconnaissance, disaster rescue and other fields.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Path search planning for unmanned aerial vehicle clusters in communication interference environments

  • 期刊论文

  • 发表日期：

    2026年4月30日

查看全文
  • Haoyang Li,
  • Xiangke Wang

摘要

The advancement of drone technology has made it increasingly common for drone swarm communication to be affected by natural and human factors. This study will use an improved A* algorithm for path search. Aiming at the problem of communication interference, a drone swarm anti-communication interference algorithm is designed, and the network structure and related functions of the reinforcement learning algorithm are improved. This study conducted simulation tests on the improved algorithm and compared the test results with other similar algorithms. When six drones flew together along the path planned by the research method, each drone could safely reach the target location along the shortest path. Under noisy conditions, research algorithms could actively shorten the relative distance with other individuals to eliminate the influence of noise. The shortest planned path and flight time in the discrete plot were 489.23 m and 16.65 s, which are 17.30 and 32.45% shorter than the suboptimal Artificial Potential Field (APF); In the maze map, they are 786.47 m and 88.06 s, (shorter by 18.35 and 16.92% compared to APF). When the number of drones was 18, the average number of collisions and incidents beyond communication range was 112.71 and 253.97 (APF is 136.84 and 337.46). The research method can plan the shortest path of unmanned aerial vehicles in communication interference environments and enable them to safely reach the target location in a short period, which is of great significance for enhancing the application value of unmanned aerial vehicles in military reconnaissance, disaster rescue and other fields.