感知范围受限的群机器人自主围捕算法

针对在有障碍物场地中感知范围受限的群机器人协同围捕问题,本文首先给出了机器人个体、障碍物、目标的模型,并用数学形式对围捕任务进行描述,在此基础上提出了机器人个体基于简化虚拟速度和基于航向避障的自主围捕控制律.基于简化虚拟速度模型的控制律使得机器人能自主地围捕目标同时保持与同伴的距离避免互撞;基于航向的避障方法提升了个体的避障效率,避免斥力避障方法导致的死锁问题.其次本文证明了在该控制律下系统的稳定性.仿真结果表明,该算法在有效围捕目标的同时能够高效地避开障碍物,具有对复杂环境的适应性.最后本文分析了与其他方法相比该算法的优点.

Autonomous hunting algorithm for swarm robots subject to limited sensing range

In order to solve the problem of cooperative hunting by swarm robots subject to limited perception range in an environment with obstacles, this paper firstly gives the models of the robots, obstacles, and targets, and describes the task in mathematical form. On this basis, an autonomous hunting control law based on simplified virtual velocity and heading based obstacle avoidance of robots is proposed. The control law based on the simplified virtual velocity model allows the robots to autonomously hunt the target and keep the distance with companions to avoid collision; heading-based obstacle avoidance improves the efficiency of individual obstacle avoidance, and avoids the deadlock problem caused by the repulsive obstacle avoidance method. In addition, this paper proves the stability of the system under this control law. The simulations show that this algorithm can efficiently avoid obstacles while effectively rounding up targets, adapted to complex environments. Finally, this paper analyzes the advantages of this algorithm compared to other methods.