基于多机器人自组织协作的多目标跟踪

提出了一种基于局部区域内的机器人和目标密度的多机器人分布式控制算法。运用这种算法,多机器人在跟踪多目标过程中,每个机器人能在每次调整速度和方向之前,根据最新的通讯信息和感知信息,计算通讯范围内相邻机器人和感知范围内移动目标的数量及位置坐标,然后按照所受人工势场(APF)力的合力决定下一次移动的新位置坐标,这个合力乘以一个基于机器人和目标密度的可变加权系数,以适应不同状态,然后按照动力学和运动学方程计算速度及方向。通过仿真实验对该算法进行了验证,实验结果表明,多机器人可以根据多目标的移动进行自主协作跟踪观测,算法的性能优于无可变加权系数的APF法和静态感知节点(SN)法。此方法对于实时监测和目标跟踪具有实际应用意义。

Self-organizing coordination of multi-robot for multi-target tracking

A distributed algorithm based on the local density of robots and targets for multi-target tracking was proposed. The algorithm improves the tracking by the following way. Based on the latest tracking information, each robot computes the numbers and coordinates of neighbor robots within its communicating range and targets within its sensing range, and then independently moves to the next position according to the sum of force resulting from other robots and targets. The force is multiplied to a weighted factor based on the local density of robots and targets. The performance of the algorithm was evaluated through simulation experiment. Experimental results indicate that robots are able to distribute themselves appropriately in response to the movement of targets. The algorithm performs better than the artificial potential field (APF) approach without the weighted factor and the static sensor nodes approach for multi-target tracking. This method can be applied to real-time monitoring and target tracking.