基于危险指数最小化的机器人实时安全避碰策略

为避免机器人在作业过程中撞到人,提出了一种能够主动提高机器人安全性的方法,即基于危险指数最小化的机器人安全避碰控制策略。该方法通过实时评估机器人运行过程中的危险指数,当危险指数达到最大安全允许值时,将在危险指数最小化的方向产生一个虚拟推动力使机器人迅速远离人,从而避免机器人与人发生碰撞。仿真结果表明,基于危险指数最小化的机器人实时安全避碰控制能够保证动态人机共处环境中人的安全性。

Real Time Robot Safe Anti collision Control Based on Minimized Danger Index

In order to prevent robot colliding to human during operation, a method of robot safe anti-collision control is presented based on minimized danger index, which can actively improve robot safety. Through real time evaluation of the danger index during robot motion, and once reaching the maximum allowable value of the danger index, a virtual driving force was produced by this method, which could push the robot rapidly away from human in the direction of the minimized danger index, and avoid robot collisions with human. The simulation result verifies that robot safe anti-collision control based on minimized danger index can guarantee human safety in dynamic environment of human-robot coexisting.