多机器人体间动态碰撞检测方法仿真研究

Simulation Research on Dynamic Collision Detection Method Between Multiple Robots

当前多结合包围求碰撞检测法、Average-Case法、K-DOPs法等实现多机器人体间动态碰撞的检测,均存在寻优性能较差、检测效率较低的问题。为此提出一种基于动态粒子群的多机器人体间动态碰撞检测方法。采用OBB层次包围盒方法,缩小多机器人之间需要动态碰撞检测的区域,同时把动态碰撞检测问题转换为物体特征对间距离机制的非线性优化问题,进而构建层次拓扑框架进行局部碰撞检测,将机器人体引入到粒子群算法中建立混合进化算法,找到动态碰撞检测的最优解,实现多机器人体间动态碰撞检测。仿真结果证明,所提方法的检测效率高达96%,且具有较高的寻优性能。

At present, there are some problems such as poor optimization performance and low detection efficiency in multi-robot dynamic collision detection by Average-Case method and K-DOPs method. Therefore, a dynamic collision detection method between multiple robots based on dynamic particle swarm optimization(DPSO) was proposed. OBB hierarchical bounding volumes were used to reduce the area of dynamic collision detection between multiple robots. Meanwhile, the problem about dynamic collision detection was transformed into the problem about non-linear optimization for the mechanism of distance between object feature pairs, and then a hierarchical topological framework was constructed for local collision detection. In addition, a mixed evolutionary algorithm was established by introducing the robot body into particle swarm optimization(PSO) algorithm. Finally, the optimal solution of dynamic collision detection was found. Thus, the detection of dynamic collision between multi-robot bodies was achieved. Simulation results show that the detection efficiency of the proposed method is more than 96%, and it has high optimization performance.