结合多层次碰撞检测与IKFast运动学的磨削平台

High Efficiency Grinding Platform Combining Multi-Level Collision Detection and IKFAST Kinematics

针对传统磨削系统中存在的低精度和低效率的痛点,提出了一种结合多层次碰撞检测与IKFast运动学的高效磨削轨迹规划算法,并成功运用于机器人磨削平台。首先,针对磨削机器人高维构型空间复杂、难解,这里采用基于采样的运动规划算法,通过随机采样及碰撞检测来有效解决高维空间的运动规划问题。其次,基于混合包围盒AABB和OBB建立碰撞检测模型,通过多层次遍历的检测方法提高规划算法的整体效率。最后,为减轻机器人磨削轨迹中奇异点的影响,建立了IKFast运动学模型,对机器人奇异位置的正确逆解进行仿真求解,结果表明,文中方法有效增加了磨削轨迹的稳定性。

Aiming at the pain points of low precision and low efficiency existing in traditional grinding system,an efficient grind-ing trajectory planning algorithm combining multi-level collision detection and IKFAST kinematics was proposed and success-fully applied to robot grinding platform.Firstly,in view of the complexity and difficulty of the high-dimensional configuration space of the grinding robot,the motion planning algorithm based on sampling was adopted in this paper to effectively solve the motion planning problem in the high-dimensional space through random sampling and collision detection.Secondly,the colli-sion detection model is established based on the hybrid bounding box AABB and OBB,and the multi-level traversal detection method is adopted to improve the overall efficiency of the planning algorithm.Finally,in order to deal with the influence of the singularity points in the grinding trajectory of the robot,the kinematics model of IKFAST was established in this paper,and a correct inverse solution to the singular position of the robot was obtained via inverse solution simulation verification,thus improv-ing the stability of the final grinding trajectory.