基于视觉的锂电池防爆片摆盘机器人轨迹规划

Trajectory Planning of Lithium Battery Explosion-Proof Pieces Handling Robot Based on Vision

摘要: 针对人工检测锂电池防爆片的正反面并组装效率低、成本高等问题，研发了一套基于机器视觉的锂电池防爆片自动摆盘系统。由定位相机采集目标的位置信息，通过Blob分析实现目标定位；由检测相机采集目标的特征信息，通过图像配准对目标位置进行修正，根据ROI视图选取区域的灰度均值来判断防爆片的正反。采用7次B样条曲线构造机械臂关节的运动轨迹，在满足机械臂约束的条件下，通过NSGA-II和MOPSO混合算法(NSGAII-MOPSO)以时间消耗少、能耗低和冲击小为优化目标对运动轨迹进行优化。实验结果表明该系统正反检测的准确率在98%以上，验证了检测算法的有效性；仿真结果显示，相较于NSGA-II和MOPSO算法，该优化方法的非支配解集分布性更好且更加接近真实Pareto前沿，最终获得的轨迹不仅满足生产线对机械臂运行时间的要求，还降低了机械臂的能耗和冲击，提高了防爆片抓取搬运的质量。

Abstract: Aiming at the problems of low efficiency and high cost of manual inspection on front and back of lithium battery explosion-proof pieces and then assembly, a set of visual location and detection system of explosion-proof pieces of lithium battery handling robot is developed. The location infor-mation of the target is collected by the positioning camera, and the target location is solved by Blob analysis. The detection camera collects the characteristic information of the target, corrects the target position through image registration, and judges the front and back of explosion-proof pieces according to the gray mean of the selected area in ROI. The motion trajectory of the joint of the ma-nipulator is constructed by seventh-degree B-spline curve. Under the condition of satisfying the constraints of the manipulator, the motion trajectory is optimized by NSGA-II and MOPSO hybrid algorithm (NSGA II-MOPSO) with less time consumption, low energy consumption and small impact as the optimization objectives. Experimental results show that front and back detection accuracy of the system is more than 98%, which verifies the effectiveness of the detection algorithm. The simu-lation results show that, compared with NSGA-II and MOPSO algorithms, the non- dominated set of this optimization method has better distribution and is closer to the real Pareto frontier. And the final trajectory not only meets the requirement of the running time of the manipulator in the pro-duction line, but also reduces the energy consumption and impact of the manipulator, and improves the quality of explosion-proof pieces grasping and handling.