目的 解决码垛机器人在抓取目标物时需要满足最大速度约束条件、较短运行时间以及运行平稳性等问题。方法 以Func200id六轴工业机器人为研究对象，通过机器人的各关节尺寸关系计算其DH参数，并建立仿真模型。对改进的HS算法通过增加变异因子VF来改变算法的搜索策略，使得算法中的每个个体不但有机会参与到迭代的过程之中，而且有概率向最优个体学习，加快算法的收敛速度。结果 通过对4种算法在6种不同测试函数下分别运行30次的对比实验可知，改进后的HS算法相较于其他3种算法的寻优能力更强，收敛速度更快。结论 改进后的HS算法能够明显缩短码垛机器人的工作时间，且运行连续平稳，在其他工业场景下也具有一定的应用价值。

The work aims to solve the problems of maximum speed constraint, short running time, and poor operation smoothness of palletizing robots when grasping target objects. With a Func200id six-axis industrial robot as the research object, its DH parameters were calculated by the relationship of each joint size of the robot, and a simulation model was established. The improved HS algorithm changed the search strategy of the algorithm by increasing the variation factor VF so that each individual in the algorithm not only had the opportunity to participate in the iterative process but also had the chance to learn from the optimal individual, which accelerated the convergence of the algorithm. Through a comparison experiment of the four algorithms running 30 times under six different test functions, the improved HS algorithm had stronger search capability and faster convergence than the remaining three algorithms. It can significantly shorten the working hours of the palletizing robot and run continuously and smoothly, which has certain application value in other industrial scenarios as well.