目的 为了实现异形热电池组的智能化装配和解决装配过程中存在装配精度不高的问题，设计了一种装配机器人，并仿真验证该机器人是否能够满足异形热电池组的装配精度需求。方法 采用改进D-H参数法构建运动学模型，通过正运动学分析机器人末端轨迹。采用改进DE算法进行逆运动学求解，并通过仿真实验分析运动学求解结果。结果 求得平均位置误差为1.4×10^–4 mm，平均姿态误差为2.4×10^–4度，平均解迭代次数为219.857，各构件最大角速度不应超过3.5 rad/s。结论 设计的装配机器人能够满足异形热电池组装配精度要求，运动学求解精度提高2个数量级，效率提高72.5%。

In order to achieve the intelligent assembly of special-shaped thermal battery packs and solve the problem of low assembly accuracy in the assembly process, an assembly robot was designed and verified whether the robot can meet the assembly accuracy requirements of special-shaped thermal battery packs. The improved D-H parameter method was used to construct a kinematic model, and the end trajectory of the robot was analyzed through positive kinematics. The improved DE algorithm is used to solve the inverse kinematics, and the kinematics solution results are analyzed through simulation experiments. The average position error is 1.4×10–4 mm, the average attitude error is 2.4×10–4 degrees, the average number of solution iterations is 219.857, and the maximum angular velocity of each component should not exceed 3.5 rad/s. The designed assembly robot can meet the assembly requirements of special-shaped thermal battery packs, the kinematics solution accuracy is improved by two orders of magnitude, and the efficiency is improved by 72.5%.