Abstract: Because of the particularity of working environment and the complicated conditions of assignment，the mathematic model is built based on the kinematics problems and the dynamic problems at first.In addition，the PSO is used to optimize the PID parameters on line.So that，the deicing robot is conducted real time control，is implied notion control.At last，a test is simulated，which is aiming at the deicing robot crawling on slope.And the result is proved that the correction of the model and the proposed motion control.

Key words: deicing robot, Partical Swarm Optimization（PSO）, motion control, model building

摘要： 针对高压输电线除冰机器人工作环境的特殊性以及工作任务的复杂性，建立了机器人的运动学与动力学模型。在此基础上，通过采用粒子群算法对PID控制器三个参数进行在线寻优，实现了除冰机器人的运动控制。最后针对除冰机器人在斜坡爬行的情况进行仿真实验，证明了所设计的模型与运动控制方法的可行性。

关键词: 除冰机器人, 粒子群算法, 运动控制, 建模