Abstract: The research object is the vehicle-mounted cleaning mobile robot arm for cleaning the surface of the heliostat in the tower solar thermal power station,establish a kinematic model and analyze its inverse solution set. The genetic algorithm is used to solve the inverse solution of the on-board cleaning mobile manipulator,when solving multi-solution problems,use the principle of "moving more small joints and less large joints" to constrain,and find the optimal solution,it provides a basis for the trajectory planning of the robotic arm. Experiments show that the algorithm used in this article can obtain a feasible optimal solution,and minimize the movement of the first joint,can improve the control of the cleaning robot arm,and meet the actual cleaning requirements.