Through accuracy tracking control the trajectory of the robot,the accuracy of robot path planning and autonomous positioning can be improved. The motion track of the robot in the process of motion is influenced by small disturbance linear error,the robot system is a multi variable nonlinear control system,and in the traditional genetic algorithm for trajectory tracking control,the steady- state tracking error appearances in the boundary layer. A kind of algorithm of trajectory tracking control of robot based on improved genetic algorithm is put forward. The simulation of motion environment model of robot is constructed,the space coordinates of robot trajectory is regarded as the virtual world of abstract genetic populations,the grid structure model of the robot motion space is obtained. Through genetic evolution way to find the target point and move,and in order to make the robot trajectory control of genetic algorithm that satisfy matching conditions and parameters perturbation error,a integral term of robot trajectory tracking error in the synovial surface is designed,and the algorithm is improved. The simulation results show that,by using the algorithm,the optimal path of robot trajectory control can be found with fast convergence speed,and it has a better robot path tracking control accuracy and convergence performance,superior performance.