Traditional robot visual servoing control system to realize the calculation of the Jacobian matrix and the solution of the inverse Jacobian matrix, a large amount of calculation, the complex structure of the system, and realize the difficult calculation. In this paper, a six joint robot visual servoing system based on genetic neural network is designed, and genetic algorithm is used to optimize the neural network. This method not only solves the system of solving the Jacobian matrix and its inverse computation problem, and does not require the camera internal parameters and robot parameters are determined at the same time due to the addition of the genetic algorithm, to improve the performance of neural network, not only greatly simplifies the control system, improve the system speed, but also to ensure the control system precision.