Abstract: An adaptive critic structure including three neural networks was developed to so lve the two point boundary value problem of differential games．Two control neur al networks were used to optimize the controllers on two sides of the differenti al games,and a co-state neural network was used to approximate the co-state v ariables in Hamiltonian function．The output of co-state network was used to correct the output of the control networks,and the two convergent control netw o rks can be used as feedback controllers on two sides of the differential games s ystem repectively．The solution of differential games based on neural networks w a s compared with the one based on Chebyshev technique．The simulation results of the pursing-escaping differential games show that the neural network controller s are consistent with the optimal solution and present good robustness with resp ect to the initial conditions and measuring noises．