Aiming at problem that parameters of six-rotor unmanned helicopter proportion integration differentiation（ PID） controller are difficult to be optimized

a PID neural network（ PIDNN） method which has characteristics of nonlinear mapping and self-learning

is developed and employed. It can adjust attitude control parameters dynamically and promote the self-adaptability of controller system. To prove the effectiveness of the method

the mathematical modeling for six-rotor is established by Simulink module in Matlab

PIDNN controller based on back propagation（ BP） algorithm is obtained by S-function. Simulation results are compared with control effects of PID

verifying that control effect of PIDNN is more valid in reducing attitude adjustment time and decreasing overshoots.