BP Neual Network Based on Improved BFGS Algorithm in the Virtual Speed Prediction of Bio-mimetic Robotic Horse

To improve the precision and efficiency for virtual speed prediction of bio-mimetic robotic horse, this paper presents the BP neural network based on improved BFGS to predict the movement of the virtual speed of the bio-mimetic robotic horse. Experiments show that the BP network proposed in this paper can effectively avoid the traditional BP neural networks detects which easy to fall into local minima, and has faster convergence speed.