Application of neural network in bicycle robot system identification

It is difficult to establish a more accurate dynamic model of bicycle robot which is a nonlinear, time-varying, ambiguity of system, uncertainty, etc, While precise model of complex system often requires more complex control design and calculation. As the neural network can approach any nonlinear function by any precision and possesses inherent characteristics of adaptive capacity. Based on NNARMAX2 model and NNOE model, the network structure identification of a typical nonlinear, unstable, and strong coupling bicycle robot system is established, which explains the relationship between handlebar angle and the inclination angle of bicycle during bicycle robot running stably. By comparing of the identified results, the simulation results show that NNOE model is effective for neural network to identify the nonlinear bicycle robot system.