Abstract:

This paper proposes and verifies a decoupling controller of chassis dynamics integrated control system for the multi wheel distributed drive all-wheel steering unmanned vehicle， in order to further improve the dynamic performance of flexible， fast and accurate unmanned platform through decoupling control. Firstly， a vehicle coupling dynamic model is established which can accurately reflect the longitudinal， lateral and yaw motion of the vehicle， and the input-output coupling characteristics of the dynamic system are quantitatively analyzed combined with the nonparametric statistical method. Then， based on the principle of Neural Network Inverse （NNI） system， the vehicle dynamic system with decoupling linearization and decoupling composite controller are constructed. The decoupling response of the system is tested and verified successfully， and the proposed controller is verified by experiments. The results show that the interference caused by the dynamic coupling relationship between the control sub-channels of each degree of freedom of the coupling dynamic system is effectively reduced， so that relatively independent control of each sub-channel can be realized， which lays an important platform and physical foundation for efficient， accurate and stable comprehensive dynamic performance in the trajectory tracking process of distributed drive unmanned vehicle.

Key words: distributed independent drive, wheeled unmanned vehicle, chassis integrated control, neural network inverse decoupling control