A novel robust model predictive controller for improving the steering stability of electric drive mobile platform considering the effect of both yaw motion and rototiller lifting

The instability problem of electric drive mobile platform (EDMP) is extremely prominent due to the narrow workspace and complex working environment in protected horticulture. In this paper, a distributed EDMP for protected horticulture is first designed to meet the requirements of flexible operation and high maneuverability stability. To improve the yaw stability of the EDMP during rototiller lifting, a novel lateral-coupling dynamic model between the EDMP and rototiller is proposed, and a polyhedral linear parameter varying discrete model considering the time-varying rotational inertia and tire lateral stiffness is also established. Afterwards, the correctness of the proposed model is verified through numerical simulation and experimental test, and the effect of rototiller lifting on the system model parameters is analyzed in-depth using the model. Finally, a novel robust model predictive controller based on linear matrix inequality is designed, where the effect of both yaw motion and rototiller lifting is considered. To demonstrate the validity of the designed controller, a steering test system of EDMP with suspended rototiller is established, and steering stability tests are also conducted under vary vehicle speeds and rototiller lifting angles. Results reveal that under the three lifting angle conditions of the rototiller, the average absolute errors of yaw rate in the experiment and simulation are 0.029, 0.023, and 0.025, respectively, and the average absolute errors of the sideslip angle are 0.0022, 0.0028, and 0.003, respectively, which verifies the correctness of the lateral coupling dynamics model. Results of the steering control test also indicate that the proposed controller reduces respectively the average absolute errors of yaw rate and sideslip angle by 46.4% and 38.2%, thereby improving greatly the steering stability of the EDMP under the combined influence of yaw motion and rototiller lifting.