Search

Summary: To address the difficulties of under‐actuation, nonlinearity, strong coupling, disturbances and measurement noise in four‐wheel steering (4WS) vehicle tracking control, this paper presents an innovative control strategy based on differential flatness and linear active disturbance rejection controller (LADRC). The flatness property of 4WS vehicle is derived from a simplified dynamic model that ignores some complicated dynamics and external disturbances. The impact of these neglected components on the system is treated as part of the total‐disturbance, which is estimated by the observer and compensated by LADRC. Three flatness‐based control schemes, including feedback linearization controller with PID corrective term, traditional LADRC and LADRC with robust generalized proportional integral observer (RGPIO), are designed and investigated in this paper. Moreover, an overtaking scenario is selected as an application example for the tracking controllers, and the corresponding trajectory planner is designed. The tracking performances of these schemes for the overtaking trajectory are simulated on a three‐degree‐of‐freedom dynamic model and a more complete dynamic model of the 4WS vehicle, respectively. Results show the effectiveness of these schemes, especially the excellent tracking performance of LADRC with RGPIO in the presence of strong external disturbances and measurement noise. [ABSTRACT FROM AUTHOR]

Further research on motion control of a 4WS4WD path-following vehicle is carried out in this paper. Focuses are placed on understanding and testing the vehicle path-following control models developed previously at a deep level. Control models are in relation to parameters introduced, and the effects of these parameters are discovered. Control models are interpreted by dynamic simulation using a 3DOF vehicle model with three cases. Three kinds of planned paths are considered in these cases to test control performances, which include the straight, circular, and sinusoidal paths. Interesting dynamic results are obtained and analyzed qualitatively, e.g., various steering modes. Simulation studies are extended with consideration of a fine vehicle dynamic model established in CarSim and a complex path composed of straight and curved segments. Control models are examined in a complex problem, and results obtained show that they are validated with robustness in dynamic environment. [ABSTRACT FROM AUTHOR]