Robust Static Output Feedback With Dynamic Disturbance Feed-Forward for Lateral Control of Long-Combination Vehicles at High Speeds

A control strategy is proposed to improve the high-speed lateral performance of an A double (tractor-semitrailer-dolly-semitrailer) combination vehicle using active steering of the dolly axles. The strategy is realized as an H1-type static output feedback (SOFB) based on a previous work, combined with dynamic disturbance feed-forward (DDFF). In order to synthesize DDFF in a way to facilitate performance improvement in practical problems, the disturbance is accurately characterized with the use of a dynamic weighting filter. As a contribution to the existing literature, novel linear matrix inequality (LMI) conditions are derived for this case, which facilitate the synthesis of a robust DDFF controller when the plant depends on uncertain time-varying parameters. An alternative DDFF synthesis method is also provided based on an adaptation of a previous work. The proposed overall controller has a simple structure and is easy to implement from a practical point of view since it requires only the driver steering angle (for feed-forward) and just one articulation angle measurement (for feedback). Obtained using a high-fidelity vehicle model, the simulation results for an example synthesis confirm a significant reduction in the rearward amplifications of the yaw rate and the lateral acceleration as well as the high-speed transient off-tracking during sudden lane change maneuvers.

Automated summary

This study proposes a control strategy to enhance the high-speed lateral performance of an A double combination vehicle by actively steering the dolly axles. The strategy combines an H1-type static output feedback (SOFB) based on previous work with dynamic disturbance feed-forward (DDFF). The accurate characterization of the disturbance is achieved through the use of a dynamic weighting filter. Novel linear matrix inequality (LMI) conditions are derived for the synthesis of a robust DDFF controller in cases where the plant depends on uncertain time-varying parameters. The simulation results utilizing a high-fidelity vehicle model demonstrate a significant reduction in rearward amplifications of yaw rate, lateral acceleration, and high-speed transient off-tracking during sudden lane change maneuvers.