Analysis and robust control design of a steering system for autonomous vehicles

The paper proposes the analysis and control of a steering system which is linked to the requirements of autonomous vehicles. The steering system of an autonomous vehicle must guarantee reliable and highly efficient intervention even in extreme conditions. Therefore, the determination of the reachability domain of the steering system is important. Both simulations and frequency domain methods are applied to the calculation. The paper shows that the reachability domain significantly depends on both the steering angle and the steering speed, which is utilized in the steering control design. Two results are incorporated in the robust control design of the steering. First, the analysis results of the steering system are built into the robust ℋ_∞ control design through frequency-dependent weighting functions. Second, experimental vehicle tests have been performed to examine the steering style of the driver. This information is also incorporated in the steering control through preview information. Finally, high-fidelity simulation scenarios show that the proposed method has several advantages compared to the conventional steering control solution.