Estimation of Steering System Internal Friction: A Steering Load Observation-based Method

Steering system friction applies a significant effect on steering angle closed-loop control and road feeling improvement. However, most of the steering system control strategies presuppose the friction as a constant value, and ignore the friction variation in the duration of vehicle driving, which will degrade the control performance. To address the issue of friction variation, this paper proposes a steering load observation-based method to estimate steering system friction in real time. First, a detailed modeling of steering system is developed, and it can be found that steering system friction is highly dependent on steering resistance load by means of modeling analysis, which is the foundation to formulate a steering system friction model. Second, a composite observer, including a sliding mode observer (SMO) and a kingpin model, is designed to estimate steering resistance load in real time. Then, the accuracy of the established friction model is validated via test-bench experiments. The effectiveness of the designed SMO is verified by CARSIM - SIMULIK joint simulation experiment. Finally, the effectiveness of the proposed friction identification method is verified by test-bench experiment.