Comparative Study on Effects of Input Configurations of Linear Quadratic Controller on Path Tracking Performance under Low Friction Condition

Abstract

This paper presents a comparative study on the effects of the input configurations of linear quadratic (LQ) controllers on path tracking performance under low friction conditions. For the last decade, the path tracking controller has adopted several control inputs, input configurations, and actuators. However, these have not been compared with one another on a single frame in terms of common measures. For this reason, this paper compares input configurations of LQ controllers and available actuators in terms of common measures. For this purpose, the control inputs of the LQ controller were composed of front and rear steering and control yaw moment. By combining these control inputs, five input configurations of the LQ controller were set. If the control yaw moment is selected as a control input, then an actuator is needed to generate a control allocation, which should be adopted to convert the control yaw moment into longitudinal and lateral tire forces of actuators. As an actuator for control yaw moment generation, front/rear and 4-wheel steering, 4-wheel independent steering, braking, and driving were adopted. By applying the weighted least square based method, control allocation was formulated as a quadratic programming problem, which can be algebraically solved. For comparison on path tracking performance, new measures were adopted. To check the path tracking performance of each input configuration, a simulation was conducted on vehicle simulation software. From the simulation results, it was shown that front or 4-wheel steering itself is enough for path tracking on low friction roads and that the control yaw moment or an additional actuator is not recommended as a control input for path tracking on low friction roads.

Keywords

• autonomous vehicle
• path tracking control
• linear quadratic regulator
• input configuration
• low friction road