Anti-Rollover Air Suspension System Using Height Adjustment Control

One of the most serious road accidents is vehicle rollover, which often results in severe injuries and fatalities, loss of properties, and significant traffic delays. Vehicle rollovers are often caused by high speed cornering, especially for trucks and vehicles with high center of gravity (CG), or in sudden maneuvering situations such as objective avoidance in autonomous driving, or trip over on uneven pavements. Understanding of the dynamics and controls of rollover is also critical for the development of Advanced Driver Assistance Systems (ADAS) for the next-generation vehicles. As a result, there has been significant efforts on the research and development of rollover resistance systems in the past decades. Five major anti-rollover methodologies have been proposed: four-wheel steering, differential braking, active roll-bar, in-wheel motor and semi-active/active suspension control. While active control of rollover demanding auxiliary energy supply can be costly and requires substantial power consumption, it has been shown that semi-active control of air suspension could be an economic and effective method to prevent rollover by employing a robust optimized PID controller to quickly recover the height of vehicle CG in the events of rollover instability caused by high speed cornering or sudden crosswind. Electronically controlled air suspension (ECAS) with height adjustment control has been widely applied in trucks and city buses to provide better ride comfort performance. Building on the configuration of ECAS, this thesis presents a novel rollover prevention method by controlling the heights of the left and right air springs individually with optimized PID controllers. A comprehensive nonlinear air spring mathematical model and a precise air flow model through the magnetic valve are established as the control plant. The PID controller can adjust the height of air spring to the target height by outputting a pulse-width modulation (PWM) signal based on the input height error variation. The simulation of this controller is implemented in MATLAB Simulink and tested for a two-degree-of-freedom quarter-car suspension model traversing on a random road profile. It is shown that the controller performs well on height tracking with minimal response oscillation. A second simulation is conducted on a rollover dynamics model with controlled air springs to examine the efficacy of this rollover prevention mechanism. To improve the response performance, the PID parameters are optimized. Results show that the optimized PID controller for height tracking is able to respond quickly enough to recover the roll angle of the vehicle and assist the vehicle to be stable