Abstract: In order to eliminate the adverse effects of severe vibration on the ride comfort and stability of the vehicle, the Lagrange's equation was used to establish the 17 degree-of-freedom (DOF) dynamic equations of the road-vehicle coupling. A time domain model of road unevenness excitation was established by using linear filter white noise method. According to modern control theory, the state equation of suspension system with multiple inputs and multiple outputs was established. An active suspension controller was designed based on linear quadratic Gaussian (LQG) control theory, and the weighted parameters of the controller were optimized based on particle swarm optimization (PSO). The simulation results indicate that the ride comfort of the vehicle can be greatly improved under the control of the active suspension system. At the same time, the handling stability and structural stability of the vehicle can be enhanced, and the overall performance of the vehicle can be comprehensively improved.