<div>Taking the semi-active suspension system as the research object, the forward
model and inverse model of a continuous damping control (CDC) damper are
established based on the characteristic test of the CDC damper. A multi-mode
semi-active suspension controller is designed to meet the diverse requirements
of vehicle performance under different road conditions. The controller
parameters of each mode are determined using a genetic algorithm. In order to
achieve automatic switching of the controller modes under different road
conditions, a method is proposed to identify the road roughness based on the
sprung mass acceleration. The average of the ratio between the squared sprung
mass acceleration and the vehicle speed within a specific time window is taken
as the identification indicator for road roughness. Simulation results show that
the proposed road roughness identification method can accurately identify smooth
roads (Class A–B), slightly rough roads (Class C), and severely rough roads
(Class D–H). The designed multi-mode semi-active suspension controller
automatically adapts to the identified road roughness, resulting in improved
ride comfort on severely rough roads and improved handling performance on smooth
roads. Finally, a real vehicle test is performed. The test results show that the
proposed road roughness identification method can effectively distinguish
between a well-paved roads and rough roads. In addition, the ride comfort of the
vehicle is significantly improved in the comfort mode of the controller on rough
roads.</div>