LIDAR BASED ACTIVE SUSPENSION

Abstract

A car's suspension system is the device that physically separates the vehicle's body from its wheels. Traditional suspension systems have a limited ability to protect the car's chassis from the outside elements. Recent developments in this area go by the label of "active suspensions". As vehicle capabilities have risen, the suspension system's performance has significantly improved. Active suspensions are effective but by no means flawless. There are several things that can be done to make them better. Herein is the significance of this research paper. Using sensors like global navigation satellite systems (GNSSs), light detection and ranging (LiDAR), and inertial measurement units, autonomous cars are able to localize accurately and perceive real-time route information (IMUs). Using route information, cars may automatically go to a certain location without causing any traffic incidents. The majority of research on autonomous cars, although being an important reference for vehicles travelling on uneven terrain, has given little attention to information on road profiles. The precision and stability of data collected by LiDAR and IMUs are decreased by the strong vibrations that the majority of vehicles encounter when travelling over uneven terrain. On the other side, active suspension systems enable vehicles to retain stability on inclines, further ensuring sensor accuracy. In this article, we present a brand-new approach to estimating road profiles that makes use of active suspension systems on automobiles and LiDAR. In the former, an elevation map was supplemented with real-time cloud data points that were collected using 3D laser scanners, IMU, and GPS. In the latter, a Kalman filter method was used to further analyze the elevation map and fuse several cloud data points at a single map cell. The active suspension system is intended to be controlled using the model predictive control (MPC) approach in order to maintain vehicle stability and further lessen drifts of LiDAR and IMU data. The trial findings showed that the proposed approach was accurate and efficient when used in outdoor settings.