Implementation of car-following system using LiDAR detection

Abstract

Listen

Translate article

This paper presents a ACC-like car following system to track a preceding car and keep a safe distance based on a laser imaging, detection, and ranging (LiDAR) sensor. The function of car-following has the ability of tracking control, including longitudinal and lateral control. System has two tasks which are environment identification and steering control. To accomplish the car following sensing task, environment information provided by LiDAR and monocular vision is used to verify the correctness of LiDAR detection. A camera is an embedded device to map LiDAR data onto image and verify the detection ability. LiDAR sensor which has one-dimensional scanning ability can measure the relative distance from preceding vehicle by scanning the horizontal plane with laser beams. The speed variation tracking is implemented by Fuzzy algorithm, and vehicle control is applied as the steering control strategy to keep up with the trajectory in lateral control. The control actuator gets related data using CAN. Environmental clutter becomes the main challenge in data processing when LiDAR tries to track the desired vehicle. From LiDAR data and camera information, this paper also provided a data fusion algorithm to do verified consideration. From relative position, the steering control is applied with curvature in pure pursuit control. In related experiments on ARTC campus, this system shows its ability to enhance driver assistance service and realize an adaptive tracking safety by using sensing information.