中图分类号：P242.6;TP273文献标识码：ADOI:10.19339/j.issn.1674-2583.2026.07.006 中文引用格式：许志远,侯浩天,岳斌斌,等. 激光雷达巷道中线提取与差速小车双环控制实现［J］.集成电路应用，2026，43（7）：29-33. 英文引用格式：Xu Zhiyuan,Hou Haotian,Yue Binbin,et al. LiDAR-based roadway centerline extraction and dual-loop control of a differential-drive vehicle［J］.Application of IC，2026，43（7）：29-33.
LiDAR-based roadway centerline extraction and dual-loop control of a differential-drive vehicle
Xu Zhiyuan,Hou Haotian,Yue Binbin,Zhang Yakun
School of Intelligent Manufacturing and Electrical Engineering, Yellow River Transportation College
Abstract： To meet the demand for high-precision centering driving of automatic guided vehicles in narrow environments such as roadways, an embedded differential-drive vehicle system is designed. The system is based on STM32H7 as the main controller, equipped with M10P PHY 2D LiDAR and a geared motor with an encoder. The LiDAR transmits point cloud data to the STM32H7 in real time via the network port UDP protocol. After parsing, an occupancy grid map of the roadways is generated. The Zhang-Suen thinning algorithm is used to extract the skeleton of the passable area, obtaining a single-pixel-width centerline, and based on this, the lateral deviation and angular deviation are calculated. The control part adopts a cascade double-loop PID structure, with the outer loop responsible for position control and the inner loop for angle control. The differential steering quantity is output to drive the geared motor with an encoder, achieving smooth steering. The system is also equipped with an RGB LCD screen to display the 2-D grid map of the tunnel and the centerline in real time. Experimental results showed that the vehicle could quickly converge to the centerline of the roadways from different initial offset positions, with a control period of approximately 30 ms. The differential steering process was smooth, and the display effect was intuitive, demonstrating good engineering practicability.
Key words : LiDAR; STM32; Zhang-Suen thinning algorithm
