Machine-vision-based Online Self-optimizing Control System for Line Marking Machines

Guanxu LONG¹, Lei SHI², Gongfeng XIN¹, Shenqi GAO³*, Wenliang ZHANG¹, Jicun XU²
¹ Shandong Gaosu Group Co., Ltd. Innovation Research Institute,
Longao North Road, Lixia District, Jinan, Shandong, 250101, China
longgx5@163.com, gfxin@163.com, wenliangzhang@163.com
² Transportation Infrastructure Construction Engineering Research Center, Shangdong Jiaotong University,
5001 Haitang Road, Jinan, Shandong, 250357, China
shil@sdjtu.edu.cn, xjctz@126.com
³ School of Information Science and Electrical Engineering, Shangdong Jiaotong University,
5001 Haitang Road, Jinan, Shandong, 250357, China
feize0624@163.com (*Corresponding author)

Abstract: In order to increase speed and efficiency of making line markings, a vision-navigation-based self-optimizing control system is proposed for an unmanned line marking machine (ULMM). A new Haar-like-feature based algorithm is used to detect a guide line (GL) for the ULMM, and to reduce the influence of complex road surfaces and light. For the problems of an inaccurate ULMM model and local navigation information, an online self-optimizing control algorithm is presented, and its self-learning rules are given. Results of simulations and real machine experiments reveal that the proposed navigation algorithm accurately detects the GL, and the precision of the control system satisfies the requirements of the line marking work.

Keywords: Unmanned Line Marking Machine, Vision navigation, Self-optimizing control, Model-free.

>>FULL TEXT: PDF

CITE THIS PAPER AS:
Guanxu LONG, Lei SHI, Gongfeng XIN, Shenqi GAO, Wenliang ZHANG, Jicun XU, Machine-vision-based Online Self-optimizing Control System for Line Marking Machines, Studies in Informatics and Control, ISSN 1220-1766, vol. 32(2), pp. 93-104, 2023. https://doi.org/10.24846/v32i2y202309