Autonomous Driving of Trackless Transport Vehicles: A Case Study in Underground Mines

• Published in: Sensors (Basel, Switzerland) Vol. 25; no. 10; p. 3189
• Main Authors: Sun, Yunjie, Zhang, Linxin, Liu, Junhong, Xu, Yonghe, Li, Xiaoquan
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 19.05.2025; MDPI
• Subjects: Accuracy; autonomous driving; Bar codes; Cameras; Case studies; Construction costs; Efficiency; image recognition; Light; light band; Localization; Medical imaging equipment; metal mine; Methods; Mineral industry; Mines; Mining; Mining industry; Sensors; trackless transport; underground mines; Vehicles; autonomous operation; underground mines; metal mine; image recognition; light band; trackless transport; autonomous driving
• ISSN: 1424-8220; 1424-8220
• DOI: 10.3390/s25103189

The introduction of autonomous vehicles in underground mine trackless transportation systems can significantly reduce safety risks for personnel in production operations and improve transportation efficiency. Current autonomous mining vehicle technology is characterized by complex algorithms and high deployment costs, which limit its widespread application in underground mines. This paper proposes a light-band-guided autonomous driving method for trackless mining vehicles, where a continuous, digitally controllable light band is installed at the tunnel ceiling to provide uninterrupted vehicle guidance. The light band is controlled by an independent hardware system and uses different colors to indicate vehicle movement status, enabling vehicles to navigate simply by following the designated light trajectory. We designed the necessary hardware and software systems and built a physical model for validation. The system enabled multiple vehicles to be guided simultaneously within the same area to perform diverse transportation tasks according to operational requirements. The model vehicles maintained a safe distance from tunnel walls. In GPS-denied environments, positioning was achieved using dead reckoning and periodic location updates at designated points based on the known light-band trajectory. The proposed method demonstrates high potential for practical applications.