Velocity-Free Localization of Autonomous Driverless Vehicles in Underground Intelligent Mines

Abstract

The rapid progress of science and technology has created favorable conditions for the development of autonomous driverless vehicles. However, the complex conditions in the cities greatly limit the application and promotion of autonomous driving technology. To solve the problem, the underground mine is recommended to apply as a pilot for the promotion of autonomous driverless vehicles. The autonomous vehicles are all connected to the internet to form a vehicle-to-vehicle (V2V) system. Because of the multi-levels and multi-stopes in underground mines, the key point is to obtain the precise locations of the vehicles in real-time. Therefore, the microseismic monitoring, image recognition technology, artificial intelligence training, and smart sensors are comprehensively utilized, based on the internet of things and cloud computing, to locate the autonomous driverless vehicles. Virtual sources localization and pencil lead break tests are conducted to simulate and eliminate the effectiveness and accuracy of the velocity-free localization method. Virtual sources localization demonstrate that the velocity-free localization method is accurate and reliable. Results of pencil lead break tests show that the average locating errors of X coordinates, Y coordinates, and absolute distance are 0.4318 cm, 0.1136 cm, and 0.5188 cm, respectively. The application and promotion of the autonomous driverless vehicles in underground mines can not only solve the problems of deep mining and reduce the frequent disasters caused by the harsh conditions but also can protect the life and property of the workers, as well as provide technical support for the safe and efficient recycling of deep resources.