Unmanned Systems Technology 003 | UAV Solutions Talon 120 | Cable harnesses | Austro Engine AE50R and AE300 | Autonomous mining | AUVSI 2015 show report | Transponders | Space systems

59 Autonomous mining | Insight The mine management software is also a key element of the autonomous operation of the whole mine. To this end, ASI has developed a module, called Haulage AI, for its Mobius command and control software. The module uses algorithms to automatically allocate tasks to multiple robotic vehicles, freeing up operators to handle more vehicles or perform other duties. The AI module then maintains safe vehicle spacing, manages queues of vehicles and dynamically positions the loading and dumping areas. Of course, the navigation and control systems for mining equipment sometimes have to operate underground, where GPS signals can’t be used and wireless signals struggle to reach. Underground operation tends to use the remote control option with direct line of sight between the operator and the vehicle, although a carefully designed wireless network can allow the teleop approach to be used underground. The easiest way to design and implement an underground network is to use a mesh, which has multiple nodes that feed data back to a central wireless access point. However, the data rate drops for each additional node used, which can limit the use of video for a teleop application, and it also increases the latency as the signal has to travel through several nodes. Equipment suppliers such as Swedish drill maker Sandvik have therefore been looking at autonomous control underground that needs much less data to be transferred, limiting the data to vehicle system updates and destinations, and reducing the requirements on the network. To do this, Sandvik has patented a system for guiding self-propelled equipment through underground mine corridors, which change regularly as a result of the mining. The vehicle contains a signal generator – which could be Lidar, infrared or even sonar – that bounces signals off the walls of the passageway; a receiver on the vehicle collects the returned signals to determine the distance to the walls. A central processor then determines all the possible routes through the mine, and selects the one that takes it towards its assigned destination. The processor then uses the positioning data to steer the vehicle through the passageways. This avoids the need to keep updating accurate maps of the mine in both the central management software and on the vehicle. Conclusion Mining brings its own set of challenges for autonomous equipment. Over the past decade, the industry has moved from remote control through teleop to full autonomy, with self-driving haulage trucks being a particular highlight of the mine of the future. Being able to run fleets of autonomous trucks, bulldozers and excavators efficiently – both above and below ground – has meant bringing together many sensor, navigation and control technologies to provide the data the vehicles need. These technologies have been tested out in the hostile environments that represent a typical mine and are now being rolled out in real-world applications. Behind the essential technology in the vehicles is a significant software infrastructure for their management, which is ensuring that all the equipment is working as efficiently as possible, with excavators and haulage trucks tightly synchronised. At the same time, the sensors in the equipment can monitor the health of the vehicles and return them to base for repair before they fail, with a replacement being allocated and operating immediately. All this allows the mine to operate as safely and efficiently as possible. Unmanned Systems Technology | Summer 2015 Drilling equipment specialist Sandvik has patented a technique to steer autonomous vehicles through mine tunnels (Courtesy of Sandvik) The easiest way to design an underground network is to use a mesh, which has multiple nodes that feed data back to a central point