ABSTRACT
Open-pit mining is very difficult to automate because the bot must be efficient enough to track itself and then extract the minerals even with poor visibility in the mine. In this chapter, we propose methods to make an autonomous bot to work in an open-pit mine. This research is done on wall follower, line follower driving through lane detection, and AprilTag. It compares all the existing techniques and proposes a new approach, i.e., AprilTag. These work much better in comfort driving the bot based on the coordinates of the mining pit. The live coordinates from the camera above the pit are compared with the coordinates of the path on which it was previously driven. These coordinates are stored to and accessed from a comma-separated values (CSV) file. A python-based graphical user interface was also designed to make the process of training and testing the bot more efficient. A walk-through of the same is also provided in this chapter. The autonomous bot is also capable with some fallback conditions to prevent itself from falling inside the pit and getting damaged. The bot exchanges data between Arduino and Raspberry Pi using serial communication. Our proposed approach is suitable for industrial use and robust in real-time scenarios; therefore, it can be applied in modern industrialized systems.
