Design of an Unmanned Aircraft Localization System in a Boiler Furnace

In response to the inspection method in the boiler hearth of China's coal-fired units, the facility is usually inspected for defects by manual hand-held instruments or naked eyes. Not only are the conditions difficult and intense, but also the efficiency is low, which can no longer adapt to the safe operation and development of the modernized hearth. In recent years, with the rapid development of drone technology and related sensor technology, drones have a wider application in hearth inspection. In view of this, a scheme using sensors such as LiDAR combined with UAVs to realize real-time positioning and map construction in the boiler hearth of coal-fired units is proposed. The scheme uses LiDAR as the sensor for environment sensing and Gaussian Newton matching algorithm for secondary point cloud matching, while this paper improves the original bilinear interpolation method to bicubic interpolation method for map construction. The experimental results show that within the boiler hearth environment, using the ideal GPS environment as a reference, the trajectory errors of the UAV position in the X-axis, Y-axis, and Z-axis directions are within 0.3m, and the flight can be accurately controlled.