Development of a real-time 3D position measurement device using camera image analysis and UWB

To improve the efficiency of decommissioning work at the Fukushima Daiichi Nuclear Power Plant, there is a need for technology that can measure indoor 3D positions with an accuracy of about 10 cm inside buildings. In this study, we developed an embedded system based on a positioning system using multiple wireless UWB devices, which have recently gained attention. This system is combined with a positioning function based on camera image analysis (deep learning) to compensate for its shortcomings. The goal is to achieve real-time 3D positioning with 10 cm accuracy. The embedded system includes a wireless UWB, a camera, a computer, an AR marker, a wireless Zigbee, and a battery. Real-time 3D positioning is detected by strategically arranging multiple devices in space. The proposed system automatically recognized the positions of four newly installed devices from three initially installed devices based on the camera image analysis function. This expansion allowed for improved camera image analysis and UWB location detection capabilities.