Self-localization method combining laser devices and image sensors in non-GNSS environments

In the promotion of Information and Communication Technology (ICT) in construction projects (i-Construction), as well as in the digitization of existing infrastructure (DX; Digital Transformation,) many improved distance measurement technologies enhance the efficient development and management of societal infrastructure. Infrastructure maintenance and management incorporate such technologies as small unmanned aerial vehicles (drones) and 3D laser scanners. Recently, the utilization of Mobile Mapping Systems (MMS) equipped with laser scanners has facilitated the acquisition of three-dimensional data in areas surrounding roads. MMS proves effective in measuring the three-dimensional shape of roadways as it acquires cloud point data. The data, represented by three-dimensional coordinates (x, y, z) aligned with the world geodetic system, can be seamlessly integrated into cyberspace on a computer. MMS utilizes a laser beam from a scanner installed on the vehicle, resulting in a measurement error of a few centimeters in the acquired cloud point data. However, the accuracy of the cloud point data significantly decreases in tunnels due to the absence of satellite positioning correction, leading to potential errors exceeding 1 meter. Presently, long distance [1 m to 30 m] measurement commonly relies on the Doppler effect. However, notable issues are the requirement for a stationary state and several seconds of measurement time to ensure accuracy. Hence, when mounted on a moving vehicle, poor quality data would be obtained. Similarly, the use of beacons or radar yields unreliable measured values for the vehicle movement. Even with autonomous navigation systems employing a gyro sensor and a speed sensor mounted on the vehicle to determine the point of travel, the cumulative error in position estimation gradually increases, particularly in locations where catching GPS satellites is challenging. Some essential sites on and around roads, such as tunnels and areas under elevated railway tracks, remain inaccessible for position information through GNSS (Global Navigation Satellite System,) limiting the full utilization of ICT technology. This study proposes a novel self-positioning system that integrates two laser devices and an image sensor to accurately acquire position information even in non-GNSS environments.