Tokyo-based Kudan Inc. demonstrated the use of a 3D-lidar simultaneous localization and mapping (SLAM) device to create a sharp point cloud without using an external GNSS receiver or inertial measurement unit (IMU).
Kudan is a research and development company specializing in algorithms for artificial perception. For the demonstration, Kudan used its localization and mapping software Kudan 3D-Lidar SLAM, or KdLidar.
Using an Ouster OS1-64 lidar and only its internal IMU, Kudan demonstrated how it can create a crisp high-resolution point cloud with KdLidar.
For the demonstration, the company used a handheld scanner with the Ouster lidar. Handheld scanners can introduce noise and fuzziness in typical point clouds generated from SLAM because of natural vibration, movement and limited field of view. However, Kudan was able to capture sharp wall definition of buildings and structures, as well as the fine detail of powerlines.
While the demonstration highlights KdLidar’s basic performance without any external sensors, the company said its algorithms can further increase the performance and quality of lidar-based scanners by fusing GNSS receivers and external IMUs or inertial navigation systems.