ROS Based Adjustable Resolution Compact 3D Scanner
Scott Dion, Akin Tatoglu · 2019
Abstract This research study aims to design and develop a compact multi-sensor 3D scanning system for a mobile robot which can detect object types. Sensor suite is equipped with a 3D Lidar with 360-degree field of view and a stationary camera mounted on top of a 2DOF turret. The sensor data is fused to generate a 3D point cloud. Mechanism is required to sweep the 3D space in pitch and roll axes. While servo controller can adjust generated point cloud resolution — and scan duration —, camera images can also be registered to generate a panoramic image of the environment followed by radial distortion corrections. System can generate variable density scans for specific application such mapping or self-driving. This study is the perception portion of our full-scale self-driving golf cart research project of Autonomous Mobile Robotics Research Group.