Planning Collision-Free Robot Motions in a Human–Robot Shared Workspace via Mixed Reality and Sensor-Fusion Skeleton Tracking

The paper describes a method for planning collision-free motions of an industrial manipulator that shares the workspace with human operators during human–robot collaborative application strict safety requirements. proposed workflow exploits advantages mixed reality to insert real entities into virtual scene, wherein robot control command is computed and validated by simulating without risks human. motion planner relies on sensor-fusion algorithm improves 3D perception humans inside workspace. Such merges estimations pose bones reconstructed means pointcloud-based skeleton tracking orientation data acquired from wearable inertial measurement units (IMUs) supposed be fixed bones. provides final reconstruction position can used include in simulation robotic workcell. A dynamic motion-planning processed within such mixed-reality environment, allowing computation joint velocity robot.