Mixed Reality Enhanced User Interactive Path Planning for Omnidirectional Mobile Robot

This paper proposes a novel control system for the path planning of an omnidirectional mobile robot based on mixed reality. Most research on mobile robots is carried out in a completely real environment or a completely virtual environment. However, a real environment containing virtual objects has important actual applications. The proposed system can control the movement of the mobile robot in the real environment, as well as the interaction between the mobile robot's motion and virtual objects which can be added to a real environment. First, an interactive interface is presented in the mixed reality device HoloLens. The interface can display the map, path, control command, and other information related to the mobile robot, and it can add virtual objects to the real map to realize a real-time interaction between the mobile robot and the virtual objects. Then, the original path planning algorithm, vector field histogram* (VFH*), is modified in the aspects of the threshold, candidate direction selection, and cost function, to make it more suitable for the scene with virtual objects, reduce the number of calculations required, and improve the security. Experimental results demonstrated that this proposed method can generate the motion path of the mobile robot according to the specific requirements of the operator, and achieve a good obstacle avoidance performance.