Autonomous Navigation of AGV by Tracking the Motion of UAV*

Heterogeneous UAV-AGV (Unmanned Aerial Vehicle – Autonomous Ground Vehicle) combination is one of the most focused two-agent systems for the industrial applications. Vision-based UAV-AGV is prominent to execute tasks in GPS (Global Positioning System) inaccessible areas. One of the advantages of UAV in the vision-based system is guiding the navigation of AGV. Map-based and mapless/reactive navigation methods exist in the literature for AGV. Mapless techniques require communication of position of obstacles to the AGV by UAV during navigation. These are suitable if the agents are cooperative/communicative where data communication between agents is possible. In the present work, we propose a mapless technique for the navigation of AGV assisted by UAV without communication of obstacles to AGV. We assume that the AGV is under sudden sensor failure and completely dependent on UAV for its safe navigation. The goal of UAV is to take the AGV to the destination while guiding it to avoid obstacles. We exploit the collaborative tracking task between UAV and AGV for the obstacle avoidance. Tracking can be either UAV tracking AGV or vice-versa. Particularly, AGV tracking the motion of UAV is exploited for the navigation of AGV. Sliding mode control method is implemented for the tracking motion of AGV and obstacle avoidance control. The job of UAV is to localize obstacle in the image plane and guide AGV to avoid it while AGV tracking UAV. Experimental results are presented to validate the proposed method. This proves to be significant technique for safe navigation of AGV when it is non-communicating and under sudden sensor failure.