A Simple and Effective Human-Robot Shared Control Approach for Agricultural Robots Employed in Table-Grape Vineyards

This paper introduces a simple yet effective shared control architecture to enable a human operator to enhance robot autonomy in complex scenarios, such as those encountered in agriculture. Specifically, the proposed approach investigates dynamically exchanging control between the robotic system and a human operator depending on the uncertainty in the environment perception. This is achieved by envisioning three control modes that exhibit varying levels of autonomy for the robot: $i$) autonomous, ii) teleoperation and iii) hand-guiding control modes. The approach is embedded into a Hierarchical Quadratic Programming (HQP) control framework, that allows the robot to simultaneously address different control objectives with different priorities. Experimental results conducted as part of the EU-funded CANOPIES project, involving a mobile robot equipped with a mobile base and a dual-arm torso engaged in harvesting operations, demonstrate the effectiveness of the proposed method in real-world conditions.