Increased Autonomy and Situation Awareness for ROV Operations

This paper proposes a semi-autonomous mission planning and management architecture for a Remotely Operated Vehicle (ROV). The work has focused on three main aspects to increase the autonomy of ROV operations: developing a plan-based mission management architecture for ROV global navigation and local intervention, integrating path planning as a refinement of actions, and real-time communication with visual SLAM (VSLAM) systems for obstacle detection and motion estimation. The architecture is inspired by the hybrid agent architecture [1] using a deliberative and a reactive layer to perform planned tasks and handle contingencies. Hardware-in-the-loop (HIL) simulations are carried out to test and verify the capability and feasibility of the proposed layered mission management architecture. The results demonstrate that the mission planning and management system can automatically generate the correct plan and guide the ROV to achieve its mission goals.