Hybrid vision/force control for underwater vehicles landing on unknown surfaces

In this work, a hybrid vision/force controller is proposed for landing an underwater vehicle on a surface without exact surface attitude angles (e.g., dam or ship bottom) and maintaining the desired contact force. The visual impedance model is employed to adjust the reference HVS state vector according to the external force to ensure the dynamic performance during the landing process. The external contact force is estimated by a force observer in the absence of force sensors. The desired velocity is generated by a linear feedback control and modified by an optimization-based approach considering the camera constraints. In the dynamic velocity control, dynamic inversion with an adaptive neural network is used to overcome uncertain disturbances. Finally, numerical simulation results are presented to validate the effectiveness of the proposed landing strategy.