| Motion planning of underwater vehicle-manipulator systems (UVMS) is one of the most important aspects for autonomous task execution. The motion planning of a UVMS is a difficult problem because of several reasons. First, a UVMS is a kinematically redundant system. Such a system admits infinite number of joint space solutions for a given Cartesian space coordinates, and thus makes motion planning difficult. Second, a UVMS is composed of two dynamic subsystems, the vehicle and the manipulator, whose bandwidths are vastly different. Thus any successful motion planning algorithm must consider this different dynamic bandwidth property of the UVMS.; In this research work, a new way of deriving the dynamic equations of motion for a UVMS has been developed using the Quasi-Lagrange formulation. Several new algorithms for motion planning of a UVMS have been developed exploiting kinematic redundancy. One of them is Drag Minimization algorithm that is based on acceleration level redundancy resolution. In this work a novel approach to fault tolerant motion planning and control has been developed using a weighted pseudoinverse technique. A new Thruster and Actuator Saturation algorithm has been developed that considers the saturation limits of the thrusters and the actuators. A new Dynamics-Based Motion Planning algorithm has also been developed that directly incorporates the different dynamic bandwidth characteristics of the subsystems. Finally, these algorithms have been integrated to develop a Unified Motion Planning algorithm. This approach provides an important motion planning strategy for a UVMS in hazardous and unstructured underwater environments. The effectiveness of the proposed methods is investigated by computer simulation. The Thruster Fault Tolerant algorithm has also been experimentally verified for an AUV. |
