A kinematically redundant manipulator possesses more joints than those strictly required to execute its task. This provides the robot with an increased level of dexterity that may be used to avoid singularities, joint limits, and workspace obstacles, but also to minimize joint torque, energy or, in general, to optimize suitable performance indexes.Application of robots in industrial sectors has been very impressive with new advances in technology. Nowadays, mobile robots are used in many applied industrial applications due to their high mobility. One of the properties of these robots is its redundancy of degrees of freedom. Although, this redundancy makes the kinematics, dynamics and control of these robotic systems more complicated, however, it reduces singular points or it has the ability to do secondary tasks during robot motion. Path planning of a redundant robot is highly dependent on the kinematic and dynamic model of the robot. In this thesis, an adaptive path planning and control of a wheeled mobile manipulator which is inherently a redundant robot has been investigated. Also, the problem of path planning and control of the robot for multi-priority task operations has been considered.First of all, optimal path planning problem for a set of kinematic and dynamic indices is done analytically. Due to the model-based nature of optimal path planning problem, the design and adaptive control of an optimal path have been used as an approach to tackle the problem. Study shows that using improved parameters of an adaptive controller in the path planning section caused more optimal designed paths. This study has been performed for correcting an optimal path planning problem with dynamic indices, since one can expect more effect of the adapted parameters.In a path planning problem with local indices, a new method is provided for real time and approximate optimization and implemented on the wheeled mobile manipulator. In the proposed method, the optimizer section has been combined to the tracker section. Tracker has the responsibility of robot's end effector trajectory control on the desired path, while the approximate optimizer has the role of optimizing the kinematic and dynamic indices. This optimizer has the capability to be implemented online for a wide range of indices.Also, in this thesis, for multi-task operation, robot's dynamics has been decoupled into the task space motion and configuration space motion .Considering priority of the task space motion to the configuration space motion, control of the robot in this operation has been investigated. This approach can be used in a wheeled mobile robot to perform an obstacle hitting free task. Key Words: Redundant manipulator , Wheeled Mobile Manipulator, Optimal Path Planning, Adaptive Control
