A novel design of a mobile parallel manipulator (MPM), which is composed by a multi-degree of freedom (DOF) parallel robot and an autonomous wheeled mobile platform, is proposed in this paper to overcome the shortcomings of the parallel robot and perform high accuracy tasks. Both the position and differential kinematics problems for the hybrid system are solved in details, and the dynamic model for the MPM is derived via Lagrangian formulation approach with simplifying hypotheses. Furthermore, taking the MPM self motion into consideration due to its redundancy, the dynamic control in task space is implemented by employing a model-based controller. Simulation results illustrate that not only the control algorithms are effective well, but also the established dynamic models are accurate enough as well. © 2006 IEEE.