No abstract

This paper presents an integrated motion planning and control framework for a nonholonomic wheeled mobile manipulator (WMM) system taking advantage of the (differential) flatness property. We first develop the kinematic model of the system and analyze its flatness properties. Subsequently, a statically feedback linearizable system description is developed by appropriately choosing the flat outputs. Motion-planning can now be achieved by polynomial curve fitting to satisfying the terminal conditions in the flat output space while control design reduces to a pole-placement problem for a linear system. A case study of point-to-point motion is considered to study the effectiveness of pose stabilization in the WMM. The simulation and experimental results highlight the ease-of-implementation of proposed method for online real-time integrated motion-planning/control within a hardware-in-the-loop (HIL) electro-mechanical testing.

The potential for use of robotic systems in remote applications arenas has long motivated development of robust and stable means of teleoperated control of slave systems. However, telerobotic systems face challenges stemming from the devices themselves, environmental factors, communication and control complexities. To address these challenges, we will adopt the passivity based synchronization framework [1] and study its applicability to safely synchronize two heterogeneous Lagrangian systems. Within this framework, an adaptive controller identifies and stabilizes the dynamics of the master and slave systems and renders the dynamics passive to a secondary coupling input. The passive mapping used to couple the output states of the master and slave systems and is made insensitive to lossy and delayed communication medium. Specifically, an adaptive passive synchronization teleoperation controller is developed between an Omni haptic device that serves as our master and a differentially driven nonholonomic Wheel Mobile Robot (WMR) as the slave system. A battery of hardware-in-the-loop simulations are used to verify the proposed controller.

No abstract