Fuzzy Logic Deadzone Compensation for a Mobile Robot

Abstract

Mobile manipulators have been introduced as a way of expanding the effective workspace of robot manipulators. Robots with moving vehicle such as macro-micro manipulators, space manipulators, and underwater robotic vehicles can be used for extending the workspace in repair and maintenance, inspection, welding, cleaning, and machining operation. Mobile manipulators possess strongly coupled dynamics of mobile vehicles and manipulators. With the assumption of known dynamics, much research has been carried out. Yamamoto & Yun (1996) addressed the coordination of locomotion and manipulator motion between the base and the arm, and the problem of following a moving surface. Khatib (1999) proposed the coordination and control of the mobile manipulator with two basic task-oriented controls: end-effector task control and platform self posture control. In (Bayle et al., 2003), the concept of manipulability was generalized to the case of mobile manipulators and the optimization criteria in terms of manipulability were given to generate the controls of the system. Most approaches require the precise knowledge of dynamics of the mobile manipulator, or, they simplify the dynamical model by ignoring dynamics issues, such as vehicle dynamics, payload dynamics, dynamics interactions between the vehicle and the arm, and unknown disturbances such as the dynamic effect caused by terrain irregularity. To handle unknown dynamics of mechanical systems, robust, and adaptive controls have been extensive investigated for robot manipulators and dynamic nonholonomic systems. Dixon et al. (2000) developed a robust tracking and regulation controller for mobile robots. In (Li et al., 2008), adaptive robust output feedback motion/force control strategies were proposed for mobile manipulators under both holonomic and nonholonomic constraints in the presence of uncertainties and disturbances. Impedance control of flexible base mobile manipulator using singular perturbation method and sliding mode control law was presented in (Salehi & Vossoughi, 2008). Because of the difficulty in dynamic modeling, adaptive neural network control has been studied for different classes of systems, such as robotic manipulators (Lewis el al., 1996) and mobile robots (Jang & Chung, 2009). In (Lin & Goldenberg, 2001), adaptive neural network controls have been developed for the motion control of mobile manipulators subject to kinematic constraint. In (Mbede et al., 2005), intelligent navigation is presented for mobile manipulator using adaptive neuro-fuzzy systems. In these schemes, the controls are designed at kinematic level with velocity as input or dynamic level with torque as input, but the actuator dynamics are ignored. Therefore, the actuator nonlinearity deteriorates the system