Applications of mobile MANIPULATORs over uneven terrain, for doing complex tasks autonomously, will be increased. In many cases, robot control algorithms and path planning methods ignore some characteristics of a robot and its environment. Thus they have a limited performance over uneven terrain. Therefore, robot instability that leads to roll over is not an unexpected matter. In this paper, mobile MANIPULATOR rollover recovery, by using a MANIPULATOR and a vision system, is presented. A genetic algorithm and a quasi static model are used to find the optimal recovery process. The criterion of optimization is the power consumption of actuators of the mobile MANIPULATOR. The power consumption depends on the path that the end effectors negotiate over the surface. To estimate the surface curve, a human like stereo vision is used. Finally, simulation results are presented.

This article will introduce a robust vision system which was implemented on a mobile MANIPULATOR. This robot has to find objects and deliver them to pre specified locations. In the first stage, a method which is named color adjacency method was employed. However, this method needs a large amount of memory and the process is very slow on computers with small memories. Therefore since the previous methods, had used statistical method for object detection, the samples for the connectionist were extracted by this method. The obtained neural network is very robust against light changes and can detect objects very quickly. This neural network has the advantage of being extremely simple to implement, and astonishingly quick in practice.

A nonlinear adaptive control algorithm for a multi-task redundant MANIPULATOR is developed in this paper. The method considers the parametric uncertainties in the system and defines a proper filtered error signal according to the allocated priority. Based on this error analysis, the asymptotic stability and convergence of the tracking error, both for the main task as well as the sub-tasks are shown using Lyapunov approach. In order to extend the algorithm for the case of orientation control in the operational task, quaternion feedback has been exploited, and the stability is shown. The results of the paper are verified in several simulations on 4DoF planar arm as well as 7DoF KUKA lightweight robot arm.

A mobile MANIPULATOR robot is a complex system due to properties such as coupling between the MANIPULATOR and mobile chassis, holonomic and nonholonomic constraints, multivariable and nonlinear dynamics. The control of robot faces the external disturbance, parametric uncertainty and unmodeled dynamics. Therefore, the use of an adaptive fuzzy system is suggested for its capability in overcoming uncertainties and approximating of nonlinear functions based on the universal approximation theorem. However, the tracking error does not converge asymptotically to zero due to the approximation error of fuzzy system. This paper presents a novel adaptive fuzzy control for a mobile MANIPULATOR robot. The novelty of paper is compensating the approximation error of fuzzy system for asymptotic convergence in tracking the desired trajectory in the presence of uncertainties. For this purpose, the closed loop system in the error space converges to a linear system with poles having negative real parts. The control design consists of two parts; the kinematic control and dynamic control. Advantages of the proposed design are the simplicity and very good performance in tracking of the desired trajectory in the presence of uncertainties. The stability of control system and convergence to the desired trajectory are proven by the Lyapunov method. The simulation results show the superiority of the proposed control over a robust adaptive control.