Tipover Stability Enhancement of Wheeled Mobile Manipulators Using an Adaptive Neuro- Fuzzy Inference Controller System

Abstract

In this paper an algorithm based on the adaptive neuro-fuzzy controller is provided to enhance the tipover stability of mobile manipulators when they are subjected to predefined trajectories for the end-effector and the vehicle. The controller creates proper configurations for the manipulator to prevent the robot from being overturned. The optimal configuration and thus the most favorable control are obtained through soft computing approaches including a combination of genetic algorithm, neural networks, and fuzzy logic. The proposed algorithm, in this paper, is that a look-up table is designed by employing the obtained values from the genetic algorithm in order to minimize the performance index and by using this data base, rule bases are designed for the ANFIS controller and will be exerted on the actuators to enhance the tipover stability of the mobile manipulator. A numerical example is presented to demonstrate the effectiveness of the proposed algorithm. ANY investigations are done in the fields related to mobile manipulation systems such as path planning, motion planning, trajectory tracking, obstacle avoidance, etc. The subject of optimal stability has a significant role in autonomous robot systems. In this case, an interface between the manipulator and the vehicle plays a vital role in the stability investigation. The effect of dynamic interaction on the coordinated control of mobile manipulators has been studied in (1). This effect is examined on the tracking of a mobile manipulator. A nonlinear feedback controller is designed that is capable of fully compensating the dynamic interactions. Stability analysis of mobile manipulators is considered in (2, 3). The stability degree and the valid stable regions based on the Zero Moment Point (ZMP) criterion are