Fuzzy-Neuro Optimal Time-Energy Control of a Three Degrees of Freedom Planar Manipulator

Abstract

In this paper, the optimal time-energy trajectory planning problem for a three degrees-of-freedom planar manipulator is considered using a multi-level fuzzy-neural system. First, a neural network is achieved to solve the inverse kinematics problem (IKP). After pre-processing steps characterizing the minimum time trajectory and the corresponding torques, a fuzzy-neuro controller is developed to deal with the minimum energy trajectory planning. The basic function of this controller is to select the minimum energy link actuator torques allowing the robot end-effector to move from a starting point S to a target point T. This optimization is performed under actuators and workspace constraints. It starts with a fuzzy clustering of input-output data, uses Tsukamoto fuzzy inference and back-propagation algorithm to train the resulting adaptive neural network. The premises parameters (antecedent membership functions parameters) as well as rule-consequence parameters are then learned and optimized, giving the optimal energy torques. Simulation results are provided to prove the efficiency of the proposal