Discrete‐time neural control for electrically driven nonholonomic mobile robots

Abstract

An inverse optimal neural controller for discrete-time unknown nonlinear systems, in the presence of external disturbances and parameter uncertainties, is presented. It is based on a discrete-time recurrent high-order neural network trained with an extended Kalman filter-based algorithm. The applicability of the proposed approach is first tested via simulations for an electrically driven nonholonomic mobile robot, and finally, the proposed methodology is implemented on real time. DISCRETE-TIME NEURAL CONTROL FOR MOBILE ROBOTS 631 for measurement; then, the RHONN is used to design an online adaptive recurrent neural identifier for nonlinear systems, whose mathematical model is assumed to be unknown. It is important to note that because [10] neural identification has been discussed in many publications ([4, 8, 11, 12] and references therein). The learning algorithm for the RHONN is implemented using an EKF. Then, a discrete-time inverse optimal controller is synthesized for the neural model. The applicability of the proposed scheme is illustrated first via simulation results and then experimentally for an electrically driven nonholonomic mobile robot.Traditionally, control of mobile robots only considers its kinematics. It has been well known that the actuator dynamics is an important part of the design of the complete robot dynamics. However, most of the reported results in literature do not consider all of the parametric uncertainties for mobile robots at the actuator level. This is due to the fact that the control problem would become extremely difficult as the complexity of the system dynamics increases and when the mobile robot model includes the uncertainties of the actuator dynamics as well as the uncertainties of the robot kinematics and dynamics [13][14][15][16].x.k/, %.k// C ´i , i D 1, , n

Real-Time Inverse Optimal Neural Control for Image Based Visual Servoing with Nonholonomic Mobile Robots

“…Theorem 1 (see [23]). The RHONN (18) trained with the EKFbased algorithm [20] to identify the nonlinear plant (17) ensures that the identification error [20] is semiglobally uniformly ultimately bounded (SGUUB); moreover, the RHONN weights remain bounded.…”

Adaptive control of electrically‐driven nonholonomic wheeled mobile robots: Taylor series‐based approach with guaranteed asymptotic stability

“…In order to design the closed-loop system regarding to current variables, i.e. 𝑰 𝑚 , we have used equations (30) and (31) to finally obtain the closed-loop system of inner loop (equation 37).…”

Neural Control for Driving a Mobile Robot Integrating Stereo Vision Feedback

“…where χ 11 = x, χ 12 = y are the coordinates of P 0 and χ 13 = θ is the heading angle of the mobile robot, χ 21 = v 1 , χ 22 = v 2 represent the angular velocities of right and left wheels, respectively and χ 31 = i a1 , χ 32 = i a2 represent motor currents of right and left wheels, respectively. R is half of the width of the mobile robot and r is the radius of the wheel, d is the distance from the center of mass P c of the mobile robot to the middle point P 0 between the right and left driving wheels.…”