Hybrid adaptive tracking control method for mobile manipulator robot based on Proportional–Integral–Derivative technique

Abstract

In this research, an adaptive tracking control method for the nonholonomic robot system is addressed based on the hybrid Proportional–Integral–Derivative (PID) technique. The proposed hybrid PID scheme first applies the merits of the traditional PID method, with the online self-learning capability for the PID – gains, to force tracking errors to zero in the presence of uncertainties. Then, in order to improve the tracking performance, an adaptive Fuzzy Neural Networks (FNN) approximator and an adaptive robust controller type-compensator are utilized to relax the uncertainties problems of the robot control system. Moreover, the nonholonomic constraint force stability of the mobile manipulator robot is also considered by an adaptive control scheme. The design of online updating laws for the proposed controllers and FNN approximator are designed by applying the Lyapunov stability theorem. Thus, besides the improvement for tracking control performance, the stability of the proposed control system is also maintained. The effectiveness, robustness and adaptability of the proposed control strategy are verified by comparative numerical simulation results.

Adaptive fault control for robot manipulators based on nonsingular terminal sliding mode control technique

“…Based on the designed algorithms in (10), the result in (A4) can be rewritten as: And by applying the results, 18 as fTf 4b…”

“…This advantage improves adaptability more effectively than that of the related method, such as in Refs. 13,18 where the sliding surface/filtered error functions are fixed. Therefore, when abrupt faults occur, all the control gains of the proposed main NTSM controller are self-updated first to actively assist to ensure the tracking control performance.…”

“…In order to deal with the problems of fixed control gains, the adaptive tuning algorithms based on adaptive/optimal/soft-computing techniques are appropriate approaches. [11][12][13][14][15][16][17][18][19][20] In fact, the control gains in all controllers are one of the proportional (P) -integral (I) -derivative (D) parts of the traditional PID controller. Pan et al 11 recommended a simple method for adjusting the PID gains that were done offline based on the pole placement technique.…”

“…It can realize that these methods have advantages in finding optimal solutions for the control gains, however, the problems of computation burden and deviation in learning/updating processes need to be carefully considered to guarantee for the stability and adaptability of the controlled system. In previous our work, 18 the control gains, the as PI and PD gains, were self-tuned by applying the well-known adaptive and filtered error techniques. This work still has some disadvantages that can be improved as: the gains in filtered error function, as a sliding surface function, are fixed and the updating errors are not considered thoroughly.…”

“…11,12 It noted that the proposed NTSM main controller contains a proportional – derivative part, which can speed up the finite time convergence of the state variables of the sliding surface. In the other hand, different with the previous method in Refs., 14,16,18 the proposed main NTSM controller also shows the advantage that all control parameters are adaptively self-updated based on the information of NTSM surface. Next, the unknown RMs’ dynamics and the uncertain faults are also estimated by the designed adaptive strategy.…”