Self-adaptive grey wolf optimization based adaptive fuzzy aided sliding mode control for robotic manipulator

As the robotic manipulators are highly nonlinear, it is a challenging task to design, in particular, the PUMA 560 robotic arm with acceptable performance. This paper intends to show the design and development of an adaptive sliding mode controller (SMC) for a robotic manipulator. Since it is not realistic to match the SMC operations with the system model at every time instant, this paper adopts fuzzy inference to replace the system model. This approach successfully achieves the objectives of the experiment, carried out in two stages. In the first stage, it acquires the precise characteristics of the system model for the diverse samples and adequately represents the robotic manipulator. Subsequently, we derive the acquired characteristics in the form of fuzzy rules. In the second stage, we represent the derived fuzzy rules on the basis on adaptive fuzzy membership functions. Further, the approach introduces the self-adaptiveness into a recent algorithm called Grey Wolf Optimization (GWO) in order to establish the adaptive fuzzy membership functions. We then compare the effectiveness of the proposed method with the identified experimental model and the known methods, like SMC, Fuzzy SMC (FSMC), and GWO-SMC. Finally, the comparison with the known methods establishes the effectiveness of the proposed SAGWO-FSMC technique.