Although flexible joint robots are widely used in the industry, they are not without problems. It is especially so in their joints, links and complex dynamic where the interaction between loops, non-linearity, and flexibility in the joints can be difficult. The purpose of the present paper is to improve the tracking performance of flexible joint robots. Therefore the physical relations of the system dynamics need to be used to determine a non-linear model for the flexible joint robot. This paper attempts to achieve the desired performance flexible joint robot based on Fuzzy Logic Self-Tuning PID controller. Generally, the classic PID controller is different from the newly introduced form of PID. In classic PID, the parameter values are calculated based on various methods such as Ziegler-Nichols, while in fuzzy logic self-tuning PID, they are obtained by intelligent methods such as fuzzy logic. After deriving the system model, this logic self-tuning PID controller is designed in two cases: using error and its derivative and employing error and its integral for the inputs. The simulation results indicate that the proposed controllers can improve the overall efficiency of the system.
Neural Network and Fuzzy-logic-based Self-tuning PID Control for Quadcopter Path Tracking