<![CDATA[This study examines and compares three Autonomous Underwater Vehicles (AUV) steering control techniques utilizing the following three control algorithms: Proportional-Integral-Derivative (PID), Fractional Order PID (FOIPD), and Neural Network-PID (NN-PID). The objective of this investigation is to gain a comprehensive understanding of each controller's response in terms of step input scenarios, trajectory changes, and when encountering disturbances. The response analysis will evaluate the strengths and weaknesses of the controller by examining parameters such as Rise Time, Settling Time, Settling Min, Settling Max, Overshoot, Peak, and Peak Time for each controller response. To determine the accuracy performance of each controller strategy, the root mean square error (RMSE) technique will be applied, allowing users to confidently select the most suitable controller option. FOPID displays the best settling time of 3.2218 seconds, and PID stands out in rise time, achieving 0.4725 seconds. The results indicate that NN-PID is the top performer as it reduces overshoot to 0.3022%. Among the three controllers that were tested, FOPID had the smallest RMSE value, while the NN-PID control's slower response and larger error resulted in a smaller overshoot than PID and FOPID. This factor is due to the online learning process on NN-PID, which requires time. Based on the simulation results, FOPID outperforms PID in settling time and produces the smallest error due to the inclusion of parameters λ and μ, leading to improved control performance.]]>
Copyrights © 2024
