Neural Network-based Control System Enhanced with a Kalman Filter

Intelligent control systems are increasingly incorporating neural networks to enhance precision and adaptability. This research introduces an optimization approach that integrates an Extended Kalman Filter (EKF) with a Recurrent Fuzzy Neural Network (RFNN) to improve control system output accuracy. The methodology utilizes Matlab simulations, demonstrating the RFNN-EKF algorithm's superiority over traditional RFNN in path planning, obstacle avoidance, and system robustness. The study contributes to the theoretical understanding of neural network optimization and provides empirical validation, paving the way for more advanced autonomous mobile robots.