Disturbance Estimation

Disturbance estimation focuses on accurately identifying and quantifying unexpected forces or effects impacting a system's performance, aiming to improve control and prediction accuracy. Current research emphasizes diverse approaches, including deep learning models (e.g., convolutional neural networks) for image-based disturbance mapping, model predictive control integrated with Gaussian processes or other data-driven methods for real-time estimation, and observer-based techniques like sliding mode observers or Kalman filters for various robotic and mechanical systems. These advancements have significant implications for improving the robustness and reliability of robotic systems, autonomous vehicles, and other complex dynamic systems across diverse applications, from underwater exploration to human-robot collaboration.