Hysteresis Compensation

Hysteresis compensation aims to mitigate the inaccuracies and unpredictable behavior caused by hysteresis—a phenomenon where a system's output depends not only on the current input but also on its history. Current research focuses on developing accurate hysteresis models using various machine learning approaches, including neural networks (feedforward, recurrent, and operator networks), temporal convolutional networks, and fuzzy logic systems, often applied to specific systems like soft robots, magnetic devices, and continuum manipulators. These advancements are crucial for improving the precision and reliability of control systems in diverse fields, ranging from robotics and medical devices to materials science and traffic management. The ultimate goal is to enable more accurate and robust control of systems exhibiting hysteresis, leading to improved performance and efficiency in various applications.

Papers

March 10, 2023

A Data-Driven Model with Hysteresis Compensation for I2RIS Robot
Mojtaba Esfandiari, Yanlin Zhou, Shervin Dehghani, Muhammad Hadi, Adnan Munawar, Henry Phalen, Peter Gehlbach, Russell H. Taylor, Iulian Iordachita
Data Driven Snake Robot Hysteresis Compensation Retinal Microsurgery

March 9, 2023

Segmentation method for cerebral blood vessels from MRA using hysteresis
Georgia Kenyon, Stephan Lau, Michael A. Chappell, Mark Jenkinson
Hessian Matrix Vessel Segmentation Segmentation Method Hysteresis Compensation Tof Mra

February 10, 2023

Discovery of sparse hysteresis models for piezoelectric materials
Abhishek Chandra, Bram Daniels, Mitrofan Curti, Koen Tiels, Elena A. Lomonova, Daniel M. Tartakovsky
Scientific Discovery Hysteresis Compensation Ultrasonic Motor Material Domain

February 6, 2023

Traffic Shaping and Hysteresis Mitigation Using Deep Reinforcement Learning in a Connected Driving Environment
Rami Ammourah, Alireza Talebpour
Reinforcement Learning Deep Reinforcement Learning Multi Agent Deep Reinforcement Learning Connected Vehicle Hysteresis Compensation Congestion Mitigation Traffic Shaping

November 23, 2022

Hybrid Learning of Time-Series Inverse Dynamics Models for Locally Isotropic Robot Motion
Tolga-Can Çallar, Sven Böttger
Rigid Body Local Path Planning Torque Sensor Hysteresis Compensation Hybrid Learning Inverse Dynamic Inverse Dynamic Model

April 29, 2022

Hysteretic Behavior Simulation Based on Pyramid Neural Network:Principle, Network Architecture, Case Study and Explanation
Yongjia Xu, Xinzheng Lu, Yifan Fei, Yuli Huang
Neural Network Line by Line Explanation Surrogate Model Neural Network Architecture General Principle Pyramid Network Network Architecture Hysteresis Compensation Structural Analysis

April 1, 2022

Hysteresis-Based RL: Robustifying Reinforcement Learning-based Control Policies via Hybrid Control
Jan de Priester, Ricardo G. Sanfelice, Nathan van de Wouw
Reinforcement Learning Proximal Policy Optimization Deep Q Network Robust Reinforcement Learning Hybrid Control Hysteresis Compensation

Hysteresis Compensation

Papers

A Data-Driven Model with Hysteresis Compensation for I2RIS Robot

Segmentation method for cerebral blood vessels from MRA using hysteresis

Discovery of sparse hysteresis models for piezoelectric materials

Traffic Shaping and Hysteresis Mitigation Using Deep Reinforcement Learning in a Connected Driving Environment

Hybrid Learning of Time-Series Inverse Dynamics Models for Locally Isotropic Robot Motion

Hysteretic Behavior Simulation Based on Pyramid Neural Network:Principle, Network Architecture, Case Study and Explanation

Hysteresis-Based RL: Robustifying Reinforcement Learning-based Control Policies via Hybrid Control