☆ 4.6 Article

A Closed-Loop Output Error Approach for Physics-Informed Trajectory Inference Using Online Data

IEEE TRANSACTIONS ON CYBERNETICS (2023)

Related references

Note: Only part of the references are listed.

Article Computer Science, Artificial Intelligence

Model-free reinforcement learning from expert demonstrations: a survey

Jorge Ramirez et al.

Summary: Reinforcement learning from expert demonstrations (RLED) is a promising approach that combines imitation learning with reinforcement learning to improve sample efficiency in high-dimensional spaces. It considers prior knowledge and online knowledge as two possible sources to guide the reinforcement learning process. The survey focuses on novel methods for model-free reinforcement learning guided through demonstrations, and discusses challenges, applications, and promising approaches for improvement.

ARTIFICIAL INTELLIGENCE REVIEW (2022)

Add to Collection

Article Computer Science, Artificial Intelligence

A complementary learning approach for expertise transference of human-optimized controllers

Adolfo Perrusquia

Summary: This paper proposes a complementary learning scheme for experience transference of unknown continuous time linear systems, inspired by the learning properties of hippocampus and neocortex via the striatum. The model involves optimizing controller data for the hippocampus and implementing a Q reinforcement learning algorithm for the neocortex, with an inverse reinforcement learning algorithm for complementary learning. Convergence of the proposed approach is analyzed using Lyapunov recursions, and simulations are carried out to verify its effectiveness.

NEURAL NETWORKS (2022)

Add to Collection

Article Automation & Control Systems

Stable robot manipulator parameter identification: A closed-loop input error approach

Adolfo Perrusquia et al.

Summary: This paper presents an on-line parametric estimation method for robot manipulators, which improves the convergence and robustness of parameter estimation through the estimation algorithm and controller.

AUTOMATICA (2022)

Add to Collection

Article Automation & Control Systems

Neural H2 Control Using Continuous-Time Reinforcement Learning

Adolfo Perrusquia et al.

Summary: This article discusses the application of continuous-time H-2 control in unknown nonlinear systems. We use differential neural networks to model the system and apply H-2 tracking control based on the neural model. Due to the sensitivity of neural H-2 control to neural modeling errors, we use reinforcement learning to improve control performance. The stability of neural modeling and H-2 tracking control is proven, and the convergence of the approach is also given. The proposed method is validated with two benchmark control problems.

IEEE TRANSACTIONS ON CYBERNETICS (2022)

Add to Collection

Article Automation & Control Systems

Finite-Time-Prescribed Performance-Based Adaptive Fuzzy Control for Strict-Feedback Nonlinear Systems With Dynamic Uncertainty and Actuator Faults

Huanqing Wang et al.

Summary: This article discusses finite-time-prescribed performance-based adaptive fuzzy control for a class of strict-feedback systems in the presence of actuator faults and dynamic disturbances. An adaptive fuzzy fault-tolerant control strategy is introduced to deal with the difficulties associated with actuator faults and external disturbances. A modified performance function called the finite-time performance function (FTPF) is presented, which ensures that all signals of the closed-loop system are bounded and the tracking error converges to a predetermined region in finite time. The effectiveness of the controller is verified through simulation results.

IEEE TRANSACTIONS ON CYBERNETICS (2022)

Add to Collection

Article Engineering, Civil

Graph-Based Spatial-Temporal Convolutional Network for Vehicle Trajectory Prediction in Autonomous Driving

Zihao Sheng et al.

Summary: This paper proposes a graph-based spatial-temporal convolutional network (GSTCN) to predict future trajectory distributions of neighbor vehicles. The network combines graph convolutional network (GCN) and convolutional neural network (CNN) to capture spatial interactions and temporal features between vehicles, resulting in accurate trajectory predictions.

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS (2022)

Add to Collection

Article Computer Science, Artificial Intelligence

Multi-agent reinforcement learning for redundant robot control in task-space

Adolfo Perrusquia et al.

Summary: This paper proposes a fully cooperative multi-agent reinforcement learning to solve the kinematic problem of redundant robots. Each joint of the robot is regarded as one agent, avoiding function approximators and large learning space. Experimental results show that the proposed MARL outperforms classic methods such as Jacobian-based methods and neural networks.

INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS (2021)

Add to Collection

Article Mechanics

Design and validation of a dynamic parameter identification model for industrial manipulator robots

Claudio Urrea et al.

Summary: This article introduces a regression model for identifying dynamic parameters in manipulator robots, which has advantages in implementation by utilizing position, speed, and voltage data instead of acceleration and torque calculation. The model is simulated and validated using MATLAB/Simulink software for 2-DoF and 5-DoF manipulator robots, and experimentally validated on a real 5-DoF redundant manipulator robot, demonstrating significant advantages in implementation.

ARCHIVE OF APPLIED MECHANICS (2021)

Add to Collection

Article Computer Science, Artificial Intelligence

Identification and optimal control of nonlinear systems using recurrent neural networks and reinforcement learning: An overview

Adolfo Perrusquia et al.

Summary: This paper reviews the identification and optimal control problems using recurrent neural networks and reinforcement learning for nonlinear systems in both discrete and continuous time, discussing methods such as the use of Lyapunov and Riccati equations, linear quadratic regulator formulation, and the normalized gradient descent algorithm.

NEUROCOMPUTING (2021)

Add to Collection

Article Chemistry, Analytical

Particle Filters: A Hands-On Tutorial

Jos Elfring et al.

Summary: The particle filter algorithm has become popular since the early 1990s and is widely used for solving estimation problems. Extensive research has improved the standard algorithm, making it more applicable, but selecting and implementing an advanced version of the particle filter can be challenging for those with limited hands-on experience. Lack of implementation details in theory-oriented papers further complicates this task.

SENSORS (2021)

Add to Collection

Article Computer Science, Artificial Intelligence

Discrete-Time H2 Neural Control Using Reinforcement Learning

Adolfo Perrusquia et al.

Summary: This article discusses the H-2 control approach for unknown nonlinear systems in discrete time using a discrete-time recurrent neural network to model the system and applying H-2 tracking control based on the neural model. To improve tracking accuracy and robustness, reinforcement learning and another neural approximator are utilized. The stability of the neural identifier and the H-2 tracking control are proven, and the convergence of the approach is demonstrated through control of pan and tilt robots and surge tanks.

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS (2021)

Add to Collection

Article Automation & Control Systems

Learning Self-Triggered Controllers With Gaussian Processes

Kazumune Hashimoto et al.

Summary: This article investigates the design of self-triggered controllers for networked control systems with unknown plant dynamics using Gaussian process regression. It formulates an optimal control problem to jointly design control and communication policies based on the GP model. An implementation algorithm based on reinforcement learning framework is provided for learning plant dynamics and self-triggered controller. Numerical simulation demonstrates the effectiveness of the proposed approach.

IEEE TRANSACTIONS ON CYBERNETICS (2021)

Add to Collection

Article Automation & Control Systems

Gaussian conditionally Markov sequences: Modeling and characterization

Reza Rezaie et al.

Summary: This paper investigates basic issues and applications of conditional Markov (CM) sequences, with a focus on their application in trajectory modeling. It characterizes nonsingular Gaussian CM sequences and presents their simple yet complete recursive dynamic models. The application of CM sequences in trajectory modeling with destination/waypoint information is discussed.

AUTOMATICA (2021)

Add to Collection

Article Computer Science, Information Systems

Nonlinear control using human behavior learning

Adolfo Perrusquia et al.

Summary: This paper discusses a human behavior learning approach for nonlinear systems control, utilizing neural reinforcement learning algorithms and other cognitive models to accelerate the learning process. By introducing persistent exciting signals and experience replay methods, the learning accuracy is improved and the sensitivity problem of human actions is addressed. The stability and convergence of neural network based reinforcement learning is also discussed, with simulation results confirming the effectiveness of the approach in two benchmark control problems.

INFORMATION SCIENCES (2021)

Add to Collection

Proceedings Paper Computer Science, Cybernetics

An Input Error Method for Parameter Identification of a Class of Euler-Lagrange Systems

Adolfo Perrusquia et al.

Summary: This paper proposes an input error identification algorithm for a class of Euler-Lagrange systems, using a state-observer structure to identify the difference between the real system and an estimated model through input error. The convergence of parameter estimates is ensured by PD controllers and excitation signal, with stability and convergence assessed via Lyapunov stability theory. Simulation studies were conducted to validate the approach.

2021 18TH INTERNATIONAL CONFERENCE ON ELECTRICAL ENGINEERING, COMPUTING SCIENCE AND AUTOMATIC CONTROL (CCE 2021) (2021)

Add to Collection

Proceedings Paper Acoustics

EKFNET: LEARNING SYSTEM NOISE STATISTICS FROM MEASUREMENT DATA

Liang Xu et al.

Summary: In this paper, a new learning framework EKFNet is proposed to automatically estimate the best process and measurement noise covariance pair, reducing the time and manpower spent on fine-tuning an EKF. The method outperforms existing methods and a manually tuned EKF, showcasing its superior performance using real GPS data.

2021 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP 2021) (2021)

Add to Collection

Article Automation & Control Systems

Continuous-time reinforcement learning for robust control under worst-case uncertainty

Adolfo Perrusquia et al.

Summary: In this paper, a novel continuous-time reinforcement learning method using neural network approximation is proposed to train algorithms with worst-case uncertainty, resulting in a more robust control policy. Experimental results demonstrate that this approach is more robust than the model-based LQR method and the well-known AC method.

INTERNATIONAL JOURNAL OF SYSTEMS SCIENCE (2021)

Add to Collection

Article Robotics

Simplified Stable Admittance Control Using End-Effector Orientations

Wen Yu et al.

INTERNATIONAL JOURNAL OF SOCIAL ROBOTICS (2020)

Add to Collection

Article Automation & Control Systems

State-Space Realizations and Optimal Smoothing for Gaussian Generalized Reciprocal Processes

Langford B. White et al.

IEEE TRANSACTIONS ON AUTOMATIC CONTROL (2020)

Add to Collection

Article Automation & Control Systems

Robust control under worst-case uncertainty for unknown nonlinear systems using modified reinforcement learning

Adolfo Perrusquia et al.

INTERNATIONAL JOURNAL OF ROBUST AND NONLINEAR CONTROL (2020)

Add to Collection

Article Automation & Control Systems

Output Error Methods for Robot Identification

Mathieu Brunot et al.

JOURNAL OF DYNAMIC SYSTEMS MEASUREMENT AND CONTROL-TRANSACTIONS OF THE ASME (2020)

Add to Collection

Article Engineering, Multidisciplinary

A Robust Extended Kalman Filter Applied to Ultrawideband Positioning

Chuanyang Wang et al.

MATHEMATICAL PROBLEMS IN ENGINEERING (2020)

Add to Collection

Article Robotics

Variational Inference With Parameter Learning Applied to Vehicle Trajectory Estimation

Jeremy Nathan Wong et al.

IEEE ROBOTICS AND AUTOMATION LETTERS (2020)

Add to Collection

Article Automation & Control Systems

Reinforcement Learning-Based Linear Quadratic Regulation of Continuous-Time Systems Using Dynamic Output Feedback

Syed Ali Asad Rizvi et al.

IEEE TRANSACTIONS ON CYBERNETICS (2020)

Add to Collection

Article Engineering, Industrial

Position/force control of robot manipulators using reinforcement learning

Adolfo Perrusquia et al.

INDUSTRIAL ROBOT-THE INTERNATIONAL JOURNAL OF ROBOTICS RESEARCH AND APPLICATION (2019)

Add to Collection

Article Computer Science, Artificial Intelligence

Output Feedback Q-Learning Control for the Discrete-Time Linear Quadratic Regulator Problem

Syed Ali Asad Rizvi et al.

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS (2019)

Add to Collection

Article Automation & Control Systems

An improved instrumental variable method for industrial robot model identification

M. Brunot et al.

CONTROL ENGINEERING PRACTICE (2018)

Add to Collection

Article Computer Science, Artificial Intelligence

Optimal and Autonomous Control Using Reinforcement Learning: A Survey

Bahare Kiumarsi et al.

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS (2018)

Add to Collection

Article Automation & Control Systems

Q-learning for continuous-time linear systems: A model-free infinite horizon optimal control approach

Kyriakos G. Vamvoudakis

SYSTEMS & CONTROL LETTERS (2017)

Add to Collection

Article Engineering, Electrical & Electronic

Cooperative parallel particle filters for online model selection and applications to urban mobility

Luca Martino et al.

DIGITAL SIGNAL PROCESSING (2017)

Add to Collection

Article Automation & Control Systems

Comparison Between the CLOE Method and the DIDIM Method for Robots Identification

Alexandre Janot et al.

IEEE TRANSACTIONS ON CONTROL SYSTEMS TECHNOLOGY (2014)

Add to Collection

Article Automation & Control Systems

A Generic Instrumental Variable Approach for Industrial Robot Identification

Alexandre Janot et al.

IEEE TRANSACTIONS ON CONTROL SYSTEMS TECHNOLOGY (2014)

Add to Collection

Article Automation & Control Systems

Identification of Physical Parameters and Instrumental Variables Validation With Two-Stage Least Squares Estimator

Alexandre Janot et al.

IEEE TRANSACTIONS ON CONTROL SYSTEMS TECHNOLOGY (2013)

Add to Collection

Article Automation & Control Systems

A New Closed-Loop Output Error Method for Parameter Identification of Robot Dynamics

Maxime Gautier et al.

IEEE TRANSACTIONS ON CONTROL SYSTEMS TECHNOLOGY (2013)

Add to Collection

Article Automation & Control Systems

DC servomechanism parameter identification: A closed loop input error approach

Ruben Garrido et al.

ISA TRANSACTIONS (2012)

Add to Collection

Article Automation & Control Systems

Online actor-critic algorithm to solve the continuous-time infinite horizon optimal control problem

Kyriakos G. Vamvoudakis et al.

AUTOMATICA (2010)

Add to Collection

Article Robotics

Real-time adaptive control for haptic telemanipulation with Kalman active observers

Rui Cortesao et al.

IEEE TRANSACTIONS ON ROBOTICS (2006)

Add to Collection

Article Automation & Control Systems

Continuous-time model identification from sampled data: implementation issues and performance evaluation

H Garnier et al.

INTERNATIONAL JOURNAL OF CONTROL (2003)

Add to Collection

© Peeref 2019-2024. All rights reserved.