A5002359724- Adaptive Control of Nonlinear Systems
- Hydraulic and Pneumatic Systems
- Iterative Learning Control Systems
- Adaptive Dynamic Programming Control
- Control Systems in Engineering
- Mechanical Circulatory Support Devices
- Model Reduction and Neural Networks
- Additive Manufacturing and 3D Printing Technologies
- Welding Techniques and Residual Stresses
- Additive Manufacturing Materials and Processes
- Real-time simulation and control systems
- Advanced Algorithms and Applications
- Dynamics and Control of Mechanical Systems
- Advanced Sensor and Control Systems
- Guidance and Control Systems
- Oil and Gas Production Techniques
- Aerospace Engineering and Control Systems
- Extremum Seeking Control Systems
Nanjing University of Science and Technology
2021-2025
Even though the unprecedented success of AlphaGo Zero demonstrated reinforcement learning as a feasible complex problem solver, research on control hydraulic systems is still void. We are motivated by challenges presented in to develop new model-based controller that achieves high-accuracy tracking at performance level and with asymptotic stability guarantees system level. In this article, proposed design consists two frameworks: A recursive robust integral sign error (RISE) approach...
As hydraulic Euler–Lagrange systems evolve to encompass high-dimension and intricate structures, the formulation of mechanical dynamics model becomes impractical, even though dynamic information remains accessible. In this article, we make first attempt establish model-data hybrid driven controller for systems. The proposed strategy commences with model-based nonlinear robust controller, followed by introduction actor–critic reinforcement learning effectively tackle uncertainties associated...
Due to the difficulty in ensuring launch safety under unfavorable site conditions, restrictions regarding selection of sites significantly weaken maneuverability unsupported random vertical (URVL) mode. In this paper, a nonlinear robust control strategy is proposed missile attitude by actively adjusting oscillation launcher through hydraulic actuator, enhancing launching and adaptability VMLS site. Firstly, considering interaction among canister, adapters, missile, 6-DOF dynamic model system...
This paper investigates the disturbance observer based actor-critic learning control for a class of uncertain nonlinear systems in presence unmodeled dynamics and time-varying disturbances. The proposed algorithm integrates filter-based design method with architecture to circumvent dynamic disturbance. To be specific, actor network is employed estimate unknown system dynamic, critic developed evaluate performance, leveraged provide efficient estimation compounded which includes approximation...
Manipulators are multi-rigid-body systems composed of multiple moving joints. During movement, the Coriolis force, centrifugal and gravity system undergo significant changes. The last three degrees freedom (DOFs) wrist joint a manipulator control end attitude. Improving command tracking accuracy is key challenge in controlling attitude manipulators. In this study, dynamics model mechanical arm–wrist established based on Lagrange method. An adaptive continuous robust integral sign error...
This paper mainly focuses on the development of a learning-based controller for class uncertain mechanical systems modeled by Euler-Lagrange formulation. The considered system can depict behavior large engineering systems, such as vehicular robot manipulators and satellites. All these are often characterized highly nonlinear characteristics, heavy modeling uncertainties unknown perturbations, therefore, accurate-model-based control approaches become unavailable. Motivated challenge,...
This paper focuses on high-accuracy motion control for electromechanical servo systems with time-varying parameter estimation. A robust indirect adaptive controller based a nonlinear X-swapping technique is proposed to achieve the online estimation of parameters. The parametric uncertainties can be well compensated via feedforward cancellation technique, and residual model are suppressed by law. Lyapunov-based stability analysis illustrates that method obtains prescribed estimating result an...
In this article, the nearly optimal tracking control scheme is established for dc motors with partially uncertain dy-namics, including unknown drift dynamic and known direction. The proposed approach adopts inversion concept to acquire feedforward compensation then reinforcement learning techniques feedback. former derived by neural network leveraged estimate dynamic, latter generated direct heuristic programming (dHDP) solve a Hamilton-Jacobi-Bellman (HJB) equation. Eventually, theoretical...
Considering parametric uncertainties and unmodeled disturbances in hydraulic systems, this paper presents an asymptotic adaptive motion control scheme by adopting full state feedback. To accomplish study, a novel parameter law based on both estimation errors tracking is developed to deal with of which can attenuate the major enhance performance. Then composite robust integral sign error (RISE) feedback controller employed restrain rest disturbances. The continuous RISE-based free chattering,...