On the ESO Based Reinforcement Learning for Pure Feedback Systems

2017 ◽  
Author(s):  
Dazi Li ◽  
Wei Wang ◽  
Zhiqiang Gao
Keyword(s):  
Reinforcement Learning ◽  
Learning Algorithm ◽  
Integral Form ◽  
Feedback System ◽  
State Observer ◽  
Extended State Observer ◽  
Reference Trajectory ◽  
Extended State ◽  
Feedback Systems ◽  
Detailed Model

The control of pure feedback system, which is widely used but has non-affine property, has always been an important and challenging problem. In order to achieve precise tracking control of pure feedback system through improving the disturbance rejection ability of existing reinforcement learning algorithm, a reinforcement learning (RL) control strategy based on extended state observer (ESO) is proposed in this paper. In the proposed method, the extended state observer can reject the total disturbances and transform the pure feedback system which is in an input-output predictor from to overcome the non-causal problem into a cascade integral form. This allows the continuous reinforcement learning strategy of the actor-critic (AC) structure not to depend on the detailed model information, which makes it practically data-driven. It is worth noting that, in order to further improve the ability to track the changing reference trajectory, a novel curvature acceleration factor is proposed, which can adjust the learning speed of the reinforcement learning controller according to the curvature of the reference trajectory. The validity of the proposed algorithm is verified by the simulation results.

Adaptive gearshift control of wet dual clutch transmission based on extended state observer and H∞ robust control

2021 ◽  
pp. 095440622110099
Author(s):  
Kun Cheng ◽  
DaTong Qin ◽  
Junhang Jian ◽  
Bangzhi Wu
Keyword(s):  
Robust Control ◽  
Control Method ◽  
State Observer ◽  
Adaptation Effect ◽  
Extended State Observer ◽  
Reference Trajectory ◽  
Extended State ◽  
Dual Clutch Transmission ◽  
Simulation And Experiment ◽  
Gearshift Control

The clutch characteristics of dual clutch transmission (DCT) will change as the service time increases, which will lead to the deterioration of gearshift performance. To reduce the influence of the change in clutch characteristics on the gearshift performance, an adaptive gearshift control method based on the extended state observer and H∞ robust control is proposed. First, the gearshift problem of the DCT is transformed into the reference trajectory tracking problem, and the gearshift reference trajectory is designed using the minimum principle. The uncertain term related to the change in clutch characteristics in the DCT gearshift dynamic model is defined, and an extended state observer is designed to estimate the uncertain term. On this basis, the gearshift controller is designed using the backstepping method, and H∞ robust control is introduced to further improve the adaptation effect of the controller, then the adaptive control laws of the clutch pressure and engine torque are obtained. Finally, the adaptation effect of the proposed method was verified by both simulation and experiment. The results show that the proposed adaptive gearshift control method can effectively avoid the gearshift delay caused by the change in clutch characteristics, and the gearshift jerk in the simulation and experiment is reduced by 55.01% and 34.8%, respectively.

Output Power Leveling Control of WECS Based on Extended State Observer for All Operating Regions

2012 ◽  
Vol 263-266 ◽  
pp. 588-594
Author(s):  
Bo Yu Jiang ◽  
De Liang Zeng ◽  
Ji Zhen Liu
Keyword(s):  
Wind Speed ◽  
Output Power ◽  
Power Grid ◽  
State Observer ◽  
Extended State Observer ◽  
Linear Matrix ◽  
Reference Trajectory ◽  
Random Disturbance ◽  
Extended State ◽  
Grid Scheduling

This paper designs the pitch controller synthesis to ensure the stability and the controllability in the whole regions as the power grid scheduling instructions. Because of the wind speed measurement and accurate system model are hard to obtain, the range that the actual working point deviating from the balance point is very large, so that the system dynamic performance cannot be described with linear time variable (LTI). This paper utilizes the linear matrix inequalities' techniques (LMI) to design LPV (linear parameter varying) gain-scheduled controllers based on criteria, which provide guarantees of stability performances along the whole reference trajectory. The LPV pitch controller replaces the traditional speed controller to dynamically compensate the pitch angle. According to the errors between the power grid scheduling instructions and measuring the output power, power controller adjusts generator torque to achieve the adjustable power in the whole regions. In view of unmodeled dynamics and wind speed random disturbance, the feed forward linearization. robust pitch regulation using extended state observer (ESO) is proposed to restrain comprehensive disturbances. The proposed controller is seen to be more efficient.

Robust output feedback trajectory tracking for quadrotors

2018 ◽  
Vol 233 (5) ◽  
pp. 1596-1610 ◽  
Author(s):  
Xiling Shi ◽  
Yunqiang Sun ◽  
Xingling Shao
Keyword(s):  
Sliding Mode ◽  
State Observer ◽  
Lyapunov Theory ◽  
Extended State Observer ◽  
Reference Trajectory ◽  
Extended State ◽  
Control Laws ◽  
Virtual Control ◽  
Output Tracking Control ◽  
The Stability

This paper focuses on robust output tracking control for quadrotors exposed to parametric uncertainties and external disturbances. Based on the back-stepping control principle, the quadrotor dynamics is decomposed into translational and rotational subsystems. To handle the limitation of traditional extended state observer that can only be effective for integral-chain systems, a high-order extended state observer with special structure is developed to estimate the unmeasurable states and the lumped disturbances in rotational subsystem simultaneously. To avoid the tedious analysis and repeated differentiation of virtual control laws in the back-stepping technique, a first-order sliding mode differentiator is introduced to compute the derivative of virtual control law at each step in the presence of disturbances. The stability analysis is established using the Lyapunov theory. Simulation results demonstrate the effectiveness of the proposed control scheme in achieving a guaranteed tracking performance with respect to an 8-shaped reference trajectory.

scholarly journals Fault-tolerant control for a class of n-order systems based on fast terminal sliding mode and extended state observer

2021 ◽  
pp. 002029402110286
Author(s):  
Pu Yang ◽  
Peng Liu ◽  
ChenWan Wen ◽  
Huilin Geng
Keyword(s):  
Fault Tolerant ◽  
Sliding Mode ◽  
Singular Control ◽  
Fault Tolerant Control ◽  
State Observer ◽  
Extended State Observer ◽  
Extended State ◽  
Terminal Sliding Mode ◽  
Fast Terminal Sliding Mode ◽  
Terminal Sliding Surface

This paper focuses on fast terminal sliding mode fault-tolerant control for a class of n-order nonlinear systems. Firstly, when the actuator fault occurs, the extended state observer (ESO) is used to estimate the lumped uncertainty and its derivative of the system, so that the fault boundary is not needed to know. The convergence of ESO is proved theoretically. Secondly, a new type of fast terminal sliding surface is designed to achieve global fast convergence, non-singular control law and chattering reduction, and the Lyapunov stability criterion is used to prove that the system states converge to the origin of the sliding mode surface in finite time, which ensures the stability of the closed-loop system. Finally, the effectiveness and superiority of the proposed algorithm are verified by two simulation experiments of different order systems.

Sign in / Sign up

Export Citation Format

Share Document