Learning Tracking Control Research Articles

Event-triggered compound learning tracking control of autonomous surface vessels in the measurement network

This paper investigates the tracking control of the autonomous surface vessels (ASVs) with the time-varying disturbances. While the echo state network (ESN) accounts for the unknown dynamics in the model, a model-based event-triggered control (MBETC) scheme is presented by using the compound learning technique, which combines the learning of the ESN and the estimation of the compound disturbance. Different from the existing compound learning, the proposed scheme updates the estimates of the ESN weights and the compound disturbances in an event-triggered manner, in which two novel prediction errors are involved in their update laws. The values of prediction errors are obtained by using the online-recorded data during the inter-event time. To solve the recently proposed problem of “jumps of virtual control laws” arising in the backstepping-based event-triggered control (ETC), an event-triggered adaptive model is established to generate the continuous estimates of the states and direct the control laws. By the merit of ETC, the proposed scheme can importantly reduce the communication traffic in the measurement network compared with its continuous substitutes. By the merit of online-recorded-data-based compound learning, the proposed scheme can achieve the good understanding of synthetic uncertainties. All the errors in the closed-loop system are proved to be semi-globally uniformly ultimately bounded (SGUUB). Finally, a numerical example corroborates the proposed scheme.

Iterative Learning Tracking Control of Nonlinear Multiagent Systems with Input Saturation

A tracking control algorithm of nonlinear multiple agents with undirected communication is studied for each multiagent system affected by external interference and input saturation. A control design scheme combining iterative learning and adaptive control is proposed to perform parameter adaptive time-varying adjustment and prove the effectiveness of the control protocol by designing Lyapunov functions. Simulation results show that the high-precision tracking control problem of the nonlinear multiagent system based on adaptive iterative learning control can be well realized even when the input is saturated. Finally, the validity of the proposed algorithm is verified by numerical analysis.

Learning Tracking Control for Cyber–Physical Systems

This article investigates the problem of optimal tracking control for cyber-physical systems (CPSs) when the cyber realm is attacked by Denial-of-Service (DoS) attacks which can prevent the control signal transmitting to the actuator. Attention is focused on how to design the optimal tracking control scheme without using the system dynamics and analyze the impact of DoS attacks on tracking performance. First, a Riccati equation for the augmented system, including the system model and the reference model is derived under the framework of dynamic programming. The existence and uniqueness of its solution are proved. Second, the impact of the successful DoS attack probability on tracking performance is analyzed. A critical value of the probability is given, beyond which the solution to the Riccati equation cannot converge. The tracking controller cannot be designed. Third, reinforcement learning is introduced to design the optimal tracking control schemes, in which the system dynamics are not necessary to be known. Finally, both a dc motor and an F16 aircraft are used to evaluate the proposed control schemes in this article.

Preliminary Evaluation of Composite Learning Tracking Control on 7-DoF Collaborative Robots

Abstract With the increase of computational power, adaptive control becomes more and more attractive for industrial robots due to the ability to handle parametric uncertainty and the potential to achieve higher tracking accuracy. This paper implements a recently developed adaptive control scheme termed composite learning robot control (CLRC) for tracking control of a 7-degree-of-freedom industrial collaborative robot named Franka Emika Panda. The desired trajectory is specifically designed to evaluate the performance of the CLRC compared with the standard adaptive robot control and composite adaptive robot control via both simulations and experiments. Comparative results demonstrate that the CLRC meets the requirement of real-time computation and outperforms the other two control schemes with respect to tracking accuracy and parameter convergence.

Compound learning tracking control of a switched fully-submerged hydrofoil craft

This paper addresses the tracking control of a fully-submerged hydrofoil craft (FSHC) on the horizontal plane. The motions of a FSHC can be classified into three cruising conditions, namely the hulborne, the taking off, and the foilborne. Considering that the FSHC has the different model dynamics and disturbances between the three cruising conditions, we treat the FSHC model as a switched system between the hullborne and the foliborne conditions, in which the taking off condition is conveniently omitted and deemed as the fast transient. The compound learning technique is introduced to offset the uncertainties, and it is comprised of two parts. First, the fuzzy logic system (FLS) is employed to approximate the unknown model dynamics in each degree of freedom. Based on a serial-parallel estimation model (SPEM), the composite adaptive laws are fabricated to estimate the weights of the FLS by combining the tracking errors with the prediction errors between the estimation model and the dynamic loop of the FSHC. Second, the disturbance observer (DOB) is fabricated to estimate the compound disturbance, which is comprised of the approximation error of the FLS and the disturbance. Based on the compound learning, we develop the switched adaptive backstepping control laws of the FSHC. By employing the average dwell-time method, all the tracking and the estimation errors in the switched system are proved to be semi-globally uniformly ultimately bounded (SGUUB). Finally, the numerical experiment is conducted to verify the effectiveness of the proposed scheme.

Iterative Learning Model-Free Control for Networked Systems With Dual-Direction Data Dropouts and Actuator Faults.

In this article, we study the tracking problem for networked nonlinear discrete systems with actuator faults and dual-direction data dropouts. A novel adaptive fault-tolerant iterative learning model-free control strategy is designed. First, by utilizing the method called compact form dynamic linearization, the original nonlinear system model is transformed into an equivalent data-driven model, and the data model contains only one unknown parameter. Both the actuator fault and the system dynamics information are included in this parameter. Then, to model the physical processes of data dropout, a new mathematical relationship is constructed. Furthermore, an adaptive fault-tolerant iterative learning tracking control scheme is developed with only randomly received input/output data. Noting that the high learning rate or convergence rate is required in actual applications, a new varying parameter approach is designed to improve such rate. Finally, it is rigorously proved that the closed loop is stable in the sense of uniform ultimate boundedness, and numerical simulation results are conducted to validate the effectiveness of the designed control strategy.

Reinforcement Learning Tracking Control for Robotic Manipulator With Kernel-Based Dynamic Model.

Reinforcement learning (RL) is an efficient learning approach to solving control problems for a robot by interacting with the environment to acquire the optimal control policy. However, there are many challenges for RL to execute continuous control tasks. In this article, without the need to know and learn the dynamic model of a robotic manipulator, a kernel-based dynamic model for RL is proposed. In addition, a new tuple is formed through kernel function sampling to describe a robotic RL control problem. In this algorithm, a reward function is defined according to the features of tracking control in order to speed up the learning process, and then an RL tracking controller with a kernel-based transition dynamic model is proposed. Finally, a critic system is presented to evaluate the policy whether it is good or bad to the RL control tasks. The simulation results illustrate that the proposed method can fulfill the robotic tracking tasks effectively and achieve similar and even better tracking performance with much smaller inputs of force/torque compared with other learning algorithms, demonstrating the effectiveness and efficiency of the proposed RL algorithm.

Anti-disturbance iterative learning tracking control for space manipulators with repetitive reference trajectory

Purpose This paper is concerned with the repetitive trajectory tracking control for space manipulators under model uncertainties and vibration disturbances. Design/methodology/approach The model uncertainties and link vibration of manipulators will degrade the tracking performance of space manipulators; in this paper, a new hybrid control scheme that consists of a composite hierarchical anti-disturbance controller and an iterative learning controller is developed to solve this problem. Findings The composite hierarchical controller can effectively attenuate model uncertainties and reject vibration disturbances, whereas the iterative learning controller is able to improve the tracking accuracy for repetitive reference trajectory. Originality/value The proposed scheme compensates for the shortcomings of iterative learning control which can only deal with repetitive disturbances, ensuring the accuracy and repeatability of space manipulators under model uncertainties and random disturbances.

Iterative learning tracking control for a class of MIMO nonlinear time-varying systems

Tracking control nonlinear systems with model-less controllers have attracted extensive attention in the research community. One frequently applied model-less controller technique is iterative learning tracking control. However, this technique has mostly been applied for controlling time-invariant systems, where the dynamic behaviour of the system would not change over time. In this article, the authors demonstrate a variable gain iterative learning control (VGILC) technique that is capable of continuously adjusting and effectively controlling multi-input and multi-output (MIMO) time-varying nonlinear systems. VGILC incorporates variable gain into the PD-type updating law to accelerate convergence speed. The convergence condition of the new learning law is presented and proved. The proposed approach improves the output tracking performance within a few trials. The effectiveness of the presented approach is demonstrated using modified robotic manipulators systems and its validity is proven.

Forgetting Factor Nonlinear Functional Analysis for Iterative Learning System with Time-Varying Disturbances and Unknown Uncertain

This paper focuses on the iterative learning tracking control problem for a class of nonlinear system with time-varying disturbances. First, because of the mismatches in time-varying disturbances functions, a high-order feed-forward iterative learning control (ILC) is employed to change the original system into an iterative system. Secondly, a variable forgetting factor is developed to stabilize the system. Based on the feed-forward iterative learning controller, a memory controller is constructed for the nonlinear system. By choosing a new variable forgetting factor, we show that the designed continuous adaptive controller makes the solutions of the closed-loop system convergent to a ball exponentially. Finally, a numerical example is given to show the feasibility and effectiveness of the proposed method.

Iterative learning tracking control for a class of MIMO nonlinear time-varying systems

Iterative learning tracking control for a class of MIMO nonlinear time-varying systems

Iterative Learning Tracking Control of a Hybrid-Driven Based Three-Cable Parallel Manipulator

The aim of this work was to present iterative learning tracking control of a hybrid-driven based three-cable parallel manipulator (HDCPM). The HDCPM has the advantages of both cable parallel manipulator and hybrid-driven planar five-bar mechanism. Design of the HDCPM is described and dynamic model is developed on the basis of Lagrange method. Consider the HDCPM system with nonlinear, time-varying characteristics and repetitive unknown disturbances; an adaptive iterative learning control strategy, which can control gains change with the iterations, is designed. By means of Lyapunov function, the stability of the controller is proved. Finally, simulation results indicate that a perfect trajectory tracking of the HDCPM is achieved by the adaptive iterative learning controller, which also illustrate the correctness of the dynamic modeling and effectiveness of the proposed control strategy.

Application of Online Iterative Learning Tracking Control for Quadrotor UAVs

Quadrotor unmanned aerial vehicles (UAVs) have attracted considerable interest for various applications including search and rescue, environmental monitoring, and surveillance because of their agilities and small sizes. This paper proposes trajectory tracking control of UAVs utilizing online iterative learning control (ILC) methods that are known to be powerful for tasks performed repeatedly. PD online ILC and switching gain PD online ILC are used to perform a variety of manoeuvring such as take-off, smooth translation, and various circular trajectory motions in two and three dimensions. Simulation results prove the ability and effectiveness of the online ILCs to perform successfully certain missions in the presence of disturbances and uncertainties. It also demonstrates that the switching gain PD ILC is much effective than the PD online ILC in terms of fast convergence rates and smaller tracking errors.

Adaptive learning tracking control of robotic manipulators with uncertainties

An adaptive learning tracking control scheme is developed for robotic manipulators by a synthesis of adaptive control and learning control approaches. The proposed controller possesses both adaptive and learning properties and thereby is able to handle robotic systems with both time-varying periodic uncertainties and time invariant parameters. Theoretical proofs are established to show that proposed controllers ensure asymptotical tracking performance. The effectiveness of the proposed approaches is validated through extensive numerical simulation results.