Adaptive Dynamic Control Research Articles

Learning-Based Neural Dynamic Surface Predictive Control for MMC

Reinforcement learning technique was developed recently as an interesting topic in designing adaptive optimal controllers. This technique explicitly provided a feasible solution to circumvent the “curse of dimensionality” and requiring a system model inherent in the classical dynamic programming algorithm. By virtue of this property, in our work, by introducing this technique into a predictor-based online adaptive neural dynamic surface predictive control architecture, we concentrate on a novel robust predictive control framework subject to system uncertainties. To be specific, in this presented framework, an adaptive dynamic programming control strategy utilizing a critic neural network point of view is developed to learn the optimal control policy. Our modification is able to facilitate the alleviation of performance deterioration caused by system uncertainties and enable the smooth and fast learning, while keeping the merits of the finite control-set model predictive control. Finally, the interest and applicability of the proposed control methodology are verified by performance evaluation.

Dynamic Event-Triggered Communication-Based Adaptive Consensus Tracking of Uncertain Nonlinear Multiagent Systems in Pure-Feedback Form

This study is aimed at investigating a dynamic event-triggered communication-based adaptive distributed consensus control problem for a class of uncertain pure-feedback nonlinear multiagent systems with limited communication resources. The nonaffine nonlinear functions of multiagent systems are assumed to be unknown and heterogeneous, and intermittent inter-agent communication is considered to occur within a directed network. A novel adaptive distributed dynamic surface control strategy using neural networks is developed to manage non-differentiable virtual control laws associated with the intermittent communication between agents with uncertain nonaffine nonlinear parts. The key contribution of this research is the derivation of dynamic event-triggering conditions using distributed tracking errors to efficiently adjust inter-event times, while ensuring the consensus tracking performance and robustness against unexpected external disturbances. Compared with the existing static event-triggered communication approach, the proposed controller can alleviate the communication burden among agents. Using technical lemmas, it is shown that all signals of the considered system are semiglobally uniformly ultimately bounded, and Zeno behavior is strictly excluded. Comparative simulation results illustrate the effectiveness of the proposed dynamic event-triggered control approach.

Dynamic Event-triggered Adaptive Control for Uncertain Nonlinear Switched Systems Based on MDADT Method

Dynamic Event-triggered Adaptive Control for Uncertain Nonlinear Switched Systems Based on MDADT Method

Adaptive Bearing-Only Control of Multiple Euler-Lagrange Systems for Static Geometric Formation⋆

This paper studies the formation control problem of multi-agent systems each of which the dynamics are modeled by the Euler-Lagrange equation with unknown parameters. The objective of each agent is to develop a control law using the bearing information with respect to its neighbors, such that the multi-agent system forms a static geometric formation defined by the bearing constraints among agents. An adaptive dynamic control law with a leader-following type is proposed such that the objective can be achieved in the presence of unknown systems’ parameters. Finally, simulation examples are presented to illustrate the effectiveness and feasibility of the main results.

Dynamic event-triggered adaptive control for uncertain stochastic nonlinear systems

In this paper, the dynamic event-triggered adaptive tracking control problem is investigated via backstepping technology for uncertain stochastic nonlinear systems. First, the stochastic nonlinear system with unknown parameter is considered. By introducing an additional dynamic variable, a dynamic event-triggered adaptive controller is designed such that the closed-loop signals are uniformly ultimately bounded in the sense of the fourth moment. Then, a more general partially unknown stochastic nonlinear system is further considered, and the designed adaptive neural network control scheme ensures that the closed-loop signals are fourth moment semi-globally uniformly ultimately bounded (SGUUB). The proposed dynamic event-triggering mechanism (DETM) guarantees that the lengths of time intervals between each two consecutive events are lower-bounded by a positive constant. It is necessary to point out that the DETM is better at saving resources than the static event-triggering mechanism (SETM). Finally, two simulations are conducted to show the validity of the control strategies for these two systems, respectively.

Event-triggered-based adaptive asymptotic tracking control for stochastic nonlinear time-delay systems with actuator and sensor faults

The issue of dynamic event-triggered adaptive asymptotic tracking control for stochastic nonlinear time-delay systems with actuator and sensor faults is investigated in this article. Utilizing fuzzy logic systems and projection operator, unknown time delays are calculated. Compared with the existing methods to overcome the effect of time delay, the estimation of the time delay information is considered to avoid the controller singularity problem. Moreover, to reckon the upper bound of fuzzy logic system approximation errors and other residuals, an error adaptive law is developed in the controller construction. With the assistance of stochastic Barbalat’s lemma and adaptive fuzzy control strategy, a novel asymptotic tracking fuzzy controller is proposed under multiple faults, which guarantees that the whole signals are bounded in probability. Besides, a dynamic event triggering control (DETC) is employed to save the communication resources. A practical example is supplied to demonstrate the effectiveness of the proposed control approach.

Adaptive event-triggered dynamic distributed control of switched positive systems with switching faults

This paper designs the dynamic output-feedback controller of switched positive systems subject to switching faults using an improved adaptive event-triggering mechanism. An adaptive event-triggering condition is addressed in the form of 1-norm by virtue of the measurable outputs of distributed sensors and the corresponding error. An error-based closed-loop control system whose dynamic variable relies on a state observer is obtained. A multiple copositive Lyapunov function is constructed to deal with the positivity and stability of the systems. The matrix decomposition and linear programming approaches are used to design and compute the controller and observer gains. An improved average dwell time scheme is proposed to handle the switching faults. The contributions of this paper lie in that: (i) An adaptive event-triggering mechanism is established for switched positive systems, (ii) A framework on the fault of switching signal is constructed, and (iii) A dynamic distributed controller is proposed for the considered systems. Finally, two illustrative examples are given to verify the effectiveness of the obtained results.

Finite-time adaptive dynamic surface synchronization control for dual-motor servo systems with backlash and time-varying uncertainties

The dual-motor driving servo system is continuously developed to satisfy strict safety and reliability requirements. However, several factors may degrade the system’s performance, such as transmission backlash, parameter drift, and motor dynamic characteristic differences. To overcome these factors, this study proposes a finite-time tracking and synchronization control method for dual-motor servo systems that suffer from backlash and time-varying uncertainties. Our solution utilizes an adaptive dynamic surface and cross-coupling control scheme to deal with tracking and synchronization control issues and compensate for the unknown time-varying uncertainties. Through synchronizing the speed and acceleration states, the proposed controller guarantees high control performance and eliminates the force fighting caused by the motor’s dynamic characteristic differences. In addition, finite-time control ensures the tracking error converges to an arbitrarily small neighborhood of zero in finite time. Moreover, the singularity problem in the derivative of the virtual control signal is avoided by introducing a new compensation term. Several simulations prove the proposed controller’s stability and effectiveness.

On the positive and negative damping effect of piezoelectric structure using switched-mode bidirectional energy conversion circuit

The introduction of a switched-mode bidirectional energy conversion circuit (BECC) facilitates the development of piezoelectric devices toward integrated and energy-efficient multi-functional designs. These new designs realize the organic combinations of two or more functions among energy harvesting (EH), vibration excitation (VE), and dynamic sensing. Yet, the structural effect after applying the BECC was not comprehensively investigated and strictly quantified. This letter analyzes the electromechanical joint dynamics of a piezoelectric structure using the BECC, which was developed after the synchronized triple bias-flip (S3BF) technique, under different EH and VE operation modes. It shows that the EH modes electrically induce an extra positive damping effect, while the VE modes induce a negative one. A lumped model and impedance analysis are used to evaluate the electrically induced damping and the energy conversion efficiency by the BECC. The closed-form expression of the vibration displacement under different operation modes is derived. Experiments are carried out under different operation modes and frequencies. The theoretical and experimental results show good agreement. They validate the damping tuning capability of BECC in either positive or negative damping directions. This switched-mode interface circuit offers a promising solution for building an adaptive dynamic control of piezoelectric structures with high energy efficiency.

A trajectory and force dual-incremental robot skill learning and generalization framework using improved dynamical movement primitives and adaptive neural network control

Due to changes in the environment and errors that occurred during skill initialization, the robot's operational skills should be modified to adapt to new tasks. As such, skills learned by the methods with fixed features, such as the classical Dynamical Movement Primitive (DMP), are difficult to use when the using cases are significantly different from the demonstrations. In this work, we propose an incremental robot skill learning and generalization framework including an incremental DMP (IDMP) for robot trajectory learning and an adaptive neural network (NN) control method, which are incrementally updated to enable robots to adapt to new cases. IDMP uses multi-mapping feature vectors to rebuild the forcing function of DMP, which are extended based on the original feature vector. In order to maintain the original skills and represent skill changes in a new task, the new feature vector consists of three parts with different usages. Therefore, the trajectories are gradually changed by expanding the feature and weight vectors, and all transition states are also easily recovered. Then, an adaptive NN controller with performance constraints is proposed to compensate dynamics errors and changed trajectories after using the IDMP. The new controller is also incrementally updated and can accumulate and reuse the learned knowledge to improve the learning efficiency. Compared with other methods, the proposed framework achieves higher tracking accuracy, realizes incremental skill learning and modification, achieves multiple stylistic skills, and is used for obstacle avoidance with different heights, which are verified in three comparative experiments.

Fixed-Time Consensus Tracking of Multiagent System Under DOS Attack With Event-Triggered Mechanism

This paper focuses on fixed-time event-triggered consensus tracking of high-order MAS over digraph and under DOS attack. The considered DOS attack is divided into connectivity-maintained attack and connectivity-broken attack. To detect connectivity-broken attack under event-triggered environment, a new cyberattack detection algorithm is presented. Next, using cyberattack detection results, a fixed-time event-triggered distributed observer is developed to estimate the leader’s states in the presence of DOS attack, which conducts event-triggered update in the normal condition and under connectivity-maintained attack and stops updating in the presence of connectivity-broken attack. Then, an adaptive event-triggered prescribed-time dynamic surface consensus tracking control protocol is proposed to eliminate mismatched disturbances caused by estimation error and achieve prescribed-time convergence of the tracking error. Stability analysis proves fixed-time convergence of the developed consensus tracking strategy and non-zeno analysis shows the developed distributed observer and consensus tracking control strategy exclude zeno behavior. This paper contributes to proposing a fixed-time leader-follower consensus control strategy for MAS under DOS attack via event-triggered approach, which eliminates the effect of DOS attack, achieves consensus tracking within a fixed time and reduces resources consumption. Finally, simulation results show the effectiveness and superiority of the developed consensus tracking approach.

Modified adaptive discrete-time incremental nonlinear dynamic inversion control for quad-rotors in the presence of motor faults

Unmanned air vehicles are intrinsically non-linear, unstable, uncertain, and prone to a variety of faults, most commonly the motor faults. The main objective of this paper is to develop a fault-tolerant control algorithm for the quadrotors with the motor faults. Accordingly, a novel adaptive modified incremental nonlinear dynamic inversion (MINDI) control is proposed to stabilize and control the quad-rotor with partial motor faults. The controller consists of a MINDI controller augmented with a discrete-time nonlinear adaptive algorithm. Since the incremental nonlinear dynamic inversion (INDI) algorithm is essentially based on the sensor measurements, it necessitates the angular rates differentiation and therefore amplifies the high-frequency noises produced by the gyroscopes. The application of derivative filters causes unavoidable internal state delays in the INDI structure. Henceforth, the performance of the controller developed for the unstable and uncertain quadrotors degrades considerably. To address this drawback, this paper proposes the MINDI controller which basically derives the angular accelerations from the angular moment estimations. Furthermore, to increase the robustness of the MINDI against motor faults, a discrete-time adaptive controller has been incorporated. The performance of the proposed controllers is verified both through the nonlinear simulations and testbed experiments. The results are compared with a recent efficient algorithm, which had been implemented on a quad-rotor model.

Control Strategy of Doubly-Fed Induction Generator under Zero Voltage Fault of Power Grid

For improving the zero-voltage ride through the capability of a doubly fed induction generator in high proportion new energy grid in extreme faults, a coordinated control scheme of hardware and optimal control strategy is proposed. A high-temperature superconductive-fault current limiter suppresses stator fault current, adaptive virtual impedance control and active dynamic reactive power support control act on the back-to-back converter of wind turbines as optimal control strategies. Optimizing the control strategy without changing the controller structure is beneficial to engineering implementation. After mathematical derivation and simulation verification, the coordinated control strategy adopted in this paper can effectively avoid the rotor current and voltage exceeding the limit when the wind turbine is facing extreme faults, actively provide reactive power support for the busbar, realize zero voltage ride through and reduce the risk of high voltage failure at the point of failure. The control effect is obviously better than the traditional virtual impedance control.

Dynamic Event-Based Adaptive Finite-Time Tracking Control for Nonlinear Stochastic Systems Under State Constraints

This article focuses on the problem of adaptive finite-time tracking control for nonlinear stochastic systems under asymmetric constraints based on dynamic event-triggering control. Different from the existing works, a novel adaptive tracking control algorithm is proposed with asymmetric time-varying constraints and dynamic event-triggering mechanism. First, to constrain the state variable within given time-varying boundaries, a novel predefined-time performance function is constructed. Second, a novel barrier function related to state variable is constructed, by means of which the state variable is directly constrained within the asymmetric time-varying boundaries without the virtual controller. In addition, by establishing a novel dynamic function, we propose a dynamic event-triggering mechanism, and then design controller accordingly, which can reduce computation burdens and save the network resources. By the aid of the Lyapunov stability theory, it is proved that the system tracking error converges to an adjustable bounded set in probability in a finite time and all state variables are successfully constrained into the asymmetric time-varying boundaries. Finally, the effectiveness of the proposed control algorithm is verified by a simulation example.

Dynamic event-triggered adaptive output feedback control of uncertain nonlinear systems with intermittent observations

In this paper, the output feedback adaptive control problem of uncertain nonlinear system with event-sampled states is studied. Firstly, a novel intermittent-state observer is designed to estimate the unmeasurable states. Secondly, the nature of event-triggered makes the state signals discontinuous, rendering the traditional backstepping technique inapplicable as the derivatives of virtual control signals do not exist. To overcome this non-trivial obstacle, we consider the case without state-triggering in the previous n−1 steps of Lyapunov stability analysis, and reconstruct the actual controller by utilizing the triggered states signals in the last step. It can be demonstrated that all signals of the resulting closed-loop system are semi-globally uniformly ultimately bounded (SGUUB) under the derived event-triggered laws and the system performance can be improved by tuning design parameters properly. Simulation examples are shown to validate the efficacy of the proposed control scheme.

Sliding Mode Disturbance Observer-Based Adaptive Dynamic Inversion Fault-Tolerant Control for Fixed-Wing UAV

Unmanned aerial vehicles (UAVs) have been widely applied over the past decades, especially in the military field. Due to the unpredictability of the flight environment and failures, higher requirements are placed on the design of the control system of the fixed-wing UAV. In this study, a sliding mode disturbance observer-based (SMDO) adaptive dynamic inversion fault-tolerant controller was designed, which includes an outer-loop sliding mode observer-based disturbance suppression dynamic inversion controller and an inner-loop real-time aerodynamic identification-based adaptive fault-tolerant dynamic inversion controller. The sliding mode disturbance observer in the outer-loop controller was designed based on the second-order super-twisting algorithm to alleviate chattering. The aerodynamic identification in the inner-loop controller adopts the recursive least squares algorithm to update the aerodynamic model of the UAV online, thereby realizing the fault-tolerant control for the control surface damage. The effectiveness of the proposed SMDO enhanced adaptive fault-tolerant control method was validated by mathematical simulation.

Improved Neural Adaptive Control for Nonlinear Oscillatory Dynamic of Flapping Wings

Modeling uncertainties and oscillatory dynamics are control challenges for flapping vehicles. Flapping-wing aerial vehicles are nonlinear time-varying oscillatory systems with flexible multibody dynamics. As such, accurate modeling of these complex systems considering unsteady aerodynamics is a formidable task. Though adaptive controllers can work well with uncertain approximate models, inherent oscillation of flapping systems can degrade their control performance and create undesired control actuations. This paper addresses improvements of neural adaptive dynamic inversion controller toward efficient performance for flapping flight via effective utility of actuator capacity. In this respect, first an assumptive model of a bird-mimetic flapping-wing is considered for simulation and analysis. It is shown that the usage of oscillatory feedback data produces additional error in trajectory tracking and unnecessary oscillatory control actuations. Subsequently, the control loop is modified by adding an adaptive notch filter for real-time estimation, tracking, and removal of dominant oscillatory modes from the feedback data and consequently from the controller output. As a result, the required control effort is effectively reduced and the controller performance is significantly improved. Simulations are provided to demonstrate the efficiency of the proposed scheme in presence of noise, variation of system frequency, disturbances, as well as delay in the control loop.

New Adaptive Dynamic Output Feedback Control of Double-Pendulum Ship-Mounted Cranes With Accurate Gravitational Compensation and Constrained Inputs

Output feedback control without using velocities is an effective way of dealing with problems caused by unmeasurable velocities and noise magnification by numerical difference, which are practical issues associated with crane control. This article proposes a new adaptive dynamic output feedback control approach for double-pendulum ship-mounted cranes in the presence of unknown parameters (e.g., cargo masses). The problems caused by the unknown masses are addressed through a new adaptive gravitational compensation law. Also, the input constraints are tackled by introducing bounded functions. Specifically, to handle the complicated disturbances caused by ship motions, new state variables are constructed to transform the dynamic model. Then, Lyapunov-based stability analysis is proceeded to design the controller and the adaptive law. To the best of authors’ knowledge, this is the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"/> first <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"/> control method proposed for double-pendulum ship-mounted crane systems <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"/> without <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"/> using velocity signals, which can simultaneously eliminate steady errors, suppress unexpected cargo swing, achieve accurate gravitational compensation, and satisfy input constraints. The closed-loop system’s state variables asymptotically converge to the equilibrium point, which is rigorously proven. A series of experiments are implemented on a hardware testbed, which further illustrate the satisfactory performance of the proposed method.

Finite-Time Adaptive Extended State Observer-Based Dynamic Sliding Mode Control for Hybrid Robots

This brief investigates the finite-time adaptive extended state observer (ESO) based dynamic sliding mode control (SMC) for the hybrid robot subject to uncertain problem, e.g., modeling error, joint friction, external disturbance and so on. A barrier function-based adaptive law is defined to adjust the switching gain of SMC system with unknown upper bound of the disturbance. Such an adaptive law-based control method ensures the finite-time convergence of the sliding variable to an adjustable vicinity of zero and realizes chattering reduction of SMC. Then, for estimating and compensating system uncertainties, the time-varying gain ESO (TESO) is designed without the condition of the disturbance change rate tending to zero. A system error-based adaptive law is designed to adjust the observer bandwidth gain, which can reduce the peaking value. Such a TESO reduces the burden of SMC and further weakens the chattering of SMC, so as to improve the control system robustness. The finite-time Lyapunov stability of the sliding variable and the boundness of the TESO error have been proved theoretically. The practical effectiveness is illustrated in simulations and experiments with the prototype of hybrid robot.

L1 Adaptive Structure-Based Nonlinear Dynamic Inversion Control for Aircraft with Center of Gravity Variations

L1 Adaptive Structure-Based Nonlinear Dynamic Inversion Control for Aircraft with Center of Gravity Variations