Search

This paper studies the adaptive neural networks (NNs) tracking control problem for a class of mobile robot systems with full-state constraints. First, to compensate for the adverse effects of the unknown dead-zone input, which is ubiquitous in mobile robot motors, a new robust control algorithm is put forward by the use of adaptive control technique. Then, a new unified barrier function (UBF) is constructed to deal with the problem of state constraints. Different from traditional barrier Lyapunov function (BLF) methods, which can only constrain the error of the system state and virtual controller, our proposed control method can constrain the system state directly by introducing a novel nonlinear transformation function and a new coordinate transformation. It's worth noting that the UBF can be utilized to deal with both constrained and unconstrained cases by resizing parameters without changing the control structure. Furthermore, adaptive NNs are introduced to approximate the uncertainty of the system. Finally, based on the Lyapunov stability theory, it is proved that all signals in the closed-loop system are ultimately bounded and the full-state constraints are never violated. The effectiveness of the control method is verified on simulation examples and a mobile robot experimental platform. [ABSTRACT FROM AUTHOR]

In this paper, the problem of finite-time trajectory tracking control is studied and addressed for a 6 degree of freedom (DOF) autonomous underwater vehicle (AUV) subjected to unknown dynamic model, stochastic perturbations, external disturbances (matched and mismatched) and saturation input nonlinearities. Based on the backstepping control approach, novel finite-time control inputs are designed and proposed. Artificial neural networks (ANNs) and finite-time adaptation laws are exploited to approximate the nonlinear dynamics of AUV, the stochastic perturbations and the upper bound of external disturbances. To handle the destructive effects of saturation input nonlinearities, finite-time auxiliary system method is utilized. To overcome the explosion of complexity problem of backstepping control strategy, compensator-based finite-time command filter approach is exploited. By utilizing the Lyapunov stability theorem, it is mathematically proven and demonstrated that the suggested nonlinear control inputs are able to guarantee the semi-global finite-time stability in probability (SGFSP) of the closed-loop AUV system. Finally, numerical simulations are carried out to illustrate and depict the effectiveness and performance of the proposed neuro-adaptive robust finite-time control scheme. [ABSTRACT FROM AUTHOR]

In this paper, for the safety–critical systems with unmatched disturbances, a safe optimal robust control method based on neural network is proposed to ensure that the safety–critical system operates within its safe region and learns the optimal control strategy. The cost function which is the goal of the designers is augmented by a control barrier function (CBF) to achieve both safety and optimality. This method does not directly regard security as a constraint on the system state, but influences the cost function through a security penalty mechanism. An additional function is used to approximate the effect of the unmatched disturbances on the safety–critical systems. On the premise of satisfying the security and robustness, the neural network approximation method is used to learn the optimal control strategy. Based on Lyapunov stability theory, it is shown that the neural network-based safe optimal robust controller can guarantee all the signals of the resulting closed-loop systems to be uniformly ultimately bounded. Finally, two simulation examples are given to demonstrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

In this paper, to resist stochastic cyber attacks and suppress the influence of the exogenous disturbance in synchronization of multiplex neural networks, an adaptive quantization control strategy is proposed. To overcome uncertainties and represent constraints in communication, an adaptive quantitative controller is designed, which is used to arrive outer synchronized behavior with an H ∞ performance between the multiplex neural networks under stochastic cyber attacks and exogenous disturbance. The node of the multiplex networks is neural networks which is named neural sub-networks. It is assumed that the neural sub-networks in the multiplex networks with delay topology are coupled by nonlinear functions. Based on Lyapunov stability method, the conditions for the multiplex networks to achieve synchronization with an H ∞ performance are presented. Finally, numerical example illustrates the effective of the theoretical framework. [ABSTRACT FROM AUTHOR]

In this article, the event-triggered near-optimal control issue is studied for the input-constrained uncertain system with the input-to-state stability (ISS) attribute. In the proposed approach, we design a near-optimal controller and an adaptive event-triggered mechanism (ETM) to optimize the cost function and reduce the redundant computational consumption. Toward this end, the robust control problem of the uncertain system is first converted into the near-optimal control issue for the nominal system via the designed cost function. Then, for reducing the redundant computational consumption for uncertain systems, the adaptive ETM is distinguishingly devised for the matched uncertainty. Subsequently, a sufficient condition is offered to ensure the asymptotical stability for the discrete-time uncertain system with the ISS attribute under the event-triggered controller. To further improve the control performance for the closed-loop systems, the goal representation adaptive critic design approach is proposed in the event-triggered context with non-periodic weight updating rules for neural networks are designed for the controlled system. Moreover, the corresponding convergence of the proposed control approach is considered. Finally, the effectiveness of the suggested approach is demonstrated by two simulation cases. [ABSTRACT FROM AUTHOR]

This study processes the adaptive robust control problem in a task space for a mobile manipulator with uncertain dynamics and external disturbances based on radial basis function neural networks(RBFNN) and a nonlinear state observer. Owing to the high nonlinearity, strong coupling, and unknown uncertainty characteristics of the mobile manipulators, their control poses a considerable challenge. An adaptive robust control strategy for nonlinear mobile robot systems with unknown uncertainties and disturbances based on the RBFNN and state observer is proposed in the operating space. First, a feedforward-feedback virtual speed generator is developed based on the feedforward control and backstepping method to implement virtual speed tracking control in the operation space to make the positional error asymptotically stable. Then, a model-independent RBFNN based adaptive robust controller (ARBFNNC) with the state observer is proposed, which converts the virtual speed input into the control torque of the actual mobile manipulator to achieve precise position tracking in the task space, and theoretical analysis performed through Lyapunov stability theory shows the global asymptotic stability of a system under the control of the proposed method. Finally, simulation results confirm the effectiveness of the proposed control method in position regulation and tracking control of the nonlinear mobile manipulator. [ABSTRACT FROM AUTHOR]

In this paper, distributed state and output feedback control law based on the dynamic event-triggered mechanism are proposed to solve the robust cooperative output regulation problem for a class of general linear uncertain multi-agent systems (MASs) subject to external disturbances. The discrete-time control law proposed for each agent is equivalent to a discrete one and thus directly implementable in the digital platform. Our design can handle parametric uncertainties, reject external disturbances and track a dynamic leader while excluding the Zeno phenomenon by providing the existence of a lower bound for inter-event times of each agent, the efficiency of which is illustrated by a numerical example. [ABSTRACT FROM AUTHOR]

This paper proposes a neural network-based robust control approach for driving disturbed Euler–Lagrange systems (e.g., robot manipulator) in a directed network to perform consensus tracking of a heterogeneous leader's output signal. The parameter matrices and external disturbances in each follower's Euler–Lagrange dynamics are unmeasurable, thus we unify them into one lumped uncertainty term. Neural networks are then employed to online estimate these uncertainties, with adaptive update laws ensuring the boundedness of the estimation errors. Subsequently, a set of distributed observers are developed to adaptively estimate the leader's states as well as the unidentified parameter vector and output matrix in the leader's model. With their help, the robust cooperative controller is designed. Its efficacy on directed networks is theoretically justified by designing a Lyapunov function with a special structure. This function produces symmetric positive definite matrices when its first derivative terms interact with the specified parameters in the observer and Laplacian matrices for directed graphs. Finally, numerical simulations based on two-link robot manipulators are carried out to verify that the tracking errors converge to zero when applying the neural network-based robust controller. [ABSTRACT FROM AUTHOR]

Although numerous approximate dynamic programming (ADP) methods have been proposed and applied to robust control problems, research on the use of ADP methods for robust tracking control in nonlinear systems remains limited, especially for discrete time (DT) nonlinear systems. In this paper, we propose an efficient robust learning-based output tracking control method, referred to as r-LTC, for a class of unknown DT nonlinear systems with dynamic uncertainty. The objectives are achieved by: 1) using a system identification technique based on network-type coefficients Auto-Regressive with eXogenous variables (ARX) model to identify completely unknown systems; 2) reformulating the identified model into a state space model in terms of deviations from the reference trajectory, such that the robust tracking problem can be converted into a robust regulation problem; 3) transforming the robust regulation problem into an equivalent optimal regulation problem for a nominal system with a modified utility function; 4) alternately implementing the performance evaluation and control policy update through two neural networks (NNs) to numerically solve DT Hamilton-Jacobi-Bellman (HJB) equation, such that the approximated optimal feedback control inputs for nominal system can be obtained, allowing for realization of all unknown bounded uncertainties. The proposed method does not require a state observer and any steady state knowledge, only the measurable system input and output data are utilized. The convergence of NNs' approximate weights and stability of the closed-loop uncertain systems are rigorously proven using the Lyapunov method. Lastly, a case study is performed using real data collected from the Shenzhen Metro Line 12 tunneling project to examine the performance of the proposed r-LTC algorithm in terms of the set value tracking accuracy, control system stability, and robustness against different types of disturbances. [ABSTRACT FROM AUTHOR]

In view of unknown Bilateral Teleoperators, the simultaneous satisfaction of not only synchronous control problem between two sides, but also optimal control performance under time-varying delays and external disturbance is a difficult task. Initially, in order to address this challenge, we proposed two control approaches including On-Policy and Off-Policy strategies after employing the sliding variable to reduce the order of dynamic model. Additionally, since the development of the unification between synchronous control problem and optimal control performance, it requires the consideration of discount factor addition to guarantee the bound of the infinite horizon cost function. In order to handle the dynamic uncertainties and the exponential cost function, by fully considering data collection technique in model-free On-Policy and Off-Policy strategies, Reinforcement Learning control schemes are proposed to guarantee the optimality effectiveness. Consequently, the Robust Integral of the Sign of the Error term is integrated into two proposed optimal control frames to improve the synchronous control performance and estimation capability of dynamic uncertainties. Moreover, the convergence of control policies and synchronous control problem of two proposed control frames are rigorously analyzed by Lyapunov stability theory. Finally, simulation results and the comparisons with the existing controller demonstrate the effectiveness of two proposed control frameworks. [ABSTRACT FROM AUTHOR]

This article investigates the robust control problem of affine nonlinear systems with both additive and multiplicative uncertainty. Different from existing actor-critic (AC) algorithms for adaptive dynamic programming (ADP), we introduce an uncertainty estimator and propose an actor-critic-estimator (ACE) algorithm. The proposed algorithm alternates between the value evaluation, uncertainty estimation, and policy update to generate the adaptive robust control law without knowing the system dynamics. Especially, during the step of uncertainty estimation, we approximate the uncertainty by a radial basis function neural network (RBFNN) and design the appropriate utility function accordingly instead of using the supremum of the uncertainty as in existing studies. The Lyapunov stability theorem provides theoretical demonstrations of the stability and convergence. We further demonstrate that the affine nonlinear systems with uncertainty is uniformly ultimately bounded (UUB) stable when the learned adaptive robust control law is adopted. The performance of the proposed algorithm is demonstrated through a torsion pendulum system and an inverted pendulum system. [ABSTRACT FROM AUTHOR]

Considering the fact that it is very difficult to fully model an autonomous underwater vehicle (AUV) in the complex water environment, this paper presents a model-free tracking control strategy for an AUV in the presence of unknown disturbances. We first formulate an optimized control problem by defining a tracking Hamilton–Jacobi–Isaac (HJI) equation. Then, we present a reinforcement learning (RL) algorithm to compute an optimized solution by learning from the HJI equation online. It is noted that during the learning period, no information about the AUV's dynamics is needed. In order to demonstrate the efficiency of the proposed strategy, numerical simulation is considered, results are validated and discussed. [ABSTRACT FROM AUTHOR]

This paper presents an adaptive incremental nonlinear backstepping sliding-mode (INBSM) controller, for fault tolerant tracking control of a blended wing body (BWB) aircraft with unknown disturbances and actuator faults. The INBSM controller is based on a nonlinear dynamics model of the BWB aircraft. In addition, a radial basis function neural network disturbance observer (RBF-NNDO) is proposed to enhance the disturbance attenuation ability. A fault estimator is suggested to improve actuator fault tolerant control level. The closed-loop control system of the BWB aircraft is proved to be globally asymptotically stable using Lyapunov theory. Simulations of the combined NNDO-INBSM controller are presented and compared with both the INBSM design and an adaptive fuzzy controller. The results demonstrate an improved capability of the NNDO-INBSM control for the BWB aircraft to execute realistic attitude tracking missions, even in the presence of center of gravity movement, unknown disturbances, model uncertainties and actuator faults. [ABSTRACT FROM AUTHOR]

In the paper, data-driven finite-horizon robust formation-containment control scheme is developed based on integral reinforcement learning and zero-sum game for perturbed multi-agent networks with completely unknown nonlinear dynamics. At first, distributed finite-time sliding mode estimators are designed to obtain the desired states of leaders and followers, respectively. Then finite-horizon robust leader formation control and follower containment control are achieved based on proposed model-free integral reinforcement learning algorithms implemented by critic-actor-disturbance structure, in the framework of multi-player zero-sum game where the non-quadratic performance index for each agent considers the influence of saturated inputs and disturbances of local neighbors thoroughly. Furthermore, it is proved that the whole network has bounded L 2 gain robust stability and Nash equilibrium of zero-sum game exists. Simulation results are provided to demonstrate the effectiveness of the proposed scheme. [ABSTRACT FROM AUTHOR]

Recent studies have verified that zeroing neural network is suitable for model-free feedback control of redundant robot manipulators with excellent convergence and accuracy. Unlike previous studies using a linear activation function, this paper employs zeroing neural networks activated by nonlinear functions to control redundant robots to track desired paths without knowing kinematic models of robots, and systematically investigates the finite-time convergence and robustness of the proposed control scheme. Specifically, two nonlinear-function-activated zeroing neural networks are employed to solve the Jacobian estimation problem and trajectory tracking problem respectively. After introducing a model-free control scheme generally applicable to different types of robots, theoretical analysis proves that the proposed control scheme has finite-time convergence when employing nonlinear activation functions and the tracking error will not exceed the upper bound with the bounded noise interference. Finally, simulations based on a five-link planar robot and a PUMA 560 robot reveal the finite-time convergence of the proposed control scheme and verify that nonlinear functions can effectively increase the error convergence rate and reduce the tracking error caused by noises, compared with conventional method based on linear-function-activated zeroing neural networks. [ABSTRACT FROM AUTHOR]

In this paper, a robust non-singular fast terminal sliding mode control scheme with adaptive neural networks is presented for a class of nonlinear systems with unknown bounds of uncertainties. To reduce transmission and computation burden in resource-constrained networked systems, two kinds of event-triggering mechanisms are taken into consideration in the proposed adaptive sliding mode control scheme. The one, from the sensor to the controller, can guarantee finite-time convergence of system states into a predesigned band; and ensure that Zeno behavior does not occur. The other, from the controller to the actuator, can guarantee the asymptotically stability and Zeno behavior exclusion. Simulation results are given to verify the effectiveness and feasibility of the proposed schemes. [ABSTRACT FROM AUTHOR]

This article investigates the guaranteed cost robust tracking control problem of nonlinear systems subjected to input constraint and unmatched uncertainty. The event-based adaptive dynamic programming (ADP) approach is utilized to address this problem. First, the tracking error and reference trajectory are combined to form an augmented uncertain system. Then, by decomposing the uncertainty into the matched and unmatched parts, the original tracking problem is converted into the optimal regulation problem of an auxiliary system. The cost function for the auxiliary system is defined, and the associated Hamilton–Jacobi–Bellman (HJB) equation is solved using a single critic neural network (NN). Moreover, a novel event-triggering rule is formulated, and it is shown that the designed event-based controller guarantees that the tracking error is uniformly ultimately bounded. The derivation of event-based guaranteed cost and its relation with the time-based counterpart is presented. The exclusion of the infamous Zeno behavior is guaranteed. The uniform ultimate boundedness of the critic weight estimation error is shown. Finally, the effectiveness of the proposed event-triggered ADP method is illustrated through simulations of the spring–mass–damper system and Van der Pol's oscillator with unmatched uncertainty. [ABSTRACT FROM AUTHOR]

For the robust tracking control problem of multiagent systems with linear uncertainties, most of the existing studies require full state feedback. Moreover, the designed control strategy also uses global information, such as the number of agents or the eigenvalues of the Laplacian matrix, which does not satisfy the distributed requirement for multiagent design. To address these problems, first, a Luenburger observer is designed to observe the full state of the system by using the output information. Second, the constant gain in the traditional distributed observer for estimating exosystem state information is replaced with a time-varying gain that comes from the output of a newly designed filter, which avoids the use of the global information mentioned above. Finally, based on the above observer and adaptive dynamic feedback, an event-triggered output feedback control strategy is given. This control strategy is not only piecewise constant, which is easy to implement, but also makes the robust tracking control error converge asymptotically to zero. Furthermore, it is theoretically proven that each agent does not have Zeno behaviour, and the effectiveness of the designed control strategy is verified with a simulation example. [ABSTRACT FROM AUTHOR]

An approximate optimal scheme is proposed for an uncertain n -link robot subject to saturation non-linearity. A remarkable feature is that compared with the previous results, the model uncertainty in robotic dynamic is taken into account in the paper. Under the frame of adaptive dynamic programming (ADP), an optimal control is designed for the nominal robotic system and proved to be an approximate optimal control of the unknown robotic system, and it not only stabilizes the unknown system, but also decreases the control cost. Furthermore, the saturation non-linearity is also solved with generalized non-quadratic functional. According to the Lyapunov theory, all the error signals can be proved to be uniformly ultimately bounded (UUB). Simulation examples are implemented to validate the effectiveness of the designed method. [ABSTRACT FROM AUTHOR]

• Various types of robustness used in RL and control communities are classified. • The concept of safety robustness of RL policies is proposed. • A general algorithm of approximating the safety robustness of policies is proposed. • Experiments are conducted to demonstrate how the proposed algorithm works. For a reinforcement learning (RL) problem without a specified reward function, one may specify different reward functions to better guide an agent to learn. With different reward functions, the agent can learn different policies that generally have different robustness. Both the achieved reward and the success rate have been commonly used to evaluate the robustness of policies. Safety is a concern when using RL to solve problems in many safety–critical applications (e.g., robotic manipulation). However, evaluating the robustness of policies from the perspective of safety has not been discussed in the literature. The major contributions of this paper are the proposal of a novel concept of safety robustness to evaluate the robustness of policies from the perspective of safety and an algorithm to approximate the safety robustness of policies. To demonstrate how to implement the proposed algorithm, illustrative experiments are conducted and the safety robustness of three policies for controlling the manipulation of a cable-driven parallel robot is analyzed. Experiment results show that the proposed algorithm can approximate the safety robustness of policies using the ratio of the number of safe episodes to the number of total episodes and identify the best policy from multiple policies in terms of the safety of policies. [ABSTRACT FROM AUTHOR]

This paper proposes a constrained robust control algorithm for some continuous nonlinear systems with partly-unknown dynamics using an adaptive dynamic programming (ADP) method based on event-triggered (ET) mechanism. First, the constrained robustness of partly-unknown systems problems can be transformed into event-triggered optimal control problems with a suitable value function. Second, the identifier network and the critic network are applied to approximate the partly-unknown system and the cost function, respectively. In addition, the stability proof of the closed-loop system is addressed by Lyapunov theory with an appropriate triggering condition and the uniformly ultimately boundedness (UUB) of the closed-loop system is proved under the event-triggered mechanism. Moreover, the simulation experiments results show the superiority of the event-triggered mechanism via adaptive dynamic programming. The optimal control pair are only updated at the aperiodic sampling times determined by the event triggering condition, which can greatly improve the efficiency of communication resource utilization and save computing storage space. [ABSTRACT FROM AUTHOR]

The robust identification problem of the linear parameter varying (LPV) systems with output data corrupted by outliers is considered in this paper. The local identification approach is used, and the LPV model is obtained by interpolating the local models with an exponential weighting function. In order to handle outliers that could occur in industrial processes, the corresponding probabilistic model is established with the process noise assumed to be mixture Laplace distributed, then the formulas to iteratively update the unknown model parameters and noise-free output are derived under the Variational Bayesian (VB) framework, which approximates the required posteriors and could avoid high dimensional integrals. One numerical example and a practical chemical process are employed to verify the efficacy of the developed algorithm. [ABSTRACT FROM AUTHOR]

This paper proposes a new approach to address robust control design for nonlinear continuous-time systems with unmatched uncertainties. First, we transform the robust control problem into an equivalent optimal control problem, which allows to simplify the control design. A critic neural network (NN) is then adopted to reformulate the derived Hamilton–Jacobi–Bellman (HJB) equation based on the optimal control methodology. Then, a novel adaptation algorithm is used to online directly estimate the unknown NN weights, so as to achieve the guaranteed convergence. The control system stability and the convergence of the derived control action to the optimal solution can be rigorously proved. Finally, two simulation examples are provided to illustrate the validity and efficacy of the proposed method. [ABSTRACT FROM AUTHOR]

A robust optimal control method for a class of nonlinear systems with unknown disturbances is addressed in this paper. In this framework, adaptive dynamic programming (ADP) is presented to obtain the optimal control. On-policy learning allows the performance index function and the optimal control to be obtained iteratively. It is shown that the iterative performance index function is non-increasing. A nonlinear disturbance observer is designed to estimate external disturbances. The compensation control is used to compensate for the influence of the disturbances. It is proven that the disturbance observer error is exponentially stable, under some conditions. The properties of the nonlinear system with unknown disturbance steered by the robust optimal control input are also proven. Simulation results demonstrate the performance of the proposed robust optimal control scheme for the nonlinear system with unknown disturbance. [ABSTRACT FROM AUTHOR]

This paper investigates the model reference adaptive control (MRAC) problem for unknown nonlinear discrete-time systems, which is how to guarantee adaptation to the variable reference input and robustness to perturbation. A supervised neural dynamic programming (SNDP) approach is developed to solve this MRAC problem, which includes a learning mode and a control mode. In the learning mode, a data-based adaptive critic learning algorithm is proposed, which guarantees that the controlled objective adaptively tracks the reference model on behavior. Such an algorithm also ensures flexible switching from the learning mode to the control mode under which the robustness of the closed-loop control systems is further improved. By employing a newly defined mode scheduler that regulates the learning mode and the control mode, the adaptation and the robustness of the systems are both achieved by the developed SNDP approach. Its uniformly ultimately bounded property is proved by using Lyapunov method. Simulation results verify that the developed SNDP approach ensures the adaptation to the variable reference input, and has superiority over some traditional MRAC methods in improving the robustness. [ABSTRACT FROM AUTHOR]

• This study develops and compares the end-to-end DDPG strategy and the hybrid DDPG strategy in controlling CDPRs. • The hybrid DDPG strategy is developed by integrating a DDPG algorithm and the inverse dynamics equation of the CDPR. • The hybrid DDPG strategy learns faster and is more robust to model uncertainties than the end-to-end DDPG strategy. • The end-to-end DDPG strategy can learn the optimal tension distribution of a CDPR. Deep reinforcement learning (DRL) has been proven effective in learning policies of high-dimensional states and actions. Recently, a variety of robot manipulation tasks have been accomplished using end-to-end DRL strategies. An end-to-end DRL strategy accomplishes a robot manipulation task as a black box. On the other hand, a robot manipulation task can be divided into multiple subtasks and accomplished by non-learning-based approaches. A hybrid DRL strategy integrates DRL with non-learning-based approaches. The hybrid DRL strategy accomplishes some subtasks of a robot manipulation task by DRL and the rest subtasks by non-learning-based approaches. However, the effects of integrating DRL with non-learning-based approaches on the learning speed and the robustness of DRL to model uncertainties have not been discussed. In this study, an end-to-end DRL strategy and a hybrid DRL strategy are developed and compared in controlling a cable-driven parallel robot. This study shows that, by integrating DRL with non-learning-based approaches, the hybrid DRL strategy learns faster and is more robust to model uncertainties than the end-to-end DRL strategy. This study demonstrates that, by taking advantages of both learning and non-learning-based approaches, the hybrid DRL strategy provides an alternative to accomplish a robot manipulation task. [ABSTRACT FROM AUTHOR]

Owing to the requirements of various product grades or operation conditions, most of industrial processes work with multiple modes. Gaussian mixture regression (GMR) is one of the most widely adopted methods to develop inferential sensors for these processes. However, outliers exist widely due to data points that are incorrectly observed, recorded, or imported and are very hard to be completely recognized and removed. These outliers render the predictive performance of the GMR-based inferential sensors quite disappointing. Aiming at resolving this problem, we propose a robust inferential sensing method based on variational Bayesian Student's- t mixture regression (VBSMR). In the VBSMR, we first explicitly consider the dependency of quality variables on process variables, where the Bayesian regularization is enabled to find the regression coefficients, and the Student's- t distributions are introduced for handling outliers. Subsequently, a computationally efficient parameter learning procedure for the VBSMR using the variational Bayesian expectation maximization (VBEM) technique is developed, where the optimal number of mixing components can be automatically determined. Experiments conducted on both numerical and practical industrial examples are provided to demonstrate the availability and flexibility of the developed inferential sensor. [ABSTRACT FROM AUTHOR]

Stackelberg games allow players to access system information differently and take actions asynchronously. This paper introduces a robust adaptive dynamic programming-based method to solve the nonlinear two-player Stackelberg game subject to external disturbances. Combined with a neural network identifier, our method is implemented on the actor–critic-disturbance structure to approximate the optimal value function, i.e., the corresponding Stackelberg equilibrium. With the aid of costate, we transform this leader–follower optimization problem into solving two parametric equations and a costate equation. The coefficients of critic approximators and the costate are updated simultaneously to reach the Stackelberg equilibrium. The proposed control method finds real-time approximations of the Stackelberg-Saddle equilibrium while ensuring the closed-loop system's stability. Finally, the simulation example shows the effectiveness and advantage of our method. • In our approach, a class of Stackelberg differential games with nonlinear dynamics is solved using the ADP method without knowing system drift dynamics. Notably, we take external disturbances into account, which contributes to the main difference between Mu et al. (2020) and this paper. The algorithm exhibits good robustness, expanding the scope of practical applications. • Our work extends the work (Kebriaei and Iannelli, 2018) to continuous-time nonlinear systems and avoids solving recursive difference Riccati equations. Two actor–critic-disturbance architectures, which have been frequently used in single-input system optimal control problems, are modified to approximate the Stackelberg-Saddle solution under this novel and more complicated task setting. In addition, this paper further investigates the impact of hierarchical input on the control performance in the context of multi-player games. • Furthermore, in contrast to Kebriaei and Razminia (2019), where the action of the follower was regarded as the function of the leader's action, we transform the differential game under consideration into solving two parametric Hamilton–Jacobi–Isaacs (HJI) equations and a costate equation. Then the approximator weights and the costate are updated simultaneously and continuously according to these equations. A comparison study between Kebriaei and Razminia (2019) and our approach is conducted to validate the superiority and effectiveness of our method in the simulation part. [ABSTRACT FROM AUTHOR]

Highlights • The heterogeneous nonlinear dynamical networks with sampled-data, quantization, and input saturation is investigated. • The sufficient conditions of quasi-synchronization are derived via the Lyapunov functional approach. • An effective design for the feedback gain is presented. • The upper bound of quasi-synchronization errors is derived. Abstract This paper considers the problem of quasi-synchronization of heterogeneous dynamical networks with sampled-data and input saturation constrains. By introducing a leader and designing the controller involved the quantization and actuator saturation, the heterogeneous dynamical networks can be transformed into the corresponding error systems with bounded disturbances. A sufficient criterion ensuring the error system can exponentially stable and convergent to a bounded region is established via a Lyapunov functional approach. Based on the criterion, the control gain matrix is further designed. Simulation examples are presented to illustrate the validity of the theoretical results. [ABSTRACT FROM AUTHOR]

Highlights • Adaptive fast terminal sliding mode controllers based on an improved RBF neural network are designed for precisely and quickly tracking the planned trajectories. • An online iteration algorithm is presented to correct the angular deviations caused by the parameter perturbation. • An effective robust control scheme is developed for positioning of a class of uncertain planar underactuated manipulator. Abstract The research on the robust control of uncertain planar underactuated manipulators is almost nonexistent due to the lack of actuator and the uncontrollability at equilibrium points. Taking an uncertain planar three-link passive-active-active underactuated manipulator as an example, this paper develops a robust control scheme including motion planning and adaptive tracking control to realize its position control. According to the constraints between the passive link and active links and the target position of the manipulator, differential evolution algorithm is used to solve the target angles and ratio between the angular velocities. Then, a set of motion trajectories is planned based on the target values. Considering the uncertain parameter perturbation and external disturbance exist, we use RBF neural network to online approximate the uncertainty. Meanwhile, we develop a set of fast terminal sliding mode controllers to track the planned trajectories, and design adaptive laws to guarantee the stability and convergence of the tracking system. Next, an online iteration algorithm is presented to correct the deviations of all link angles caused by the parameter perturbation, which makes the manipulator gradually approach to its target position. Finally, the simulation results verify the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Abstract Sport video analysis is gaining popularity recently owing to its importance in understanding sports and improving the performance of athletes. In this paper we focus on shuttlecock tracking algorithm. Particularly, a novel fast tracking based on object center (i.e., FTOC) method by fusing heterogeneous cues and AdaBoost algorithm are proposed to improve the tracking performance for a robot. Experimental results show that the proposed FTOC tracking method performs favorably against many other popular tracking approaches, such as TLD, MIL, KCF, DCF_CA, SMAF_CA, KCC, DSN, COKCF, etc., in term of speed, accuracy, and robustness, especially in challenging scenarios such as scale variations and background clutter. We further demonstrate the feasibility of the FTOC algorithm in a real-time ZED binocular camera based 3D shuttlecock tracking system for a robot. [ABSTRACT FROM AUTHOR]

Abstract Previous approaches for scene text detection usually rely on manually defined sliding windows. This work presents an intuitive two-stage region-based method to detect multi-oriented text without any prior knowledge regarding the textual shape. In the first stage, we estimate the possible locations of text instances by detecting and linking corners instead of shifting a set of default anchors. The quadrilateral proposals are geometry adaptive, which allows our method to cope with various text aspect ratios and orientations. In the second stage, we design a new pooling layer named Dual-RoI Pooling which embeds data augmentation inside the region-wise subnetwork for more robust classification and regression over these proposals. Experimental results on public benchmarks confirm that the proposed method is capable of achieving comparable performance with state-of-the-art methods. The code is publicly available at https://github.com/xhzdeng/crpn. [ABSTRACT FROM AUTHOR]

Abstract In this paper, a novel robust control strategy based on adaptive dynamics programming (ADP) technique is proposed for multi-player nonlinear systems with input disturbances. A pair of robust control policies is constructed by multiplying appropriate coupling gains to the Nash solution of nominal nonlinear nonzero-sum game with predefined cost functions accounting for system uncertain disturbances. Sufficient conditions for the existence of robust control strategy are derived, and it is proved that the robust control strategy can guarantee the multi-player nonlinear systems to be stable in the sense of uniform ultimate boundedness (UUB) with disturbance rejection. The single-network ADP algorithm is employed to solve the coupled Hamilton–Jacobi equations, where only requires to online tune the weights of critic neural networks (NN) for each player. By utilizing Lyapunov theory, the NN weight estimation errors are proved to be uniformly ultimately bounded, while the stability of the closed-loop nonzero-sum game system is also guaranteed. Two numerical experiments are given to demonstrate the effectiveness of the proposed approach. [ABSTRACT FROM AUTHOR]

Abstract The identification of nonlinear state-space model (NSSM) with output observations corrupted by outliers is investigated in this paper. The outlier is commonly encountered in practical industrial processes which should not be ignored in nonlinear processes modeling. The statistical scheme based on the Student's t-distribution is applied to resist the outlier and the expectation-maximization (EM) algorithm is employed to simultaneously identify the undetermined model and noise parameters. A particle smoother is introduced and used to approximately calculate the desired Q-function. The usefulness of the proposed approach is demonstrated via the numerical and mechanical examples. [ABSTRACT FROM AUTHOR]

Abstract This paper proposes a robust hierarchical controller using adaptive radical basis function neural networks (RBFNNs) based proportional derivative-sliding mode control (RPD-SMC) method and robust integral of the signum of error (RISE) approach for an under-actuated quadrotor in the presence of disturbances and parametric uncertainties. The quadrotor system is decoupled into two parts: the outer loop for position control and the inner loop for attitude control. The RPD-SMC is designed for the outer loop to ensure robust position tracking. The PRD-SMC combines the advantages of simplicity of PD control, the strong robustness of SMC and the approximation ability of arbitrary functions of RBFNNs. The RISE method is applied in the inner loop to guarantee fast convergence of the attitude angles to their desired values with continuous control signals. The capabilities of online approximating and null steady-state tracking are proved using Lyapunov stability theory. The effectiveness of the proposed control strategy is validated by comparing with the performances achieved by PD, PID, PD-SMC and RBFNNs based controllers via numerical simulations. [ABSTRACT FROM AUTHOR]

Abstract Evolutionary Computation is often used in the domain of automated discovery of trading rules. Within this area, both Genetic Programming and Grammatical Evolution offer solutions with similar structures that have two key advantages in common: they are both interpretable and flexible in terms of their structure. The core algorithms can be extended to use automatically defined functions or mechanisms aimed to promote parsimony. The number of references on this topic is ample, but most of the studies focus on a specific setup. This means that it is not clear which is the best alternative. This work intends to fill that gap in the literature presenting a comprehensive set of experiments using both techniques with similar variations, and measuring their sensitivity to an increase in population size and composition of the terminal set. The experimental work, based on three S&P 500 data sets, suggest that Grammatical Evolution generates strategies that are more profitable, more robust and simpler, especially when a parsimony control technique was applied. As for the use of automatically defined function, it improved the performance in some experiments, but the results were inconclusive. [ABSTRACT FROM AUTHOR]

Abstract This paper proposes a methodology to design a robust observer for Takagi–Sugeno descriptor (TS-D) systems with unmeasurable premise variables and its application to sensor fault detection and isolation. A robust H ∞ approach is considered to minimize the effect of uncertainties given by the unmeasurable premise variables, disturbances and sensor noise. As a result, a set of relaxed linear matrix inequalities (LMI) are derived, which provides sufficient conditions to guarantee the convergence of the proposed state observer. Finally, a conventional fault detection scheme is considered by means of residual generation and evaluation. An academic example is given to illustrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Abstract Gesture recognition is a hot topic in computer vision and pattern recognition, which plays a vitally important role in natural human-computer interface. Although great progress has been made recently, fast and robust hand gesture recognition remains an open problem, since the existing methods have not well balanced the performance and the efficiency simultaneously. To bridge it, this work combines image entropy and density clustering to exploit the key frames from hand gesture video for further feature extraction, which can improve the efficiency of recognition. Moreover, a feature fusion strategy is also proposed to further improve feature representation, which elevates the performance of recognition. To validate our approach in a "wild" environment, we also introduce two new datasets called HandGesture and Action3D datasets. Experiments consistently demonstrate that our strategy achieves competitive results on Northwestern University, Cambridge, HandGesture and Action3D hand gesture datasets. Our code and datasets will release at https://github.com/Ha0Tang/HandGestureRecognition. [ABSTRACT FROM AUTHOR]

Highlights • A new robust loss function is designed for imbalanced data sets. • The proposed model combines this loss with SVM. • The robustness of model is analyzed in theory. • The Bayes optimal solution is derived. • Alternative iterative algorithm is designed to reduce algorithm complexity. Abstract Based on minimizing misclassification cost, a new robust loss function is designed in this paper to deal with the imbalanced classification problem under noise environment. It is nonconvex but maintains Fisher consistency. Applying the proposed loss function into support vector machine (SVM), a robust SVM framework is presented which results in a Bayes optimal classifier. However, nonconvexity makes the model difficult to optimize. We develop an alternative iterative algorithm to solve the proposed model. What's more, we analyze the robustness of the proposed model theoretically from a re-weighted SVM viewpoint and the obtained optimal solution is consistent with Bayesian optimal decision rule. Furthermore, numerical experiments are carried out on databases that are drawn from UCI Machine Learning Repository and a practical application. With two different types of noise environments, one with label noise and one with feature noise, experiment results show that on these two databases the proposed method achieves better generalization results compared to other SVM methods. [ABSTRACT FROM AUTHOR]

Abstract In this paper, the grayscale image segmentation problem is investigated and a new mixture model with shifted asymmetric Laplace distribution component is proposed. Instead of the conventional Gaussian model, the shifted asymmetric Laplace distribution model is adopted to model the pixels. The spatial constraint on neighboring pixels is introduced into the proposed shifted asymmetric Laplace mixture model, which makes the model be robust to noise and outliers of the images. The unknown model parameters are estimated via the expectation-maximization (EM) algorithm, which can guarantee convergence to a local minimum. The experimental verification is performed on both synthesized images and images of real chip to prove the effectiveness of our image segmentation method. [ABSTRACT FROM AUTHOR]

Abstract Distance metric learning aims to learn a metric with the similarity of samples. However, the increasing scalability and complexity of dataset or complex application brings about inevitable label noise, which frustrates the distance metric learning. In this paper, we propose a resampling scheme robust to label noise, Re-KISSME, based on Keep It Simple and Straightforward Metric (KISSME) learning method. Specifically, we consider the data structure and the priors of labels as two resampling factors to correct the observed distribution. By introducing the true similarity as latent variable, these two factors are integrated into a maximum likelihood estimation model. As a result, Re-KISSME can reason the underlying similarity of each pair and reduce the influence of label noise to estimate the metric matrix. Our model is solved by iterative algorithm with low computational cost. With synthetic label noise, the experiments on UCI datasets and two application datasets of person re-identification confirm the effectiveness of our proposal. [ABSTRACT FROM AUTHOR]

Abstract RANSAC is a popular robust model estimation algorithm in various computer vision applications. However, the speed of RANSAC declines dramatically as the inlier rate of the measurements decreases. In this paper, a novel Adaptively Ranked Sample Consensus(ARSAC) algorithm is presented to boost the speed and robustness of RANSAC. The algorithm adopts non-uniform sampling based on the ranked measurements to speed up the sampling process. Instead of a fixed measurement ranking, we design an adaptive scheme which updates the ranking of the measurements, to incorporate high quality measurements into sample at high priority. At the same time, a geometric constraint is proposed during sampling process to select measurements with scattered distribution in images, which could alleviate degenerate cases in epipolar geometry estimation. Experiments on both synthetic and real-world data demonstrate the superiority in efficiency and robustness of the proposed algorithm compared to the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Abstract This paper investigates the robust fault detection problem for fuzzy singularly perturbed systems with pole assignment constraints. A finite frequency approach is proposed for the simultaneous achievement of the fault detection objectives and control objectives, which combines the computation of filter gains into a unified framework. By designing the fault detection filter, the residual signal for the detection of finite frequency faults can be produced, and the transient system behaviors can be improved to a satisfactory level based on the pole assignment. In addition, the estimated upper bound of the singular perturbation parameter is used to evaluate the performance of the proposed design method. Finally, a simulation example on an inverted pendulum controlled by a motor via a gear train is provided. [ABSTRACT FROM AUTHOR]

Abstract Binarization is a fundamental problem in document image analysis systems. Different from current thresholding techniques, a novel variational model is proposed for noise robust document image binarization in this work, which is inspired by two classical variational models that have been successfully applied in image segmentation and denoising. In our variational model, the energy functional consists of three terms: data fidelity term, binary classification term and regularization term. As a result, the proposed model is capable of performing image binarization and suppressing noise simultaneously. Concretely, we firstly design a variational model for the original document image, the minimizer of which is our desired binarization result. Secondly, the gradient descent flow equation of our model is derived by means of variational principle. Lastly, a simple finite difference scheme, time forward and space center difference, is employed to solve the gradient descent flow equation. Extensive experiments are conducted on three types of document image datasets to validate our model qualitatively and quantitatively. The experiment results not only demonstrate the effectiveness of our approach, but also verify its noise and illumination robustness. [ABSTRACT FROM AUTHOR]

Abstract Broad Learning System (BLS) has achieved outstanding performance in classification and regression problems. Specifically, the accuracy and efficiency can be balanced well by BLS. However, the presence of outliers in data may destroy the stability and generality of standard BLS. In this paper, we propose the robust version of BLS (RBLS) to treat the data modeling with outliers. By assuming the regression residual and output weights follow their respective distributions, the objective function for RBLS is derived and the output weights for robust modeling can be determined by maximum a posterior estimation. Then the robustness of RBLS can be enhanced further by integrating the regularization theory. The Augmented Lagrange Multiplier method is utilized to optimize the novel models efficiently, and a solid theoretical proof is given to guarantee that the proposed RBLS is more robust than the standard BLS. Extensive experiments on function approximation and real-world regression are carried out to demonstrate that our proposed RBLS model can achieve a better modeling performance in uncertain data environment than the standard BLS and other regression algorithms. [ABSTRACT FROM AUTHOR]

Abstract In this paper, the robust boundary iterative learning control for the output tracking and disturbance attenuation of the 2 × 2 nonlinear hyperbolic system is addressed. Since the measurement limitation, the control and measurement are implemented at the same boundary of the system and the disturbance is not necessary to be estimated, which makes the iterative learning control be easy in implementation and low in measurement cost. By using the characteristic method, the robust convergence with respect to iteration-varying uncertainties arising from initial states shift, external disturbances, model plants uncertainties and disturbed reference trajectories is analyzed without any model reduction, rigorously. It is shown that the robust convergence bound is continuously dependent on the bounds of the iteration-varying uncertainties. Furthermore, to implement the proposed iterative learning control, the actuator dynamic is considered, also. Finally, with the actuator dynamic, two examples are given to demonstrate the effectiveness of the proposed iterative learning control strategy for the 2 × 2 nonlinear hyperbolic system. [ABSTRACT FROM AUTHOR]

Abstract A robust adaptive neural network (NN) control law is developed for dynamic positioning (DP) of vessels with unknown dynamics and unknown time-varying disturbances under input constraints through incorporating adaptive radial basis function (RBF) NNs, an auxiliary dynamic system and a robust control term into dynamic surface control method. The developed DP control law makes the DP closed-loop system be uniformly ultimately stable and the vessel's position and heading be maintained at the desired values with arbitrarily small errors. The advantages of the proposed control scheme are that: first, the developed DP control law does not require any priori knowledge of vessel dynamics and disturbances under input constraints, and prevents the presence of input constraints from degrading control performance and even destabilizing the DP control system; second, the developed DP control law compensates for not only unknown time-varying disturbances but also NN approximation errors for unknown vessel dynamics. Simulations on two supply vessels are conducted to exhibit the efficiency and control performance of the developed DP control law. [ABSTRACT FROM AUTHOR]

Abstract Correlation filter based trackers have achieved appealing performance and high efficiency in recent years. However, for long-term tracking where target objects undergo dramatic appearance variation due to heavy occlusion or out-of-view, conventional correlation filter based tracking algorithms would be distracted by irrelevant objects. Once the trained tracker loses its way, it is impossible to recover the information for the following frames as the model has drifted. In this paper, we decompose the long-term tracking task into tracking and detection. Tracker learns separate correlation filters for explicit translation and scale estimation. Specifically, in order to improve tracking accuracy, the convolutional features for translation filter are extracted, and the scale filter is learned using the target appearance sampled at different scales. Detector trains an online long-term filter and applies it to the entire frame to generate detection proposals. By exploiting these detection proposals, it helps the tracker to recover from problems such as temporary or persistent occlusions. In this way, the proposed approach could handle the model drifting problem effectively for long-term tracking with more accurate estimation of object scale and location. Extensive experimental results on large-scale benchmark sequences have shown the robustness of the proposed method. [ABSTRACT FROM AUTHOR]

Abstract To further improve the robustness of two-dimensional LDA (2DLDA) methods against outliers, this paper proposes a new robust 2DLDA version which obtains the optimal projection transformation by maximizing the correntropy-based within-class similarity and maintaining the global dispersity simultaneously. The objective problem of the proposed method can be solved by an iterative optimization algorithm which is proved to converge at a local maximum point. The experimental results on FERET face database, PolyU palmprint database and Binary Alphadigits database illustrate that the proposed method outperforms three conventional 2DLDA methods when there are outliers. [ABSTRACT FROM AUTHOR]

Abstract Graph-based approaches have been successfully used in semi-supervised learning (SSL) by weighting the affinity between the corresponding pairs of samples. Low-rank representation (LRR) is one of the state-of-the-art methods, has been extensively employed in the existing graph-based learning models. In graph-based SSL, a weighted affinity graph induced by LRR coefficients is used to reflect the relationships among data samples. Constructing a robust and discriminative graph to discover the intrinsic structures of the data is critical for SSL, because the central idea behind the graph-based SSL approaches is to explore the pairwise affinity between data samples to infer those unknown labels. However, most of existing LRR-based approaches fail to guarantee weight consistency for each pair of data samples, which is beneficial to preserve the subspace structures of high dimensional data. In this paper, we propose a symmetric LRR with adaptive distance penalty (SLRRADP) method for the small sample size (SSS) recognition problem. The graph identified by SLRRADP can not only preserve global and local structures of data, but also exploit intrinsic correlation information of data samples. Extensive experimental results on semi-supervised classification tasks demonstrate the effectiveness of the proposed method and its superiority in performance over the state-of-the-art graph construction approaches. [ABSTRACT FROM AUTHOR]