Showing papers on "Adaptive algorithm published in 2020"

Adaptive Finite-Time Fuzzy Control of Nonlinear Active Suspension Systems With Input Delay

Abstract: This paper presents a new adaptive fuzzy control scheme for active suspension systems subject to control input time delay and unknown nonlinear dynamics. First, a predictor-based compensation scheme is constructed to address the effect of input delay in the closed-loop system. Then, a fuzzy logic system (FLS) is employed as the function approximator to address the unknown nonlinearities. Finally, to enhance the transient suspension response, a novel parameter estimation error-based finite-time (FT) adaptive algorithm is developed to online update the unknown FLS weights, which differs from traditional estimation methods, for example, gradient algorithm with ${e}$ -modification or ${\sigma }$ -modification. In this framework, both the suspension and estimation errors can achieve convergence in FT. A Lyapunov–Krasovskii functional is constructed to prove the closed-loop system stability. Comparative simulation results based on a dynamic simulator built in a professional vehicle simulation software, Carsim, are provided to demonstrate the validity of the proposed control approach, and show its effectiveness to operate active suspension systems safely and reliably in various road conditions.

Robust Adaptive Control Scheme for Teleoperation Systems With Delay and Uncertainties

Abstract: This paper proposes a robust adaptive algorithm that effectively copes with time-varying delay and uncertainties in Internet-based teleoperation systems. Time-delay induced by the communication network, as a major problem in teleoperation systems, along with uncertainties in modeling of robotic manipulators and remote environment warn the stability and performance of the system. A robust adaptive control algorithm is developed to deal with the system uncertainties and to provide a smooth estimation of delayed reference signals. The proposed control algorithm generates chattering-free torques which is one of the practical considerations for robotic applications. In addition, the achieved input-to-state stability gains do not necessarily require high gain control torques to retain the system’s stability. Experimental simulation studies validate the effectiveness of the proposed control strategy on a teleoperation system consisting of a Phantom Omni Haptic device and SimMechanics model of the industrial manipulator UR10. The validation of the proposed control methodology was executed through a real-time Internet-based communication established over 4G mobile networks between Australia and Scotland.

An Adaptive Backstepping Sliding Mode Controller to Improve Vehicle Maneuverability and Stability via Torque Vectoring Control

Abstract: To improve the maneuverability and stability of a vehicle and fully leverage the advantages of torque vectoring technology in vehicle dynamics control, a finite-time yaw rate and sideslip angle tracking controller is proposed by combining a second-order sliding mode (SOSM) controller with the backstepping method in this paper. However, existing research indicates that first-order sliding mode (FOSM) control suffers from the chattering problem, while the traditional SOSM controller requires knowing the bound of the uncertain term in advance to obtain the switching gain, which is difficult in practice. To address these problems, this paper proposes an adaptive second-order sliding mode (ASOSM) controller based on the backstepping method by adding the high-frequency switching term to the first derivative of the sliding mode variable, which implies that the actual control can be acquired after an integration process. The switching gain in the ASOSM controller is obtained by an adaptive algorithm without knowing any information of the uncertainty. The proposed algorithm is compared with FOSM and SOSM in different scenarios to demonstrate its applicability and robustness. Simulation results show that the bandwidth of the vehicle transient response can be improved by 21%. In addition, ASOSM and SOSM controllers are insensitive to vehicle mass and tire type, implying their robustness to such disturbances. Furthermore, ASOSM requires less control action because of the adaptive law when it performs similarly with SOSM and FOSM.

Model Predictive Adaptive Constraint Tracking Control for Underwater Vehicles

Abstract: In this article, in order to solve the trajectory tracking control problem with the drive saturation (thrust overrun) for the 4500-m human occupied vehicle named “Deep-sea Warrior,” a model predictive adaptive constraint control strategy is put forward. The proposed control strategy mainly consists of two controllers. The first part is a kinematics controller designed by quantum-behaved particle swarm optimization model predictive control method. The second part is a dynamic controller designed by an adaptive algorithm. In order to study the effect of the ocean current disturbance on tracking controller, the ocean current is incorporated into the kinematics and dynamics model of the 4500-m human occupied vehicle. The thrusts of four degrees of freedom under the ocean current are calculated from designed controllers. Then, the thrusts are assigned to six thrusters on the 4500-m human occupied vehicle according to its thruster arrangement. An ocean current observer based on artificial fish proportional-integral control is designed for unknown currents. The simulation results of tracking control in three-dimensional underwater environment are given, which illustrates that the proposed control strategy can not only meet the hardware requirements (drive saturation) but also achieve a stable and efficient tracking control performance because of its constraint to speed and speed increment, the effect of the ocean current on kinematics and dynamics models and the dual feedback mechanism.

A Separable Maximum Correntropy Adaptive Algorithm

Abstract: In this brief, a separable maximum correntropy criterion (SMCC) algorithm is developed by exploiting the typical separability property of tensors. Utilizing the separability property, a great number savings are obtained along with accelerated learning rate and improved estimate accuracy. In the proposed SMCC, a correntropy scheme is used to construct a adaptive algorithm to combat the impulsive noise and outliers in non-Gaussian environment. The complexity and convergence analysis of the SMCC are presented and discussed. Examples with two-way matrix and three-way tensor are carried out to verify the performance of the proposed SMCC algorithm under mixture Gaussian and Student’s t noises.

Adam revisited: a weighted past gradients perspective

Abstract: Adaptive learning rate methods have been successfully applied in many fields, especially in training deep neural networks. Recent results have shown that adaptive methods with exponential increasing weights on squared past gradients (i.e., ADAM, RMSPROP) may fail to converge to the optimal solution. Though many algorithms, such as AMSGRAD and ADAMNC, have been proposed to fix the non-convergence issues, achieving a data-dependent regret bound similar to or better than ADAGRAD is still a challenge to these methods. In this paper, we propose a novel adaptive method weighted adaptive algorithm (WADA) to tackle the non-convergence issues. Unlike AMSGRAD and ADAMNC, we consider using a milder growing weighting strategy on squared past gradient, in which weights grow linearly. Based on this idea, we propose weighted adaptive gradient method framework (WAGMF) and implement WADA algorithm on this framework. Moreover, we prove that WADA can achieve a weighted data-dependent regret bound, which could be better than the original regret bound of ADAGRAD when the gradients decrease rapidly. This bound may partially explain the good performance of ADAM in practice. Finally, extensive experiments demonstrate the effectiveness of WADA and its variants in comparison with several variants of ADAM on training convex problems and deep neural networks.

Data Rate Enhancement in Optical Camera Communications Using an Artificial Neural Network Equaliser

Abstract: In optical camera communication (OCC) systems leverage on the use of commercial off-the-shelf image sensors to perceive the spatial and temporal variation of light intensity to enable data transmission. However, the transmission data rate is mainly limited by the exposure time and the frame rate of the camera. In addition, the camera's sampling will introduce intersymbol interference (ISI), which will degrade the system performance. In this paper, an artificial neural network (ANN)-based equaliser with the adaptive algorithm is employed for the first time in the field of OCC to mitigate ISI and therefore increase the data rate. Unlike other communication systems, training of the ANN network in OCC is done only once in a lifetime for a range of different exposure time and the network can be stored with a look-up table. The proposed system is theoretically investigated and experimentally evaluated. The results record the highest bit rate for OCC using a single LED source and the Manchester line code (MLC) non-return to zero (NRZ) encoded signal. It also demonstrates 2 to 9 times improved bandwidth depending on the exposure times where the system's bit error rate is below the forward error correction limit.

A Robust Total Least Mean M-Estimate Adaptive Algorithm for Impulsive Noise Suppression

Abstract: The errors-in-variables (EIV) model is widely used in linear systems where both input and output signals are contaminated with noise. For the parameter estimation in the EIV model, the adaptive filtering algorithm using total least squares (TLS) approach has shown better performance than classical least squares (LS) approach. However, the TLS approach which is based on minimizing the mean squared total error may be irrational in the presence of impulsive noise. To address this problem, a novel robust adaptive algorithm, named as the total least mean M-estimate (TLMM) algorithm, is proposed in this brief, which combines the advantages of TLS approach and M-estimate function. In addition, to further improve the performance of the TLMM algorithm, its variable step-size (VSS) version has been developed. Moreover, we carry out the local stability analysis and the computational complexity analysis. Simulation results show that the proposed algorithms outperform some well-known algorithms.

Adaptive Visual Servoing for an Underwater Soft Robot Considering Refraction Effects

Abstract: Robots inspired from marine organisms are tremendously developed for applications of underwater exploration, rescuing, navigation, etc. In this article, we implement an uncalibrated visual servoing scheme to achieve accurate positioning performance of an octopus-tentacle-like soft robot arm. The image-based adaptive visual servoing controller is designed based on the underwater dynamic model of the robot system. An adaptive mechanism to solve the tedious camera calibration problem is also extremely complicated in underwater environment due to the refraction effect resulting in changes of optical condition. In this article, this effect is analogous to the radial distortion. The presented algorithm can linearize the distortion model, and then online iteratively estimate the unknown image mapping model based on the classical Slotine-Li adaptive algorithm. The intrinsic and extrinsic parameters can also be estimated in real time. The presented adaptive controller is verified both theoretically using the Lyapunov stability analysis to prove the stability of the dynamical system, and experimentally to prove the accuracy and rapid convergence to the target image position.

Adaptive multi-population inflationary differential evolution

Abstract: This paper proposes a multi-population adaptive version of inflationary differential evolution algorithm. Inflationary differential evolution algorithm (IDEA) combines basic differential evolution (DE) with some of the restart and local search mechanisms of Monotonic Basin Hopping (MBH). In the adaptive version presented in this paper, the DE parameters $${ CR}$$ and F are automatically adapted together with the size of the local restart bubble and the number of local restarts of MBH. The proposed algorithm implements a simple but effective mechanism to avoid multiple detections of the same local minima. The novel mechanism allows the algorithm to decide whether to start or not a local search. The algorithm has been extensively tested over more than fifty test functions from the competitions of the Congress on Evolutionary Computation (CEC), CEC 2005, CEC 2011 and CEC 2014, and compared against all the algorithms participating in those competitions. For each test function, the paper reports best, worst, median, mean and standard deviation values of the best minimum found by the algorithm. Comparisons with other algorithms participating in the CEC competitions are presented in terms of relative ranking, Wilcoxon tests and success rates. For completeness, the paper presents also the single population adaptive IDEA, that can adapt only $$\textit{CR}$$ and F, and shows that this simpler version can outperform the multi-population one if the radius of the restart bubble and the number of restarts are properly chosen.

Finite-Time Last-Iterate Convergence for Multi-Agent Learning in Games

Abstract: In this paper, we consider multi-agent learning via online gradient descent in a class of games called $\lambda$-cocoercive games, a fairly broad class of games that admits many Nash equilibria and that properly includes unconstrained strongly monotone games. We characterize the finite-time last-iterate convergence rate for joint OGD learning on $\lambda$-cocoercive games; further, building on this result, we develop a fully adaptive OGD learning algorithm that does not require any knowledge of problem parameter (e.g. cocoercive constant $\lambda$) and show, via a novel double-stopping time technique, that this adaptive algorithm achieves same finite-time last-iterate convergence rate as non-adaptive counterpart. Subsequently, we extend OGD learning to the noisy gradient feedback case and establish last-iterate convergence results--first qualitative almost sure convergence, then quantitative finite-time convergence rates-- all under non-decreasing step-sizes. To our knowledge, we provide the first set of results that fill in several gaps of the existing multi-agent online learning literature, where three aspects--finite-time convergence rates, non-decreasing step-sizes, and fully adaptive algorithms have been unexplored before.

Fast Adaptive Gradient RBF Networks For Online Learning of Nonstationary Time Series

Abstract: For a learning model to be effective in online modeling of nonstationary data, it must not only be equipped with high adaptability to track the changing data dynamics but also maintain low complexity to meet online computational restrictions. Based on these two important principles, in this paper, we propose a fast adaptive gradient radial basis function (GRBF) network for nonlinear and nonstationary time series prediction. Specifically, an initial compact GRBF model is constructed on the training data using the orthogonal least squares algorithm, which is capable of modeling variations of local mean and trend in the signal well. During the online operation, when the current model does not perform well, the worst performing GRBF node is replaced by a new node, whose structure is optimized to fit the current data. Owing to the local one-step predictor property of GRBF node, this adaptive node replacement can be done very efficiently. Experiments involving two chaotic time series and two real-world signals are used to demonstrate the superior online prediction performance of the proposed fast adaptive GRBF algorithm over a range of benchmark schemes, in terms of prediction accuracy and real-time computational complexity.

A Dynamic Adaptive Algorithm for Merging Into Platoons in Connected Automated Environments

Abstract: The automated merging control is one of the connected and automated vehicle applications that is expected to improve the operation and safety of freeway merging areas. This study introduces a dynamic adaptive algorithm for automated merging control based on a cost function of travel time for both on-ramp and mainline vehicles. The developed algorithm is designed to guide on-ramp vehicles to merge efficiently, without frequent slowdown or wait for merging gaps at the end of the ramp and with minimal disruption to the mainline traffic. The algorithm was evaluated under different mainline traffic demands, ranging from light to heavy conditions, using a simulation model for a one-mile freeway segment with two lanes and a single-lane on-ramp. Platooning is implemented in the simulation based on Cooperative Adaptive Cruise Control. The performance of the algorithm is compared to a base case where mainline vehicles cooperatively decelerate to help on-ramp vehicles merge. The results show that the proposed algorithm reduced ramp delays in the range of 38% to 91% for different traffic conditions, without disrupting mainline operation. Additionally, travel time reliability index improved in the range of 18% to 48% under different traffic conditions. According to the simulation results, the proposed automated merging algorithm is henceforth considered reliable, with its promising performance.

An Adaptive Fitness-Dependent Optimizer for the One-Dimensional Bin Packing Problem

Abstract: In recent years, the one-dimensional bin packing problem (1D-BPP) has become one of the most famous combinatorial optimization problems. The 1D-BPP is a robust NP-hard problem that can be solved through optimization algorithms. This paper proposes an adaptive procedure using a recently optimized swarm algorithm and fitness-dependent optimizer (FDO), named the AFDO, to solve the BPP. The proposed algorithm is based on the generation of a feasible initial population through a modified well-known first fit (FF) heuristic approach. To obtain a final optimized solution, the most critical parameters of the algorithm are adapted for the problem. To the best of our knowledge, this is the first study to apply the FDO algorithm in a discrete optimization problem, especially for solving the BPP. The adaptive algorithm was tested on 30 instances obtained from benchmark datasets. The performance and evaluation results of this algorithm were compared with those of other popular algorithms, such as the particle swarm optimization (PSO) algorithm, crow search algorithm (CSA), and Jaya algorithm. The AFDO algorithm obtained the smallest fitness values and outperformed the PSO, CS, and Jaya algorithms by 16%, 17%, and 11%, respectively. Moreover, the AFDO shows superiority in terms of execution time with improvements over the execution times of the PSO, CS, and Jaya algorithms by up to 46%, 54%, and 43%, respectively. The experimental results illustrate the effectiveness of the proposed adaptive algorithm for solving the 1D-BPP.

Adaptive Spatial Filtering of High-Density EMG for Reducing the Influence of Noise and Artefacts in Myoelectric Control

Abstract: Electromyography (EMG) is a source of neural information for controlling neuroprosthetic devices. To enhance the information content of conventional bipolar EMG, high-density EMG systems include tens to hundreds of closely spaced electrodes that non-invasively record the electrical activity of muscles with high spatial resolution. Despite the advantages of relying on multiple signal sources, however, variations in electrode-skin contact impedance and noise remain challenging for multichannel myocontrol systems. These spatial and temporal non-stationarities negatively impact the control accuracy and therefore substantially limit the clinical viability of high-density EMG techniques. Here, we propose an adaptive algorithm for automatic artefact/noise detection and attenuation for high-density EMG control. The method infers the presence of noise in each EMG channel by spectro-temporal measures of signal similarity. These measures are then used for establishing a scoring system based on an adaptive weighting and reinforcement formulation. The method was experimentally tested as a pre-processing step for a multi-class discrimination problem of 4-digit activation. The approach was proven to enhance the discriminative information content of high-density EMG signals, as well as to attenuate non-stationary artefacts, with improvements in accuracy and robustness of the classification.

Finite-time time-varying output formation-tracking of heterogeneous linear multi-agent systems

Abstract: In this paper, the finite-time time-varying output formation-tracking (FT-TV-OFT) problem of heterogeneous linear multi-agent systems (HL-MASs) is investigated. First, we propose an adaptive finite-time observer for each follower, aiming to estimate not only the leader’s state but also the leader’s system matrix. Compared with conventional observers, the dependence of the observer parameters on any global information is completely removed. Moreover, exact estimate can be achieved in a finite time. Then, an adaptive algorithm is used to calculate the regulation equations, such that control gains can be obtained. Next, an adaptive distributed controller is constructed to solve the FT-TV-OFT problem. Moreover, we also extend above results to the case of the directed communication topology. Finally, examples are given to demonstrate the effectiveness of the results.

A Mesh-free Method Using Piecewise Deep Neural Network for Elliptic Interface Problems

Abstract: In this paper, we propose a novel mesh-free numerical method for solving the elliptic interface problems based on deep learning. We approximate the solution by the neural networks and, since the solution may change dramatically across the interface, we employ different neural networks in different sub-domains. By reformulating the interface problem as a least-squares problem, we discretize the objective function using mean squared error via sampling and solve the proposed deep least-squares method by standard training algorithms such as stochastic gradient descent. The discretized objective function utilizes only the point-wise information on the sampling points and thus no underlying mesh is required. Doing this circumvents the challenging meshing procedure as well as the numerical integration on the complex interface. To improve the computational efficiency for more challenging problems, we further design an adaptive sampling strategy based on the residual of the least-squares function and propose an adaptive algorithm. Finally, we present several numerical experiments in both 2D and 3D to show the flexibility, effectiveness, and accuracy of the proposed deep least-square method for solving interface problems.

Online estimation of state of health for the airborne Li-ion battery using adaptive DEKF-based fuzzy inference system

Abstract: The quick and accurate estimation of the state of health (SOH) of Li-ion battery is a technical difficulty in battery management system research. For the low accuracy of Li-ion battery SOH estimation under complex stress conditions, an estimation method of SOH for Li-ion battery using the adaptive dual extended Kalman filter-based fuzzy inference system (ADEKF-FIS) is proposed. First, Li-ion battery SOH is online estimated by dual extended Kalman filter. Then the Sage–Husa adaptive algorithm and the fuzzy controller are used to correct the state noise covariance and the observed noise covariance, respectively. The algorithm is flat on the state variance and the noise variance. The recursive estimation of the square root ensures the symmetry and nonnegative nature of the state and noise variance. In the end, this paper performing the dynamic stress test condition experiment for confirmation. Experimental results show that, compared with the EKF algorithm, ADEKF-FIS algorithm can obtain state of charge estimation with higher accuracy, which further improves the prediction accuracy of SOH and makes this algorithm have higher accuracy and better convergence.

Adaptive stochastic Galerkin FEM for lognormal coefficients in hierarchical tensor representations

Abstract: Stochastic Galerkin methods for non-affine coefficient representations are known to cause major difficulties from theoretical and numerical points of view In this work, an adaptive Galerkin FE method for linear parametric PDEs with lognormal coefficients discretized in Hermite chaos polynomials is derived It employs problem-adapted function spaces to ensure solvability of the variational formulation The inherently high computational complexity of the parametric operator is made tractable by using hierarchical tensor representations For this, a new tensor train format of the lognormal coefficient is derived and verified numerically The central novelty is the derivation of a reliable residual-based a posteriori error estimator This can be regarded as a unique feature of stochastic Galerkin methods It allows for an adaptive algorithm to steer the refinements of the physical mesh and the anisotropic Wiener chaos polynomial degrees For the evaluation of the error estimator to become feasible, a numerically efficient tensor format discretization is developed Benchmark examples with unbounded lognormal coefficient fields illustrate the performance of the proposed Galerkin discretization and the fully adaptive algorithm

Adaptive fixed-time fault-tolerant control for noncooperative spacecraft proximity using relative motion information

Abstract: This paper studies the control issue of noncooperative spacecraft proximity. In particular, a pursuer spacecraft approaches a noncooperative target while synchronizing its attitude with the target. The pursuer spacecraft is subject to actuator faults and parametric uncertainties. Due to the existence of spatial disturbances on both the target and pursuer, the six-dof (degree-of-freedom) relative motion dynamics are first established in the pursuer’s body frame. Then, by merely using the relative information, a novel adaptive fixed-time fault-tolerant control strategy is proposed under a backstepping framework. Specifically, the relative attitude and position controllers are designed by using an adaptive robust control scheme against parametric uncertainties and spatial disturbances, where a virtual control coefficient-based adaptive algorithm is also introduced to offset the actuator fault effects. It is shown that the relative states driven by the proposed controllers are bounded and converge to a small neighborhood of origin in a fixed time. Simulation comparisons further highlight the proposed control strategy.

APTEEN routing protocol optimization in wireless sensor networks based on combination of genetic algorithms and fruit fly optimization algorithm

Abstract: APTEEN routing protocol exists the problems of uneven network energy consumption, premature death of some nodes, consume too much unnecessary energy and low effective coverage of the whole network. To solve these problems, this paper optimizes the APTEEN routing protocol by combining genetic algorithm with fruit fly optimization algorithm. By adding residual energy, distance from node to base station, distance from node to geometric center of the whole network, node degree and other selection factors to cluster heads selection, the genetic algorithm and fruit fly optimization algorithm is used to select cluster heads for the first time, and the second time of cluster heads selection based on density adaptive algorithm. Some nodes are selected to sleep according to the position and degree of nodes. The residual energy of cluster head, the distance between node and cluster head, and the number of cluster members are taken into account when nodes join clusters. When energy is transmitted from cluster heads to base station, the Dijkstra algorithm is used to find the optimal path. Add the rule of rotating cluster heads when the energy consumption of data transmission is too high, and the GA-APTEEN routing protocol is obtained through the above optimization. The simulation results show that the GA-APTEEN improves the 50% lifetime, 10% coverage and robustness of the network, reduces the energy consumption of the overall network system and avoids the phenomenon of the hot zone of energy.

Maximum Total Complex Correntropy for Adaptive Filter

Abstract: Nowadays, complex Correntropy has been widely used for adaptive filtering in the complex domain. Compared with the second order statistics methods, the complex correntropy based algorithms have shown the superiority in the non-Gaussian noise, especially the impulsive noise. However, the current complex correntropy based adaptive filtering algorithms have not taken the input noise into consideration, and the performances will be deteriorated when the input signals are also corrupted by the noise. In this article, we focus on the errors-in-variables (EIV) model and propose an adaptive algorithm based on the maximum total complex correntropy (MTCC). More importantly, we present the local stability analysis and derive the theoretical weight error power. Simulation results confirm the validity of the theoretical analysis and illustrate the superior performance of the propose algorithm in the EIV cases.

A Multi-Scale Tensor Network Architecture for Classification and Regression

Abstract: We present an algorithm for supervised learning using tensor networks, employing a step of preprocessing the data by coarse-graining through a sequence of wavelet transformations. We represent these transformations as a set of tensor network layers identical to those in a multi-scale entanglement renormalization ansatz (MERA) tensor network, and perform supervised learning and regression tasks through a model based on a matrix product state (MPS) tensor network acting on the coarse-grained data. Because the entire model consists of tensor contractions (apart from the initial non-linear feature map), we can adaptively fine-grain the optimized MPS model backwards through the layers with essentially no loss in performance. The MPS itself is trained using an adaptive algorithm based on the density matrix renormalization group (DMRG) algorithm. We test our methods by performing a classification task on audio data and a regression task on temperature time-series data, studying the dependence of training accuracy on the number of coarse-graining layers and showing how fine-graining through the network may be used to initialize models with access to finer-scale features.