Convex Optimization-Based Adaptive Neural Network Control for Unmanned Surface Vehicles Considering Moving Obstacles

An adaptive neural network control scheme is developed for a class of nonlinear systems in the strict-feedback form. Compared with the existing approaches, the main advantage is that the developed scheme can be implemented by utilizing only one neural network approximator. Thus, the designed controller structure is simplified. In addition, less neural network can reduce the running cost in practical application. The developed neural network control scheme can achieve that all the signals of the closed-loop system are uniformly bounded and the tracking errors converge to an arbitrary small neighborhood around zero by selecting suitably design parameters.

In this paper; a new robust adaptive neural network controller (RANNC) is presented for a class of non-affine nonlinear systems in the presence of unknown nonlinearities and disturbances. Firstly, the existence of an ideal implicit feedback linearisation control (IFLC) is established based on implicit function theory. Using Taylor series expansion, it is shown that the control of non-affine nonlinear systems is equivalent to the control ofaffine nonlinear systems in the neighbourhood of the operating trajectory under mild conditions. Then, a robust adaptive neural network control scheme is presented for the transformed nonlinear systems by using neural networks as universal approximators for the unknown system nonlinearities. The proposed RANNC can guarantee that all the signals in the closed-loop system are bounded and the tracking error asymptotically converges to zero. Simulation studies on the control of a continuously stirred tank reactor (CSTR) system are used to show the effectiveness of the scheme.

This paper investigates a scheme of adaptive neural network control for a stochastic switched system with input saturation. The unknown smooth nonlinear functions are approximated directly by neural networks. A modified approach is proposed to deal with unknown functions with nonstrict feedback form in the design process. Furthermore, by combining the auxiliary design signal and the adaptive backstepping design, a valid adaptive neural tracking controller design algorithm is presented such that all the signals of the switched closed-loop system are in probability semiglobally, uniformly, and ultimately bounded, and the tracking error eventually converges to a small neighborhood of the origin in probability. In the end, the effectiveness of the proposed method is verified by a simulation example.

This manuscript considers the finite time adaptive neural tracking control problem for a class of the nonstrict-feedback stochastic nonlinear systems with Bouc-Wen hysteresis input. During the design process, a Bouc-Wen model is first adopted to obtain the input hysteresis phenomenon. By fusion with the backstepping technique and the neural network approximation capability, the unknown nonlinearities are coped with, a constructive finite time adaptive neural network control strategy is proposed. Furthermore, the finite-time mean square stability of stochastic nonlinear systems is proved, and at last the effectiveness of the proposed control strategy is validated by the simulations.

In this paper, we present a neural adaptive control scheme for active vibration suppression of a composite aircraft fin tip. The mathematical model of a composite aircraft fin tip is derived using the finite element approach. The finite element model is updated experimentally to reflect the natural frequencies and mode shapes very accurately. Piezo-electric actuators and sensors are placed at optimal locations such that the vibration suppression is a maximum. Model-reference direct adaptive neural network control scheme is proposed to force the vibration level within the minimum acceptable limit. In this scheme, Gaussian neural network with linear filters is used to approximate the inverse dynamics of the system and the parameters of the neural controller are estimated using Lyapunov based update law. In order to reduce the computational burden, which is critical for real-time applications, the number of hidden neurons is also estimated in the proposed scheme. The global asymptotic stability of the overall system is ensured using the principles of Lyapunov approach. Simulation studies are carried-out using sinusoidal force functions of varying frequency. Experimental results show that the proposed neural adaptive control scheme is capable of providing significant vibration suppression in the multiple bending modes of interest. The performance of the proposed scheme is better than the H(infinity) control scheme.

Constrained adaptive neural network control of an MIMO aeroelastic system with input nonlinearities

This article focuses on the application issue of deep neural networks for the control strategy in unmanned surface vehicles. A novel adaptive law is developed to update the weights of the deep neural networks, thereby enhancing the control strategy’s ability to handle uncertain dynamics. We establish the relationship between the output of the deep neural networks and the dynamic response of the unmanned surface vehicle using the command filter method. And then, we develop a convex optimization-based adaptive approach to address the challenges of online learning within the deep neural network. We also prove the convergence of the signals in our controller using the Lyapunov stability theorem. Finally, simulation and experiment results are given to demonstrate the effectiveness of our proposed method.

PurposeThis paper aims to design a fixed-time disturbance observer (FTDO)-based adaptive neural fixed-time control strategy for the uncertain medium-scale unmanned autonomous helicopter (UAH) with full-state constraints and external disturbances.Design/methodology/approachFirst, the tan-type barrier Lyapunov function (BLF) and neural network (NN) are constructed to deal with the full-state constraints and system uncertainties, respectively. Subsequently, the compound disturbances are estimated by FTDOs. Simultaneously, the FTDOs and fixed-time control strategy guarantees that the errors of disturbances and states converge to the desired region in fixed-time, and the upper bound on the convergence time of the FTDOs and the designed controller can be estimated through the devised parameters. In addition, Lyapunov stability theory proves that all states of the UAH system are semiglobally uniform ultimately bounded.FindingsThe designed controller combines the fixed-time stability theory with tan-type BLF, which proves that all states of the UAH can be constrained in a predefined region. Furthermore, the state and the disturbance errors converge to a desired region in fixed time, which improves the performance of the controller.Practical implicationsThe designed controller keeps the UAH within the safe range, reduces the accident rate and improves the operation efficiency.Originality/valueThere are few research studies on the constraints and convergence time of UAH. In this paper, based on the FTDO, an adaptive NN control strategy with tan-type BLF is proposed to ensure that the state errors can be constrained to a safety region and deal with the system uncertainties and the influence of external disturbances.

This paper presents a neural network adaptive control scheme for the nonlinear systems in strict-feedback form, where the unstructured uncertainties are assumed to be unknown, though they still satisfy certain growth conditions characterized by 'bounding functions' composed of known functions multiplied by unknown constants. All adaptation laws for these unknown bounds are derived from Lyapunov based method as well as the adaptation laws for the networks' weights estimations. In addition, the unknown control gain functions are not approximated directly by neural networks. Therefore, we can avoid the possible controller singularity problems. Under a certain relaxed assumptions on the control gain functions, proposed control scheme can guarantee that all the signals in the closed-loop system are uniformly ultimately bounded (UUB). Simulation studies are included to illustrate the effectiveness of the proposed scheme, and some practical features of the control laws are also discussed.

A neural network adaptive control scheme for robot manipulators is proposed which consists of one Adaline network to identify structured system dynamics and another one to compensate for both structured and unstructured dynamic uncertainties. The former is trained offline using an LMS (least mean square) type algorithm, while the latter uses an online stable weight updating mechanism determined using Lyapunov theory. Since Adaline nets match robot regressor dynamics perfectly, the training processes of the resulting simple neural networks are computationally efficient and the proposed adaptive control scheme has very high potential in real-time applications. The proposed control scheme is illustrated by simulation and comparison studies. >

A consensus algorithm for coordinated control of unmanned surface vehicles (USVs) is studied. The USVs have different driving directions and initial velocities, so the consensus algorithm is needed to coordinate the directions and velocities of the USVs. Firstly, based on the first order linear model of USVs, get the first order nonlinear differential equations of high frequency, and the simulate environment of USVs. Then, based on the Vicsek model, the consensus algorithm in cooperative control of USVs is proposed. Aiming at the problem that USV's angular velocity lags behind the rudder angle, this paper improves the traditional PD control method, and adopts cascade PID control. So, USV can reach the same driving direction in a shorter time. Finally, simulations prove that the algorithm is feasible and compare the difference of efficiency between the PD controller and the cascade PID controller.

In this note, robust adaptive neural network (NN) control scheme is constructed for a class of unknown nonlinear systems with drift terms. The robust adaptive NN control laws are developed using backstepping technique which does not require the unknown parameters to be linear parametrizable and no regression matrices are needed. All the signals in the resulting closed-loop system are proved to be ultimately uniform bounded, and the system states are guaranteed to converge to zero.Keywordsadaptive controlbacksteppingneural networknonlinear system

In this article, a dynamic model of a six degrees of freedom (6 DOF) Eight-Rotor MAV (micro aerial vehicles) is derived on the basis of the Newton-Euler formalism.The derivation comprises determining equations of the motion of the Eight-Rotor MAV in three dimensions and approximating the actuation forces through the modelling of aerodynamic coefficients and electric motor dynamics.For Eight-Rotor MAV is inherently unstable as they are highly sensitive to external perturb, here provides an adaptive double neural network controller for the motion control of the Eight-Rotor MAV autonomous flight.This controller is developed in three parts around each of the variables.The experimental results show that the proposed adaptive double neural network controller outperforms the conventional PID controller due to its fast adaptive qualities in the presence of sensor measurement noise and the parameters variations of Eight-Rotor MAV.

In this study, a novel adaptive neural network control scheme is proposed to resolve the tracking control problem for flexible‐joint robots with random noises. More precisely, the controlled system in this study is a multi‐input and multi‐output stochastic nonlinear system, employing the traditional backstepping design to study such a system will greatly increase the amount of calculation. To resolve this problem, the command filtered technology is applied to the adaptive neural network design framework. More importantly, with the aid of the event‐triggered strategy, the proposed control algorithm can reduce the communication burden to a certain extent. Besides, the proposed method can also ensure that the tracking error converges to a small neighborhood of the origin. Finally, the simulation example is given to verify the effectiveness of the proposed algorithm.

In this paper, a novel adaptive neural network (NN) control scheme is proposed for a class of nonlinear systems with unknown control direction. By introducing some differentiable functions and high-order Lyapunov functions, the obstacle caused by unknown control direction in NN control is successfully circumvented and all closed-loop signals are shown to be uniformly bounded up to infinite time. Meanwhile, by introducing an error transformation technique, it is rigorously proved that the argument of the unknown nonlinearities remains within a compact set which can be explicitly calculated a priori, making the NN approximation always valid. Moreover, with the aid of a bound estimation approach, we effectively compress the impact of approximation errors and external disturbances and steer the tracking error into a predefined small residual set. Simulation results illustrate the effectiveness of the proposed scheme.