Filtering adaptive neural network controller for multivariable nonlinear systems with mismatched uncertainties

This paper synthesizes a filtering adaptive neural network controller for multivariable nonlinear systems with mismatched uncertainties. The multivariable nonlinear systems under consideration have both matched and mismatched uncertainties, which satisfy the semiglobal Lipschitz condition. The nonlinear uncertainties are approximated by a Gaussian radial basis function (GRBF)‐based neural network incorporated with a piecewise constant adaptive law, where the adaptive law will generate adaptive parameters by solving the error dynamics between the real system and the state predictor with the neglection of unknowns. The combination of GRBF‐based neural network and piecewise constant adaptive law relaxes hardware limitations (CPU). A filtering control law is designed to handle the nonlinear uncertainties and deliver a good tracking performance with guaranteed robustness. The matched uncertainties are cancelled directly by adopting their opposite in the control signal, whereas a dynamic inversion of the system is required to eliminate the effect of the mismatched uncertainties on the output. Since the virtual reference system defines the best performance that can be achieved by the closed‐loop system, the uniform performance bounds are derived for the states and control signals via comparison. To validate the theoretical findings, comparisons between the model reference adaptive control method and the proposed filtering adaptive neural network control architecture with the implementation of different sampling time are carried out.     

显式抑制测量误差的改进滑模控制

反馈控制系统中模型不确定性和测量误差的同时出现, 给高精度控制器的设计带来挑战. 经典滑模控制能抵抗一定程度的模型不确定性和输入干扰, 但引入的高增益使得其性能对测量噪声极为敏感, 也容易引起系统强烈抖振. 为此, 本文针对一种典型的角度和角速率测量分别包含高频和低频测量误差条件, 提出了一种改进的基于趋近律的角度跟踪控制方案. 本方案采用低通滤波器来应对高频测量噪声, 同时采用一种新颖的基于模型的测量误差估计器, 来补偿低频测量漂移. 采用Quanser Aero平台进行两自由度角轨迹跟踪控制仿真和实验验证, 并与自抗扰控制等几种典型鲁棒控制方案进行了全面对比, 证实了本文方案性能上的优越性和调参的便捷性.     

事件触发下随机非确定线性多智能体的指数同步

研究随机非确定线性多智能体系统在有向拓扑连接下的指数同步问题,为减少不必要的网络带宽资源的浪费,提出一种基于事件触发控制的协议。根据组合测量对系统中的所有节点设计相应的事件触发函数,使得节点之间的控制信号更新仅在事件触发时刻进行。基于Lyapunov稳定性理论和M矩阵理论,得到了多智能体系统指数同步结论,并给出了同步的收敛速度。同时,理论排除了事件触发控制过程中的芝诺（Zeno）现象。数值仿真结果进一步验证了理论分析的有效性。     

FTESO‐Based Finite Time Control for Underactuated System Within a Bounded Input

Finite time control problem is investigated for a class of underactuated systems with uncertainties and external disturbances. For the sake of expanding control region furthest within a bound input, finite time extended state observer (FTESO) and a novel adaptive terminal sliding mode (ATSM) controller are applied to improve the stability performance of system. Compared to the general extended state observer (ESO), FTESO makes use of fractional powers to reduce the estimation errors to zero in finite time. The coordinate transformation is made for more degrees of design freedom. Rigorous analysis of finite time convergence results has been performed through Lyapunov theory and sufficient conditions are provided for the observer/controller‐design. Finally, simulation results on the Rotating Inverted Pendulum are given to demonstrate the effectiveness of the proposed controller and observer.     

Global output‐feedback stabilization for nonlinear systems with unknown relative degree

This paper is devoted to the global stabilization via output feedback for a class of nonlinear systems with unknown relative degree, dynamics uncertainties, unknown control direction, and nonparametric uncertain nonlinearities. In particular, the unknown relative degree is without known upper bound, which renders us to research for a filter with varying dimension rather than the ones with over dimensions in the existing literature. In comparison with more popular but a bit stronger input‐to‐state stable or input‐to‐state practically stable requirement, only bounded‐input bounded‐state stable requirement is imposed on the dynamics uncertainties, which affect the systems in a persistent intensity rather than in a decaying one. In this paper, to compensate multiple serious system uncertainties and realize global output‐feedback stabilization, a design scheme via switching logic together with varying dimensional filter is developed. In this scheme, 2 switching sequences, which separately generate the gains of the controller and act as the varying dimensions of the filter, are designed to overcome unknown control direction, dynamics uncertainties and nonparametric uncertain nonlinearities, and unknown relative degree, respectively. A 2‐mass lumped‐parameter structure is provided to show the effectiveness of the proposed method in this paper.     

Neural network–based reconfiguration control for spacecraft formation in obstacle environments

This paper proposes an adaptive formation reconfiguration control scheme based on the leader‐follower strategy for multiple spacecraft systems. By taking the predesigned desired velocities and the trajectories as reference signals, a set of coordination tracking controllers is constructed by combining the reconstructed dynamic system and the neural network–based reconfiguration algorithm together. To avoid collisions between spacecraft and obstacles during the formation configuration process, the null space–based behavioral control is integrated into the control design. Since the spacecraft dynamics contains unknown nonlinearity and disturbance, it is challenging to make the system robust to uncertainties and improve the control precision simultaneously. To solve this problem, both the adaptive neural network strategy and the finite‐time control theory are employed. Finally, 2 simulation examples are carried out to verify the proposed algorithm, showing that the formation reconfiguration task can be executed successfully while achieving high control precision.     

Output feedback semiglobal practical consensus of linear systems with relative state‐dependent uncertainties

This paper proposes an output feedback consensus control law for linear multiagent systems with relative state‐dependent uncertainties. To achieve output feedback control and uncertainty attenuation, the theories of extended high‐gain observer and structural decomposition are employed. Under the proposed control scheme, semiglobal practical consensus is achieved in the sense that the synchronization errors are ultimately bounded. Besides, the synchronization errors can be kept arbitrarily small by a proper choice of tuning parameters. Finally, a simulation example is provided to verify the theoretical results.     

Robust Formation Maneuvers Through Sliding Mode for Multi-agent Systems With Uncertainties

This paper develops a robust control method for formation maneuvers of a multi-agent system. The multi-agent system is leader-follower-based, where the graph theory is utilized to describe the information exchange among the agents. The control method is exercised via sliding mode methodology where each agent is subjected to uncertainties. The technique of nonlinear disturbance observer is adopted in order to overcome the adverse effects of the uncertainties. Assuming that the uncertainties have an unknown bound, the formation stability conditions are investigated according to a given communication topology. In the sense of Lyapunov, not only the formation maneuvers of the multi-agent system have guaranteed stability, but the desired formations of the agents are also realized. Compared with other two control approaches, i.e., the basic sliding mode approach and the fuzzy sliding mode approach, some numerical results are presented to illustrate the effectiveness, performance and validity of the robust control method for formation maneuvers in the presence of uncertainties.      

Robust state estimation for singularly perturbed systems

This paper deals with the design of interval observers for singularly perturbed linear systems. The full-order system is first decoupled into slow and fast subsystems. Then, using the cooperativity theory, an interval observer is designed for the slow and fast subsystems assuming that the measurement noise and the disturbances are bounded and the singular perturbed parameter is uncertain. This decoupling leads to two observers that estimate the lower and upper bounds for the feasible state domain. A numerical example shows the efficiency of the proposed technique.     

Fault-tolerant control for uncertain linear systems via adaptive and LMI approaches

This paper proposes a fault-tolerant control scheme for linear systems with mismatched uncertainties which are assumed to be norm-bounded, affine and polytopic, respectively. The linear fractional transformation (LFT) and linear matrix inequality (LMI) techniques are introduced to handle the mismatched uncertainties, and the adaptive techniques are used to compensate actuator faults. By using the cone complementary linearisation algorithm, the resulting stability criteria are converted into solvable ones. Then, on the basis of Lyapunov stability theory, it is shown that the solutions to the closed-loop system and error system are uniformly bounded, especially, the states converge asymptotically to zero. Finally, simulations are given to illustrate the effectiveness and advantages of the proposed theoretical results.