马尔可夫跳变线性系统最优控制的研究现状与进展

马尔可夫跳变线性系统（MJLS）是一种具有多个模态的随机系统，系统在各个模态之间的跳变转移由一组马尔可夫链来决定。MJLS模型因其在表示过程中可以产生突变而更能精确的描述实际工程应用中的系统。近年来，MJLS的最优控制问题成为了研究的热点，动态规划、极大值原理以及线性矩阵不等式等成为了解决此类问题的主流方法。本文对MJLS最优控制领域的研究现状进行了综述。分别对一般情况下、带有噪声的情况下、带有时滞的情况下以及某些特定情况下的MLJS最优控制问题的国内外研究现状进行论述。最后进行了总结并提出MJLS最优控制领域未来值得关注的研究方向。     

基于神经网络算法的井下裂缝诊断与堵漏技术

在钻井过程中，常常钻遇不同宽度的井下地层裂缝。钻遇裂缝时容易发生钻井液漏失现象，甚至发生钻井液失返现象，严重影响了安全、高效钻井。目前裂缝封堵的方法常存在封堵成功率不高、堵漏承压能力低的问题，其中一个重要的原因是对井下地层的裂缝宽度等特征认识不清。基于地层裂缝产生的岩石力学机理，确定影响裂缝宽度关键的6个力学和工程因素，并利用神经网络计算的非线性、大数据特点建立了井下地层裂缝宽度的分析模型，模型包含输入层、输出层和3个隐藏层。通过该模型诊断井下裂缝宽度，提高了计算精度，平均误差仅为2.09%，最大误差为5.88%，解决钻井现场仅凭经验判断裂缝误差较大和依靠成像测井成本较高的问题。同时根据神经网络模型诊断得到的裂缝宽度优化堵漏材料的粒径配比，提高了裂缝内的架桥封堵强度和架桥的稳定性，封堵层的承压能力达到12.8 MPa，反向承压能力达到4.5 MPa。现场堵漏试验最高憋压10 MPa，经过封堵作业后大排量循环不漏，达到了裂缝性地层高效堵漏的目的，堵漏一次成功。      

基于时频域的卷积神经网络运动想象脑电信号识别方法

针对目前运动想象脑电（EEG）信号识别率较低的问题，考虑到脑电信号蕴含着丰富的时频信息，提出一种基于时频域的卷积神经网络（CNN）运动想象脑电信号识别方法。首先，利用短时傅里叶变换（STFT）对脑电信号的相关频带进行预处理，并将多个电极的时频图组合构造出一种二维时频图；然后，针对二维时频图的时频特性，通过一维卷积的方法设计了一种新颖的CNN结构；最后，通过支持向量机（SVM）对CNN提取的特征进行分类。基于BCI数据集的实验结果表明，所提方法的平均识别率为86.5%，优于其他传统运动想象脑电信号识别方法；同时将该方法应用在智能轮椅上，验证了其有效性。     

非全日制研究生线上线下混合式教学模式研究 ——以《现代电气控制技术》为例

随着教育信息化技术的发展，线上线下混合式教学模式已成为一种趋势。为了解决非全日制研究生课堂教学召集困难、效果不佳等问题，提出了基于“互联网＋虚拟仿真技术”的线上线下混合式教学模式。本文以控制工程专业学位研究生为例，结合《现代电气控制技术》课程，探讨了线上线下混合式教学实现途径、师生互动方法、项目驱动案例教学法以及电气控制系统虚拟仿真实验平台的构建方法，切实提升非全日制研究生的培养质量。     

Event-triggered distributed receding horizon control for tracking and formation of homogeneous multi-agent systems

This paper proposes an event-triggered distributed receding horizon control (DRHC) approach for the formation and tracking problems of homogeneous multi-agent systems. For each agent, an event-triggering condition, based on assumed predictive information of the neighbours, is derived from stability analysis. Considering the uncertain deviation between the assumed and true predictive information, we design a time-varying compatibility constraint for the individual optimization problem. In the event-triggered DRHC algorithm, each agent solves the optimization problem and communicates with its neighbours only when the event-triggering condition is satisfied, so the communication and computation burden are reduced. Moreover, guarantees for the recursive feasibility and asymptotic stability of the overall system are proved. A simulation example is provided to illustrate effectiveness of the proposed approach.     

Pressure control of an electro-hydraulic actuated clutch via novel hysteresis model

This paper mainly focuses on the development of pressure tracking control logic of electro-hydraulic actuators for vehicle application. This is done to improve and ensure the performance of a precise lower-level controller for evolving modern shift control logic. The required performance is obtained by hysteresis model-based feed-forward control and additional feedback control. The hysteresis and the time delay, which adversely affect pressure control, are well known nonlinear behaviors in electro-hydraulic actuators. In order to cope with the hysteresis, a novel hysteresis model is proposed based on a physical phenomenon. A mathematical model based on a characteristic curve obtained in preliminary experiments is presented using only one tuning parameter, and this model can be inverted easily to construct a feed-forward controller. In addition, a feedback controller is designed considering the stability margin of a time delay system. The feedback control inputs ensure compensation of the feed-forward errors caused by model error and uncertainty. The proposed controller is designed to lower computational cost considering applicability for production vehicles. As a result, the developed pressure controller is applied to a transmission control unit of a production vehicle and verified experimentally for various driving scenarios.     

Adaptive compensation of persistent actuator failures of nonlinear systems

In this article, adaptive compensation designs are developed for nonlinear systems with uncertainties from the system functions and persistent actuator failures of characterizations that (i) some unknown system inputs are stuck at some unknown fixed or varying values at unknown time instants and (ii) the failure pattern always switches from one to another and the switching does not stop. Such a controlled plant is described by an uncertain time-varying nonlinear system, and some robust adaptive feedback linearization based failure compensation results are studied for closed-loop system stabilization and bounded output tracking for some specific conditions. To improve the tracking performance in the presence of persistent actuator failures, a new adaptive control scheme is developed, using the failure indicator function which contains the failure pattern and failure time in the formulation. Detailed stability and tracking performance are shown. Simulation results are shown to verify the effectiveness of the proposed adaptive actuator failure compensation method.     

Semirigid inclusions in elastic bodies: Mechanical interplay and optimal control

The paper concerns an analysis of an equilibrium problem for 2D elastic body with two semirigid inclusions. It is assumed that inclusions have a joint point, and we investigate a junction problem for these inclusions. The existence of solutions is proved, and different equivalent formulations of the problem are proposed. We investigate a convergence to infinity of a rigidity parameter of the semirigid inclusion. It is proved that in the limit, we obtain an equilibrium problem for the elastic body with a rigid inclusion and a semirigid one. A parameter identification problem is investigated. In particular, the existence of a solution to a suitable optimal control problem is proved.     

On one boundary problem for nonlinear stationary controlled system

The object of study is nonlinear stationary controlled system of ordinary differential equations with constant disturbance in the right part. The problem of constructing the synthesising control function providing the transfer of this system from the initial state to the origin is considered. The sufficiently simple for numerical implementation algorithm of solution of the above-mentioned problem is obtained. It is shown that for local null controllability of the considered system, it is sufficient that the conditions of the Kalman's type were satisfied. In addition, the estimates restricting the choice of initial conditions and external disturbances under which the transfer is guaranteed are obtained. The main idea of the method of construction of the desired control function consists in reducing the original problem to stabilisation of a special kind linear non-stationary system and solving the Cauchy problem for an auxiliary system of ordinary differential equations closed by stabilising control. The simplicity of the realisation of this algorithm is determined by the construction of the auxiliary system and its stabilisation that could be obtained by analytical methods. The effectiveness of the method is illustrated by solving the problem of crane control and its numerical simulation.     

Integration of imitation learning using GAIL and reinforcement learning using task-achievement rewards via probabilistic graphical model

The integration of reinforcement learning (RL) and imitation learning (IL) is an important problem that has long been studied in the field of intelligent robotics. RL optimizes policies to maximize the cumulative reward, whereas IL attempts to extract general knowledge about the trajectories demonstrated by experts, i.e, demonstrators. Because each has its own drawbacks, many methods combining them and compensating for each set of drawbacks have been explored thus far. However, many of these methods are heuristic and do not have a solid theoretical basis. This paper presents a new theory for integrating RL and IL by extending the probabilistic graphical model (PGM) framework for RL, control as inference. We develop a new PGM for RL with multiple types of rewards, called probabilistic graphical model for Markov decision processes with multiple optimality emissions (pMDP-MO). Furthermore, we demonstrate that the integrated learning method of RL and IL can be formulated as a probabilistic inference of policies on pMDP-MO by considering the discriminator in generative adversarial imitation learning (GAIL) as an additional optimality emission. We adapt the GAIL and task-achievement reward to our proposed framework, achieving significantly better performance than policies trained with baseline methods.