Sequential Inverse Optimal Control of Discrete-Time Systems

This paper presents a novel sequential inverse optimal control(SIOC)method for discrete-time systems,which calculates the unknown weight vectors of the cost function in real time using the input and output of an optimally controlled discrete-time system.The proposed method overcomes the limitations of previous approaches by eliminating the need for the invertible Jacobian assumption.It calculates the possible-solution spaces and their intersections sequentially until the dimension of the intersection space decreases to one.The remaining one-dimensional vector of the possible-solution space’s intersection represents the SIOC solution.The paper presents clear conditions for convergence and addresses the issue of noisy data by clarifying the conditions for the singular values of the matrices that relate to the possible-solution space.The effectiveness of the proposed method is demonstrated through simulation results.

Inverse Optimal Control of Evolution Systems and Its Application to Extensible and Shearable Slender Beams

An optimal(practical) stabilization problem is formulated in an inverse approach and solved for nonlinear evolution systems in Hilbert spaces. The optimal control design ensures global well-posedness and global practical K∞-exponential stability of the closed-loop system, minimizes a cost functional,which appropriately penalizes both state and control in the sense that it is positive definite(and radially unbounded) in the state and control, without having to solve a Hamilton-Jacobi-Belman equation(HJBE). The Lyapunov functional used in the control design explicitly solves a family of HJBEs. The results are applied to design inverse optimal boundary stabilization control laws for extensible and shearable slender beams governed by fully nonlinear partial differential equations.

Global Inverse Optimal Tracking Control of Underactuated Omni-directional Intelligent Navigators （ODINs）

This paper presents a design of optimal controllers with respect to a meaningful cost function to force an underactuated omni-directional intelligent navigator （ODIN） under unknown constant environmental loads to track a reference trajectory in two-dimensional space. Motivated by the vehicle＇s steering practice, the yaw angle regarded as a virtual control plus the surge thrust force are used to force the position of the vehicle to globally track its reference trajectory. The control design is based on several recent results developed for inverse optimal control and stability analysis of nonlinear systems, a new design of bounded disturbance observers, and backstepping and Lyapunov＇s direct methods. Both state- and output-feedback control designs are addressed. Simulations are included to illustrate the effectiveness of the proposed results.

A hyperchaotic system stabilization via inverse optimal control and experimental research

In this paper, some basic dynamical properties of a four-dimensional autonomous hyperchaotic system are investi- gated by means of Poincare mapping, Lyapunov exponents and bifurcation diagram. The dynamical behaviours of this new hyperchaotic system are proved not only by performing numerical simulation and brief theoretical analysis but also by conducting an electronic circuit experiment. An efficient approaching is developed for global asymptotic stabilization of this four-dimensional hyperchaotic system. Based on the method of inverse optimal control for nonlinear systems, a linear state feedback is electronically implemented. It is remarkably simple as compared with other chaos control ways, like nonlinear state feedback.

Stability in Probability and Inverse Optimal Control of Evolution Systems Driven by Levy Processes

This paper first develops a Lyapunov-type theorem to study global well-posedness(existence and uniqueness of the strong variational solution)and asymptotic stability in probability of nonlinear stochastic evolution systems(SESs)driven by a special class of Levy processes,which consist of Wiener and compensated Poisson processes.This theorem is then utilized to develop an approach to solve an inverse optimal stabilization problem for SESs driven by Levy processes.The inverse optimal control design achieves global well-posedness and global asymptotic stability of the closed-loop system,and minimizes a meaningful cost functional that penalizes both states and control.The approach does not require to solve a Hamilton-Jacobi-Bellman equation(HJBE).An optimal stabilization of the evolution of the frequency of a certain genetic character from the population is included to illustrate the theoretical developments.

基于逆模型预测控制的拟人驾驶控制

本文提出一种基于逆模型预测控制的拟人驾驶控制方法,利用模型预测控制产生的实轴轨迹与真实轨迹的损失函数更新控制模块代价函数的权重系数实现拟人化驾驶控制。将拟人驾驶控制构建成一个双层优化问题,在下层利用模型预测控制求解一个典型的最优控制问题产生实轴驾驶轨迹,在上层最小化所产生的实轴轨迹和真实驾驶轨迹的误差更新下层代价函数的权重系数,基于极大值微分原理构造辅助系统求解实轴轨迹关于代价函数权重系数的梯度。实车采集真实驾驶轨迹并进行模仿测试与泛化验证,结果表明:本文所提出的方法相比于两类基于虚轴轨迹的逆最优控制方法,在3个工况下与真实驾驶轨迹最大误差分别平均降低了73.52%和65.03%,驾驶行为更加拟人化,且具备泛化性能。

基于IoC模式的工作流与应用系统的集成

针对在应用系统中引入工作流技术进行集成的问题,提出应用控制反转模式对嵌入式工作流系统进行集成的方案,以降低集成耦合度、减少系统复杂性、提高组件重用率。避免不适当的集成方法导致应用系统代码的侵入、结构的破坏以及系统紧耦合所造成的系统灵活性和可扩性差、维护难度大的问题。给出开源工作流系统OSWorkflow和基于Java EE轻量级架构的应用系统的集成实例和实现过程。

基于零空间行为融合的多智能体编队控制综述

针对单智能体在民用和军用领域无法完成现实场景下复杂任务的问题,本文综述了零空间行为融合算法。阐述零空间行为融合算法的设计思想与数学模型,介绍零空间行为融合算法的文献计量分析。综述了零空间算法的研究进展,包括编队速度饱和、行为转换不连续、行为优先级分配和多种算法结合方面的改进工作,并分析当前研究存在的问题。研究表明:零空间行为融合算法在多智能体编队控制领域,提高算法的自适应性、考虑动力学特性、异构编队和行为函数优化是该算法未来的发展方向。

基于Spring框架的IoC微内核的实现机制与应用

通过模拟实现Spring的IoC容器服务,揭示了容器的微内核实现机制,利用反射原理,编写且分析了Spring用依赖注入(Dependency Injection)实现IoC服务的过程。

基于IABAP混合算法的混凝土坝浇筑仓温控措施智能优选

混凝土坝浇筑仓在施工期的温控措施选择不当会导致混凝土出现温度裂缝,将智能优化算法与温控措施结合进行多个温控因素联合分析,可实现大坝温控措施智能调控。提出了一种结合粒子群(PSO)与人工峰群(ABC)的混合优化算法IABAP,避免了单一算法后期收敛速度降低、易陷入局部最优等问题,并将其运用于白鹤滩拱坝典型坝段的参数反演,基于计算温度曲线与实测温度曲线对比结果,进行温控措施联合优选。结果表明,通过IABAP算法温控措施调整后的温度曲线预测结果与施工期采用实际温控措施参数值预测结果相比温升速率和浇筑温度更小,对控制混凝土早期拉应力更有利。研究结果为混合优化算法在大坝温控措施智能优选方面的运用提供了参考。

集成型车载充电系统并网模式模型预测控制策略

相较于传统车载充电系统,集成型车载充电系统(integrated onboard charger system,IOCS)在成本、功率密度等方面具备显著优势。文中基于六相永磁电驱系统设计了一台IOCS,并研究了模型预测电流控制(model predictive current control,MPCC)算法在该系统并网模式下的应用。首先,分析所提IOCS的电路拓扑并建立数学模型,同时介绍传统MPCC的实施流程。然后,针对传统MPCC计算量大、稳态性能差等不足,提出一种基于占空比优化的MPCC(MPCC based on duty cycle optimization,DCO-MPCC)策略。一方面,减少备选电压矢量数量,降低电流预测环节带来的计算负担;另一方面,提出一种占空比优化技术,改善系统稳态性能。最后,通过实验验证了所提算法的有效性与优越性。实验结果表明,DCO-MPCC策略能够显著提升系统稳态性能并减少算法计算量。充电与车网互动(vehicle to grid,V2G)工况下,网侧电流总谐波畸变(total harmonic distortion,THD)分别降低6.18%与5.92%,算法运行时间减少17.54μs。

NEURAL NETWORK REALIZATION OF STRUCTURAL VIBRATION CONTROL USING MR DAMPER

Magnetorheological (MR) dampers are one of the most promising new devices for civil infrastructural vibration control applications. However, due to their highly nonlinear dynamic behavior, it is very difficult to obtain of a mathematical model of inverse MR damper that has an explicit relationship between the desired damper force and the command signal (voltage). This force voltage relationship is especially required for the structural vibration control design and simulation using MR dampers. This paper focuses on using a neural network (NN) technique to emulate the inverse MR damper model. The output of the neural network can be used to command the MR damper for generating desired forces. Numerical simulations are also presented to illustrate the effectiveness of this inverse model in semi active vibration control using MR dampers.

Spring MVC设计中的IoC控制

控制反转功能有效地利用IoC容器实现了Bean的装配与控制,统一完成组件的实例化工作并管理丰富的逻辑组件资源。在Spring MVC设计中,通过外部容器获得控制层必须的逻辑组件对象,然后完成逻辑处理业务,实现了控制层、逻辑层及数据层的充分解耦,大大提高了开发效率和管理效率。理解并掌握IoC容器(即配置文件)的设计和应用,是实现Spring MVC架构的关键环节。

基于输出反馈逆强化Q学习的线性二次型最优控制方法

本文针对模型参数未知且状态不可测的线性离散系统的线性二次型最优控制问题,提出了一种数据驱动的基于输出反馈逆强化Q学习的最优控制方法,利用系统的输入输出数据同时确定合适的二次型性能指标权重和最优控制律,使得系统运行轨迹与参考轨迹一致.本文首先提出一个参数矫正方程并与逆最优控制相结合得到一种基于模型的逆强化学习最优控制框架,实现输出反馈控制律参数和性能指标加权项的矫正.在此基础上,本文引入强化Q学习思想提出了数据驱动的输出反馈逆强化Q学习最优控制方法,无需知道系统模型参数,仅利用历史输入输出数据对输出反馈控制律参数和性能指标加权项进行求解.理论分析与仿真实验验证了所提方法的有效性.

一类带积分型源项的对流-扩散方程的初值反演问题

主要讨论了一类带有积分型源项的对流-扩散方程的初值反演问题.在最优化控制理论框架下,证明了控制泛函最优解的存在性,得到了最优解满足的必要条件.利用必要条件,通过一些正问题的先验估计证明了最优解具有非局部唯一性和稳定性.

IoC模式及其在设备监控软件中的应用

控制反转(Inversion of Control,IoC)模式是一种框架设计模式,目的在于提高应用软件的可复用性和可扩展性。结合抽象模型设计思想,针对卫星地面应用系统的设备监控系统,举例对比了如何运用传统方法和IoC模式进行软件的设计和开发。结果表明,软件开发中使用IoC模式可以有效地降低组件之间的耦合,使软件框架更加易于扩展。

Minimum Time Overtaking Problem of Vehicle Handling Inverse Dynamics Based on Two Kinds of Safe Distances

Overtaking accidents caused by improper operations performed by a driver occur frequently. However, most stud?ies on overtaking safety have neglected research into driver control input. A novel method is proposed to obtain the driver control input during the overtaking process. Meanwhile, to improve the safety of overtaking, two types of safe distances, and the time of the overtaking are considered. Path constraints are established when considering the two types of safe distances. An optimal control model is established to solve the minimum time maneuver under multiple constraints. Using the Gauss pseudospectral method, the optimal control problem is converted into a nonlinear pro?gramming problem, which is then solved through sequential quadratic programming(SQP). In addition, the e ective?ness of the proposed method is verified based on the results of a Carsim simulation. The simulation results show that by adopting an inverse dynamics method to solve the manipulation problem of the vehicle’s minimum overtaking time, the manipulation capability of a vehicle in completing an overtaking safely within the minimum time can be obtained. This method can provide a reference for research into the active safety of manned and unmanned vehicles.

Decoupling control and zero dynamics stabilization for shunt hybrid active power filter

An inverse system method based optimal control strategy was proposed for the shunt hybrid active power filter （SHAPF） to enhance its harmonic elimination performance. Based on the inverse system method, the d-axis and q-axis current dynamics of the SHAPF system were decoupled and linearized into two pseudolinear subsystems. Then, an optimal feedback controUer was designed for the pseudolinear system, and the stability condition of the resulting zero dynamics was presented. Under the control strategy, the current dynamics can asymptotically converge to their reference states and the zero dynamics can be bounded. Simulation results show that the proposed control strategy is robust against load variations and system parameter mismatches, its steady-state performance is better than that of the traditional linear control strategy.

Robust Adaptive Control of PVTOL System via Minimum Projection Method

＂Dynamic extension＂ is commonly used for stabilization of the planar vertical take off and landing （PVTOL） system. Most controllers designed by the method are based on ＂dynamic＂ control Lyapunov functions （CLFs）. We design a C^∞ differentiable ＂static＂ CLF for the PVTOL system by dynamic extension and minimum projection method. Then we propose an inverse optimal controller based on the static CLF that attains a gain margin. We design an adaptive control input and show the robustness of the controller by computer simulation.

逆向强化学习研究综述

逆向强化学习(inverse reinforcement learning,IRL)也称为逆向最优控制(inverse optimal control,IOC),是强化学习和模仿学习领域的一种重要研究方法,该方法通过专家样本求解奖赏函数,并根据所得奖赏函数求解最优策略,以达到模仿专家策略的目的.近年来,逆向强化学习在模仿学习领域取得了丰富的研究成果,已广泛应用于汽车导航、路径推荐和机器人最优控制等问题中.首先介绍逆向强化学习理论基础,然后从奖赏函数构建方式出发,讨论分析基于线性奖赏函数和非线性奖赏函数的逆向强化学习算法,包括最大边际逆向强化学习算法、最大熵逆向强化学习算法、最大熵深度逆向强化学习算法和生成对抗模仿学习等.随后从逆向强化学习领域的前沿研究方向进行综述,比较和分析该领域代表性算法,包括状态动作信息不完全逆向强化学习、多智能体逆向强化学习、示范样本非最优逆向强化学习和指导逆向强化学习等.最后总结分析当前存在的关键问题,并从理论和应用方面探讨未来的发展方向.