Solving the Problem of Approach of Controlled Objects in Dynamic Game Problems

The problem of a guaranteed result in game problems of approach of controlled objects is considered. A method for solving such problems is proposed. It involves constructing some scalar functions that qualitatively characterize the course of approach of controlled objects and the efficiency of decisions. Such functions are called resolving functions. In contrast to the main scheme of the method, the case is considered where the classical Pontryagin condition does not hold. In this situation, instead of the Pontryagin selector, which does not exist, some shift functions are considered and special multivalued mappings are introduced with their help. They generate upper and lower resolving functions, which are used to formulate the sufficient conditions for the game completion in a certain guaranteed time. An example is given to illustrate the approach of controlled objects with a simple motion, in order to obtain upper and lower resolving functions in explicit form, which allows making a conclusion about the possibility of ending the game when the Pontryagin condition does not hold.

     

不同支付函数形式对MG模型行为的影响

MG是描述多个体竞争性系统的一个结构简单但行为复杂的模型。MG模型的一个重要的机制是支付函数,它决定了对模型主体的财富和主体的策略的更新规则。该文研究了不同支付函数形式对模型的影响,发现支付函数只要是奖励赢方(少数方),惩罚输方(多数方),模型的突现行为就不会改变,与奖惩的具体形式无关,同时还发现模型突现行为有两个必要条件,一是主体必须竞争有限资源,二是在竞争中必须规定“拥挤效应”。     

Time Dilation Principle in Dynamic Game Problems

Cybernetics and Systems Analysis - The authors propose a method for solving the game problem where the trajectory of a quasi-linear non-stationary system approaches a cylindrical terminal set...     

Convex Analysis and Extremum Conditions in the Theory of Extremum Problems

A review of results and the present state of the art in the theory of necessary extremum conditions are presented. Applications of specific classes of extremum problem are considered.     

A Game Theoretic Approach to the Analysis of Dynamic Networks

A model of dynamic networks is introduced which incorporates three kinds of network changes: deletion of nodes (by faults or sabotage), restoration of nodes (by actions of “repair”), and creation of nodes (by actions that extend the network). The antagonism between the operations of deletion and restoration resp. creation is modelled by a game between the two agents “Destructor” and “Constructor”. In this framework of dynamic model-checking, we consider as specifications (“winning conditions” for Constructor) elementary requirements on connectivity of those networks which are reachable from some initial given network. We show some basic results on the (un-)decidability and hardness of dynamic model-checking problems.     

基于博弈策略强化学习的函数优化算法

该文提出了一种基于博弈论的函数优化算法。算法将优化问题的搜索空间映射为博弈的策略组合空间,优化目标函数映射为博弈的效用函数,通过博弈策略的强化学习过程智能地求解函数优化问题。文章给出了算法的形式定义及描述,然后在一组标准的函数优化测试集上进行了仿真运算,验证了算法的有效性。     

An Asymptotically Exact Approach to Robust Semidefinite Programming Problems with Function Variables

This technical note provides an approximate approach to a semidefinite programming problem with a parameter-dependent constraint and a function variable. This problem covers a variety of control problems including a robust stability/performance analysis with a parameter-dependent Lyapunov function. In the proposed approach, the original problem is approximated by a standard semidefinite programming problem through two steps: first, the function variable is approximated by a finite-dimensional variable; second, the parameter-dependent constraint is approximated by a finite number of parameter-independent constraints. Both steps produce approximation error. On the sum of these approximation errors, this technical note provides an upper bound. This bound enables quantitative analysis of the approach and gives an efficient way to reduction of the approximation error. Moreover, this technical note discusses how to verify that an optimal solution of the approximate problem is actually optimal also for the original problem.      

Adaptive Dynamic Surface Control for Spacecraft Terminal Safe Approach with Input Saturation Based on Tracking Differentiator

This thesis studies the spacecraft terminal safe approach control problem considering input saturation. Based on the spacecraft relative motion model and sphere collision avoidance potential function, an anti-saturation controller and an adaptive finite-time anti-saturation controller using dynamic surface control(DSC) are presented for the situations of known and unknown upper bound of external disturbances respectively, which can guarantee that no collisions happen in the tracking process. The second-order tracking differentiator is introduced to design the controllers, which avoids the differential of the virtual control signal and ensures the tracking performance of system output signals. Meanwhile, the auxiliary system is introduced to handle input saturation. Lyapunov stability theory is adopted to prove that the states of system under the designed controllers are uniformly ultimately bounded and practical finite-time stable respectively, and the chaser spacecraft can approach to the desired position without collision. The numerical simulation results demonstrate that the chaser spacecraft using the designed controllers can realize terminal safe approach to target spacecraft, which further illustrate the effectiveness of the proposed controllers.     

具备对讲功能的矿用广播终端系统设计

本文针对现有广播终端存在的不足,设计了一种既能支持实时广播、点对点对讲,还能实现本地和在线高品质音乐播放的广播终端。该广播终端采用集成高性能 DSP 的模拟电话适配器 KSZ8342作为核心部件,利用 MP3编解码芯片、音频功放设备、本地 SD 存储卡,实现实时广播、本地和在线音乐播放功能;另外,利用用户线接口电路芯片、外挂话柄、拨号键盘和显示屏实现点对点对讲功能。     

CMAC Neural Network and Its Use in Problems of Identification and Control of Nonlinear Dynamic Objects

Efficient use of CMAC (Cerebellar Model Articulation Controller) for identification and real-time control of nonlinear dynamical systems is demonstrated. An on-line weight training algorithm is proposed. The results of modelling and controlling nonlinear objects with unknown dynamics testify to the efficiency of this network. __________ Translated from Kibernetika i Sistemnyi Analiz, No. 5, pp. 16–28, September–October 2005.     

非完备策略的演化少数者博弈模型研究

提出了一种基于非完备策略的演化少数者博弈模型,它在演化的同时允许参与者的策略包含部分随机位,并且规定当主规则随机选择时,由次规则顶替指导。对比分析和数值模拟表明,新模型由于在策略结构上采用“缺席的等级制度”,其性能相对于普通演化MG模型有显著提升,能够以更小的记忆步长和更稳定的策略组成,进化到一个近乎理想的协作状态。     

PLC虚拟仿真被控对象的设计与实现

针对PLC传统实践教学中存在的成本高、场地有限、被控挂件需求量大等问题,基于3D Studio Max软件开发了一套虚拟仿真被控对象实验系统.该系统能够有效的解决目前实验室传统教学中的一些难题,它能将被控对象的动作直观、形象的展示给学生,激发他们的学习兴趣,有效的提高了教学效果与教学质量,从而达到我们对培养高技能人才的...     

函数对象在泛型算法中的应用

C++中的函数对象是一种特殊的对象,使用起来特别灵活。掌握并使用好函数对象可以大大精简代码。该文在介绍函数对象概念的同时总结了函数对象的几点应用。     

面向动态物体场景的视觉SLAM综述

针对当前机器人导航、自动驾驶等领域中的热点问题——面向动态物体场景的视觉SLAM(同步定位与地图构建)——进行了综述.根据动态SLAM在定位与建图时对动态物体的不同处理方式,划分了3个研究方向:动态鲁棒性SLAM与静态背景重建、非刚性动态物体跟踪重建、以及移动物体跟踪与重建.对这3个研究方向分别进行了综述,并重点介绍结合了深度学习的动态SLAM方法.最后,展望了动态SLAM的未来发展方向.     

有终端约束的调节器问题的某些渐近性质

本文研究有终端约束的调节器问题,刻划了最优轨线在终端附近的性态;给出了由此导出的反馈控制律所构成的闭环系统的特征值的渐近分布;并讨论了闭环系统的增益裕度、相位裕度和时滞容限等性质。     

游戏程序设计中若干问题的探讨

游戏编程是当今程序设计的一个热点,开发过程会遇到各种各样的问题。通过限制构造函数和拷贝函数,设计一个单体全局管理类,可以避免不合适的调用。采用分块技术加栽地图,可以在时空性能上优化游戏。阐述了D3D的空间坐标变换。分析了万向节死锁问题发生的原因,并提出了解决办法。     

竞争式模糊聚类学习在函数逼近中的应用

本文通过对RCA算法中遗忘函数的修正,抑制了类间竞争迭代中的病态发散,从而实现了算法的稳健收敛。用该算法分析数据,采用TSK模糊模型对函数进行逼近,仿真表明该方法能有效地排除噪声及孤立点对系统逼近的干扰。     

游戏程序设计中若干问题的探讨

游戏编程是当今程序设计的一个热点,开发过程会遇到各种各样的问题。通过限制构造函数和拷贝函数,设计一个单体全局管理类,可以避免不合适的调用。采用分块技术加载地图,可以在时空性能上优化游戏。阐述了D3D的空间坐标变换,分析了万向节死锁问题发生的原因,并提出了解决办法。