一种基于控制流解耦的可重构阵列动态调度方法

可重构阵列依靠数据流驱动带来的能效优势,被广泛运用在特定领域的运算加速中.随着应用范围的增大,当应用中存在不同控制流区域时,采用传统的空间调度方案同时执行整个数据流图,会由于非一致性控制流的存在,造成严重的性能损失.本文提出一种基于控制流解耦的调度方法,通过将处于不同控制边界的数据流解耦成若干个相互独立的子图交替执行,...     

基于MADDPG的多无人机协同任务决策

针对多无人机任务决策方法研究中传统优化算法难以在短时间内得到期望结果的问题,基于深度强化学习提出一种无人机多智能体深度确定性策略梯度(MADDPG)算法,通过允许无人机在学习时使用全局信息,在应用决策的时候只使用局部信息的方法,从网络结构、状态空间、动作空间和奖励函数设计了 MADDPG算法的模型结构.最后通过仿真实验...     

相控阵雷达长时跟踪波束调度与波形优化策略

针对相控阵雷达多目标跟踪波束调度和波形参数优化控制的问题，本文提出了一种基于马尔可夫决策过程（MDP）的相控阵雷达跟踪波束调度与波形参数优化策略，该方法以无迹卡尔曼滤波（UKF）算法为基础来估计目标的状态。首先将本文的序列决策问题建模为马尔可夫决策过程，定义了资源的效费比和长期回报率，然后与当前实际跟踪误差综合考虑作为MDP的回报函数，进而给出了调度的优化模型，最后将长时决策问题转化为动态规划算法结构进行求解，并且提出了一种并行混合遗传粒子群优化算法来求解各决策时刻的最优策略。仿真结果表明了长时策略的先进性以及寻优算法的优越性，与传统的短时策略相比，跟踪精度可提高11.17%。     

Task allocation strategies for cooperative task planning of multi-autonomous satellite constellation

In recent years, the application of imaging satellites has entered a completely new stage, with the new demands such as rapid response to emergency events, observation of large-scale regional targets and multivariate data fusion, the multi-autonomous satellite constellation has been proposed. This paper first designs the structure of multi-autonomous satellite constellation, and a centralized-distributed structure is proposed. This structure could improve the dynamic response capacity of the whole constellation. Then, this paper adds adapted filtering mechanism to single autonomous satellite online scheduling algorithm to enhance its performance. This article also pays more attention on task allocation strategies of the master satellite in constellation, and ten different task allocation strategies based on five dimensions are analyzed by simulation experiments. At last, this paper extracts several characteristic features of the regular observation targets and designs a selector based on support vector machine (SVM). This selector could select an appropriate strategy according to the features of each experiment scenario.     

A multi-objective binary-encoding differential evolution algorithm for proactive scheduling of agile earth observation satellites

A robust model and a Multi-objective Binary-encoding Differential Evolution (MBDE) algorithm are proposed in this paper for agile earth observation satellite proactive scheduling considering satellite resource failure and emergency task insertion. Firstly, considering that the reserved slack time in schedule can absorb uncertainties, two indicators are proposed: schedule profits and slack time, based on which, a robust model of proactive scheduling is established. Secondly, to solve the multi-objective model, solutions are represented via binary-encoding, mutation, crossover, and selection operators are re-designed, besides, an external archive update strategy is adopted to store elitist solutions in the evolution process, moreover, non-dominated Pareto set evaluation metrics are improved. Finally, computational results have demonstrated that, compared with NSGA-II and SPEA-2, the MBDE algorithm is able to obtain well-distributed solutions with good convergence more efficiently. The study can provide the method support for proactive scheduling of agile earth observation satellites.     

任务表集群计算平台的实时性研究

实时计算系统的关键特征包含计算时限保证和可信度两个方面。本文从进程配置、负载分配、消息传递等方面介绍任务表算法集群平台的计算时间保证措施;从瞬间故障表现和长期运行表现介绍平台在实时应用中的可信度措施。     

飞机排班系统在航空公司生产运行领域中的应用和探讨

宏观介绍了航空公司生产运行领域飞机排班系统的主要功能组成和使用情况及其与相关系统之间的关联关系，并结合航空公司的未来发展趋势，提出了系统建设方面的一些思路和设想，为生产运行系统的持续改进提供一定的借鉴和参考。     

基于波位聚类的相控阵雷达群目标调度模型

如何有效分配有限的能量和时间资源,从而实现大量密集群目标的跟踪测量,是相控阵雷达资源调度面临的难点问题。在采用基于最小期望时间和时间槽优先占用的自适应资源调度模型的基础上,论述了基于波位动态聚类及分区识别算法的相控阵雷达群目标调度模型,详细描述了群目标调度流程图。群目标调度模型根据距离门宽度、方位俯仰波束宽度、目标速度、航迹预测误差和点迹测量误差等确定群目标的分区位置及大小,自动实现群目标的聚类和跟踪。仿真分析表明:该模型能够适应目标群的动态演化过程,实现群内外所有目标的独立跟踪,有效节约群目标跟踪时的资源开销,并已在实际工程应用中得到了验证。     

面向无人机蜂群的航电云多层任务调度模型

在航空作战体系中，基于航电云的无人机（UAV）蜂群作战是提高未来无人机综合作战能力的一种新模式。针对无人机蜂群作战的航电云架构，如何将云端作战任务派发到无人机且保证作战任务完成时间是其中关键。在无人机蜂群分层分簇网络结构和模块级资源虚拟化的基础上，对传统单层平台级任务调度模型进行改进，提出了一种细化到模块级的多层任务调度模型，将作战任务从云端逐层调度到无人机功能模块上执行。利用OMNeT++对无人机蜂群多层任务调度模型以及传统的单层任务调度模型分别进行仿真，云端以攻击使命组为例构建使命组集进行分配，并对任务吞吐量、消息平均端到端延时和任务完成时间进行性能对比。仿真结果表明：与平台级单层任务调度相比，在执行任务方面，模块级多层任务调度模型将单个任务平均完成时间降低了46.2%，将使命组完成时间降低了52.1%，在保证任务吞吐量的基础上具有对复杂任务更稳定的调度能力；在网络性能方面，模块级多层任务调度模型消息端到端延时更低，延时分布更集中，提高了网络消息传输的实时性。     

Satellite range scheduling with the priority constraint: An improved genetic algorithm using a station ID encoding method

Satellite range scheduling with the priority constraint is one of the most important problems in the field of satellite operation.This paper proposes a station coding based genetic algorithm to solve this problem,which adopts a new chromosome encoding method that arranges tasks according to the ground station ID.The new encoding method contributes to reducing the complexity in conflict checking and resolving,and helps to improve the ability to find optimal resolutions.Three different selection operators are designed to match the new encoding strategy,namely random selection,greedy selection,and roulette selection.To demonstrate the benefits of the improved genetic algorithm,a basic genetic algorithm is designed in which two cross operators are presented,a single-point crossover and a multi-point crossover.For the purpose of algorithm test and analysis,a problem-generating program is designed,which can simulate problems by modeling features encountered in real-world problems.Based on the problem generator,computational results and analysis are made and illustrated for the scheduling of multiple ground stations.