A hybrid task scheduling scheme for multicore processors based on improving the EDF algorithm

为解决多核处理器系统中的实时任务调度问题,尤其是实时任务和非实时任务的混合调度问题,在对最早截止时间优先(EDF)算法进行改进的基础上,提出多核处理器混合任务调度算法——EDF-segment算法.EDF-segment算法可以整理调度混合任务时出现的碎片,并通过对碎片的迁移、合并提高处理器的利用率,从而提高系统处理混合任务的性能.通过EDF-segment算法不但可以解决混合任务的调度问题,还可以避免使用EDF算法时造成的多核处理器利用率下降,在保证实时任务处理延迟的前提下提升多核处理器的利用率.经过理论推导和实验分析证明,EDF-segment算法可以有效地应用于多核处理器系统中.     

An efficient task scheduling algorithm for heterogeneous multi-core processors

针对现有异构多核处理器任务调度算法效率低的问题,提出了一种综合性的、高效的静态任务调度算法,即聚簇与复制列表优化调度(CDLOS)算法.该算法首先通过对任务图进行聚簇优化,降低某些特殊任务的通信开销;然后从整个任务图的拓扑结构出发计算任务的优先级权值,提高关键任务的优先级;继而采用区间插入和任务复制技术进行调度,降低处理器资源浪费;最后通过优化调度结果,消除冗余任务,减小整个任务的调度长度.实例分析和模拟实验结果表明:与以往算法相比,此新算法较高地提升了多核处理器任务调度的效率,具有更好的应用前景.     

滤光片表面缺陷视觉检测系统多处理器调度优化

针对滤光片表面缺陷视觉检测系统中在线检测实时性需求对检测速度要求较高,研究一种有效利用可用硬件资源并行处理实时工作提高处理速度的调度优化策略。基于AOE图对滤光片表面缺陷视觉检测系统进行任务级分析,优化事件、活动拓扑关系与任务间冗余的数据相关性、资源相关性,建立并行任务模型;采用关联处理器调度算法(arbitrary processor affinities,APAs)进行并行多处理器调度,指定任务只能被某个处理器集合执行,将期限紧迫、缓存敏感的任务限制在单一处理器,提高资源利用率,改进检测系统实时性。试验结果表明:在尺寸为1.20mm×1.20mm、26×28个滤光片组成滤光片面板上,采用多处理器调度可使检测速度极大提升,采用APAs调度算法后,平均缺陷识别完成时间为常规检测系统时间的36.5%,可以满足在线实时要求,证明应用多处理器调度方法,可以极大提升检测仪器实时性能的有效性。     

在Linux下改进EDF实时调度算法

EDF(Earliest Deadline First)算法由于CPU利用率高,可调度的任务集较大等优点在实时系统中的得到广泛的应用,但是EDF算法无法保证重要任务的执行。本文在EDF算法的基础上结合任务的重要性,提出了SBID(Scheduler Based Importance and Deadline)调度算法。首先从理论的角度分析了SBID算法的可调度性,并针对SBID算法对Linux2.6系统内核作一些修改。然后通过实验数据对比表明了SBID算法在保证重要任务顺利运行方面具有较好的优越性。     

基于空闲时间分配的电压调节策略

为了求解考虑系统能耗的实时动态电压调节（DVS）问题,提出了基于空闲时间分配算法（STDA）的DVS策略,该策略以时间片为单位逐步把所有空闲时间分配给各个任务,且每个时间片都被分配给产生能量减少量最大的任务。分析指出,当时间片大小设置合适时,STDA算法的能耗接近于DVS问题的最小能耗。仿真实验结果表明,对于仅考虑处理器能耗的DVS问题,STDA算法的能耗和最优策略的能耗相近;而对于考虑系统能耗的DVS问题,STDA算法减少能耗的效果要优于其他现有DVS策略。     

在Linux下改进EDF实时调度，

EDF（Earliest Deadline First）算法由于CPU利用率高，可调度的任务集较大等优点在实时系统中的得到广泛的应用，但是EDF算法无法保证重要任务的执行。本文在EDF算法的基础上结合任务的重要性，提出了SBID（Scheduler Based Importance and Deadline）调度算法。首先从理论的角度分析了SBID算法的可调度性，并针对SBID算法对Linux2．6系统内核作一些修改。然后通过实验数据对比表明了SBID算法在保证重要任务顺利运行方面具有较好的优越性。     

动态场景下基于YOLOv5和几何约束的视觉SLAM算法

目的 移动智能体在执行同步定位与地图构建(Simultaneous Localization and Mapping,SLAM)的复杂任务时,动态物体的干扰会导致特征点间的关联减弱,系统定位精度下降,为此提出一种面向室内动态场景下基于YOLOv5和几何约束的视觉SLAM算法。方法 首先,以YOLOv5s为基础,将原有的CSPDarknet主干网络替换成轻量级的MobileNetV3网络,可以减少参数、加快运行速度,同时与ORB-SLAM2系统相结合,在提取ORB特征点的同时获取语义信息,并剔除先验的动态特征点。然后,结合光流法和对极几何约束对可能残存的动态特征点进一步剔除。最后,仅用静态特征点对相机位姿进行估计。结果 在TUM数据集上的实验结果表明,与ORB-SLAM2相比,在高动态序列下的ATE和RPE都减少了90%以上,与DS-SLAM、Dyna-SLAM同类型系统相比,在保证定位精度和鲁棒性的同时,跟踪线程中处理一帧图像平均只需28.26 ms。结论 该算法能够有效降低动态物体对实时SLAM过程造成的干扰,为实现更加智能化、自动化的包装流程提供了可能。     

基于信标同步触发的机械振动WSNs传输休眠时序调度方法

针对IEEE 802. 15. 4协议中载波侦听多路访问/冲突避免机制(CSMA-CA)在机械振动无线传感器网络大量原始数据传输需求下存在多次载波侦听导致高侦听能耗、隐藏节点问题导致高重传能耗等问题,提出一种基于信标同步触发的机械振动无线传感器网络传输休眠时序调度方法。首先利用在信标网络中广播信标帧具有全网周期性同步的特点,对超帧结构进行设计,为物理上具有分散性的各节点接收时序调度信息提供基本的时间同步条件;然后以各节点网络短地址为调度信息并利用信标可携带负载的特点,将调度信息载入信标进行广播,实现低开销的调度信息通知机制;最后,各节点根据调度信息进行数据传输、休眠时序调度,改善并优化CSMA-CA在传输大量原始数据时存在无谓的传输能耗浪费缺陷。较之CSMA-CA,实验结果表明所提方法能有效降低机械振动无线传感器网络数据传输能耗。     

基于聚类状态隶属度的动态调度Q-学习

提出了一种利用Q-学习解决动态单机调度环境下的自适应调度规则选择的方法.该方法针对动态调度环境中系统状态空间大,Q-学习不易收敛的特点,首先提取系统状态特征,对系统状态进行合理聚类,有效地降低系统状态空间维数,然后在学习过程中令设备Agent根据瞬时状态向量对各聚类状态的隶属度做出综合判断,选择合适规则,并在每次迭代后根据隶属度将动作奖惩分配给各聚类状态的动作值函数.仿真结果表明,所提Q-学习算法较之传统Q-学习具有更快的收敛速度,提高了设备Agent的动态调度规则选择能力.     

生物计算网格中的在线调度技术研究

研究了在异构网格环境下的生物应用集成,定义了服务的提供者、部署者和使用者三种用户角色,设计了网格环境下的服务和资源整合机制,重点实现了用户管理以及作业调度控制等功能.根据计算资源的异构特点,设计了多种启发式调度算法.考虑到生物计算应用的不同类型,提出了自适应调度算法,该算法根据应用的特点动态选择启发式调度算法.实验表明,非阻塞调度优于阻塞调度方式;自适应调度算法比静态的在线调度算法有更好的性能,而在异构的网络中,带宽优先调度算法的性能比其他静态调度算法性能更好.     

Multi-sensor task allocation framework for supply networks security using task administration protocols

This research proposes a multi-sensor task allocation framework for security of supply networks aimed to maximise the number of correctly detected and reported security events (defined as tasks). The framework includes a double layer system consisting of a process layer and a monitoring layer. The process layer allocates sensors to tasks using an ant colony algorithm. The monitoring layer applies four task administration protocols (TAPs) specially developed and implemented to deal with high time-consuming tasks, conflicts in task priorities and sensor failure, defined in this research as overloading, deception and tampering of sensors, respectively. A system objective function for sensor to task allocation was developed to allow computation of the expected value of system performance given the sensor and the task parameters. Sensory limitations evaluated including reliability, distance coverage and the limited number of sensors are addressed in the decision-making process. The framework enables detection of tasks as soon as they occur in every location along the supply network, based on the sensor network distribution. The dual layer system analyses reveal that TAPs increase the systems performance in the scenarios of deception, tampering and overloading by more than 64% with respect to the number of unallocated tasks in comparison to a single layer system. Overall availability was analysed using Monte Carlo simulation and the fault tolerant system yielded significantly increased number of treated tasks (by 11%, p = 0.02).     

Effect of task difficulty and angle in a positioning task involving symmetrical and asymmetrical motions

Task difficulty where simultaneous hand motions are involved may be defined for a given angle(s) of hand motion, as the magnitude of hand motion and the amount of control needed for completing the hand motions. The task difficulty has traditionally been measured using the following relationship: TD = log₂ 2D/C bits where D is the distance moved and C is the target diameter. Task difficulty and its effect on element position involving symmetrical and asymmetrical simultaneous motion does not seem to have been investigated in the past. In this paper results of experimental investigations are reported. The experiments included nine levels of task difficulty and three angles of hand motion for each symmetrical and asymmetrical simultaneous motion, Models for predicting performance times for ‘position’ involving symmetrical and asymmetrical simultaneous motions have been developed. A comparison of predicted time values using these models has been made with the values developed using some of the available predetermined motion time systems.     

Multichannel communication of an industrial task

A specially constructed simulated assembly task was used to test the effectiveness of several multichannel communication combinations. The assembly task was a pegboard with a 4 × 4 matrix of ?-in. diameter, 3 in. long wooden dowels placed on ?-in. centres. Three type3 of wooden washers were made to fit on to the matrix of wooden pegs. The washers were assembled on to the pegs according to a sot of instructions presented to the subject.

The three primary communication modes under study were the verbal mode, pictorial mode, and the written mode. Instructions for the verbal mode were presented via a, prerecorded taped message. The pictorial instructions consisted of the presentation of n 35 mm coloured slide of the assembly task to be completed. The written instructions were also 35 mm slides, but of a typewritten message. The three primary modes were combined to form a series of multichannel mossagos, that is, information presented via two channels.

Two groups of 90 subjects each, 90 industrial subjects and 90 university students, performed the assembly task.

The hypothesis under investigation was that cue summation provides an initial insight into the explanation of multichannel communication. The specific hypothesis was that when multichannel cues summate there will be facilitation in performance. Also, the effects of interference and redundancy were investigated. The primary results were that the cue summation principle docs not seem to present adequate depth to form a basis for a theory of multichannel communication : however, in the absence of a ba3ic theory the pictorial channel still seems to be the most efficient communication medium for industrial procedure instructions.     

Integrated project task and manpower scheduling

The objective of project task scheduling is to determine task start dates and durations to complete a project on time with the minimum cost of performing tasks plus overhead. By altering task start dates and durations, the daily labor-demand profile can be changed. The objective of personnel scheduling is to determine how many workers must be assigned to each feasible days-off tour to satisfy a given labor-demand profile with minimum labor cost. Integrating these two problems permits the simultaneous determination of start dates, durations, labor levels and required tours for a minimum-cost and on-time schedule. Both integer programming and heuristic solution procedures to solve the integrated problem are presented. In a series of 20 test problems, the heuristic procedure outperformed the traditional two-step scheduling procedure by reducing the cost of labor and overhead by 8.6%.     

Heuristics for single-pass welding task sequencing

Welding task sequencing is a prerequisite in the offline programming of robot arc welding. Single-pass welding task sequencing can be modelled as a modified travelling salesman problem. Owing to the difficulty of the resulting arc-routing problems, effective local search heuristics are developed. Computational speed becomes important because robot arc welding is often part of an automated process-planning procedure. Generating a reasonable solution in an acceptable time is necessary for effective automated process planning. Several different heuristics are proposed for solving the welding task-sequencing problem considering both productivity and the potential for welding distortion. Constructive heuristics based on the nearest neighbour concept and tabu search heuristics are developed and enhanced using improvement procedures. The effectiveness of the heuristics developed is tested and verified on actual welded structure problems and random problems.     

Effects of practice, age, and task demands, on interference from a phone task while driving

Experimental research on the effects of cellular phone conversations on driving indicates that the phone task interferes with many driving-related functions, especially with older drivers. Unfortunately in past research (1) the dual task conditions were not repeated in order to test for learning, (2) the 'phone tasks' were not representative of real conversations, and (3) most often both the driving and the phone tasks were experimenter-paced. In real driving drivers learn to time-share various tasks, they can pace their driving to accommodate the demands of a phone conversation, and they can even partially pace the phone conversation to accommodate the driving demands. The present study was designed to better simulate real driving conditions by providing a simulated driving environment with repeated experiences of driving while carrying two different hands-free 'phone' tasks with different proximities to real conversations. In the course of five sessions of driving and using the phone, there was a learning effect on most of the driving measures. In addition, the interference from the phone task on many of the driving tasks diminished over time as expected. Finally, the interference effects were greater when the phone task was the often-used artificial math operations task than when it was an emotionally involving conversation, when the driving demands were greater, and when the drivers were older. Thus, the deleterious effects of conversing on the phone are very real initially, but may not be as severe with continued practice at the dual task, especially for drivers who are not old.     

基于Q学习的任务调度问题的改进研究

论文针对协同工作中的任务调度问题,建立了相应的马尔可夫决策过程模型,在此基础上提出了一种改进的基于模拟退火的Q学习算法。该算法通过引入模拟退火,并结合贪婪策略,以及在状态空间上的筛选判断,显著地提高了收敛速度,缩短了执行时间。最后与其它文献中相关算法的对比分析,验证了本改进算法的高效性。     

任务分配问题的对称群方法

探讨了无人飞行器（UAV）编队的任务分配问题。任务分配是UAV协同控制的基础,其解是任务区域内各任务的一个排列。求解UAV任务分配问题的有效方法是能在合理的计算时间内找到近似最优解的启发式算法。用对称群描述UAV任务分配的搜索空间,基于右乘运算构造搜索邻域。仿真结果验证了群论禁忌搜索算法的有效性。