Modeling and deadlock avoidance of automated manufacturing systems with multiple automated guided vehicles.

An automated manufacturing system (AMS) contains a number of versatile machines (or workstations), buffers, an automated material handling system (MHS), and is computer-controlled. An effective and flexible alternative for implementing MHS is to use automated guided vehicle (AGV) system. The deadlock issue in AMS is very important in its operation and has extensively been studied. The deadlock problems were separately treated for parts in production and transportation and many techniques were developed for each problem. However, such treatment does not take the advantage of the flexibility offered by multiple AGVs. In general, it is intractable to obtain maximally permissive control policy for either problem. Instead, this paper investigates these two problems in an integrated way. First we model an AGV system and part processing processes by resource-oriented Petri nets, respectively. Then the two models are integrated by using macro transitions. Based on the combined model, a novel control policy for deadlock avoidance is proposed. It is shown to be maximally permissive with computational complexity of O (n2) where n is the number of machines in AMS if the complexity for controlling the part transportation by AGVs is not considered. Thus, the complexity of deadlock avoidance for the whole system is bounded by the complexity in controlling the AGV system. An illustrative example shows its application and power.     

Deadlock Resolution in Automated Manufacturing Systems With Robots

An automated manufacturing system (AMS) contains a number of versatile machines (or workstations), buffers, and an automated material handling system (MHS). The MHS can be an automated guide vehicle (AGV) system, and/or a system that consists of multiple robots. Deadlock resolution in AMS is an important issue. For the AMS with an AGV system as MHS, the problems of deadlock resolution for part processing process and AGV system as an integrated system has been studied. It is shown that AGVs can serve as both material handling devices and central buffers at the same time to help resolve deadlocks. For AMS with robots as MHS, the existing work treated the robots just as material handling devices and showed that the robots had contribution to deadlock. In this paper, such AMS is modeled by resource-oriented Petri nets. Contrary to the existing work, it is shown that the robots have no contribution to deadlock by adopting such nets to control AMS. More interestingly, they can be used to resolve deadlock by serving as temporary part storage devices. A new deadlock control policy is proposed by treating robots as both material handling devices and buffers. The new policy outperforms the existing ones.     

自动导引车与机器集成调度问题研究现状

随着自动导引车（automated guided vehicles,AGV）的广泛应用,柔性制造车间中机器设备与AGV之间的协同配合日益受到重视。AGV与机器的集成调度主要研究机器分配、工序排序、搬运任务的AGV分配以及AGV路径规划。该问题是极为复杂的组合优化问题,对其研究具有重要的学术意义和应用价值。围绕问题特征,从模型与算法两个方面,对国内外最新的研究文献进行了梳理。对现有模型中的约束条件和优化目标进行了详细分类,从遗传算法、混合优化算法、仿真优化算法等五个方面综述了现有算法研究中的代表性成果。在此基础上,指出了现有研究中的不足,提出了未来的研究内容和方向。     

Intelligent AGV driving toward an autonomous decentralized manufacturing system

This paper proposes two methods that give intelligence to automatically guided vehicles (AGVs). In order to drive AGVs autonomously, two types of problems need to be overcome. They are the AGV navigation problem and collision avoidance problem. The first problem has been well known since 1980s. A new method based on the feature scene recognition and acquisition is proposed. The sparse distributed memory neural network (SDM) is employed for the scene recognition and acquisition. The navigation route for the AGV is learnt by use of Q-learning depending on the recognized and acquired scenes. The second problem is described as mutual understanding of behaviors between AGVs. The method of mutual understanding is proposed by the use of Q-learning. Those two methods are combined together for driving plural AGVs autonomously to deliver raw materials between machine tools in a factory. They are incorporated into the AGVs as the machine intelligence. In experimental simulations, it is verified that the first proposed method can guide the AGV to the suitable navigation and that the second method can acquire knowledge of mutual understanding of the AGVs’ behaviors.     

AGV作业调度模型及改进的DE算法研究

AGV作业调度问题的求解结果对AS/RS的运行效率具有重要影响。通过必要的简化,建立了AGV作业调度问题的静态优化模型。可知静态AGV作业调度问题实质是一种带约束的多重TSP问题,属于典型的NP完全问题,目前还不存在可在多项式时间内求解的确定算法。提出了一种改进的差分演化算法用于求解该问题。为了适应AGV作业调度问题的特点,新算法设计了新的两段编码方法,对多个DE算子进行了改造。还提出了基于生存时间的种群多样性增强机制,用于增强算法的搜索能力,避免陷入局部最优。仿真实验显示,该算法可以有效提高AGV作业调度的效率,验证了相关改进机制的有效性。     

Scheduling of no buffer job shop cells with blocking constraints and automated guided vehicles

The blocking job shop (BJS) problem is an extension of a job shop problem with no buffer constraints. It means that after a job is completed on the current machine, it remains on that machine until the next machine becomes available. This paper addresses an extension of the BJS problem, which takes into account transferring jobs between different machines using a limited number of automated guided vehicles (AGV), called a BJS–AGV problem. Two integer non-linear programming (INLP) models are proposed. A two-stage heuristic algorithm that combines an improving timetabling method and a local search is proposed to solve the BJS–AGV problem. A neighborhood structure in the local search is proposed based on a disjunctive graph model. According to the characteristics of the BJS–AGV problem, four principles are proposed to guarantee the feasibility of the search neighborhood. Computation results are presented for a set of benchmarking tests, some of which are enlarged by transportation times between different machines. The numerical results show the effectiveness of the proposed two-stage algorithm.     

Resource management in decentralized industrial Automated Guided Vehicle systems

This paper proposes an advanced decentralized method where an Automated Guided Vehicle (AGV) can optimally insert charging stations into an already assigned optimal tour of task locations. In today's industrial AGV systems, advanced algorithms and techniques are used to control the whole fleet of AGVs robustly and efficiently. While in academia, much research is conducted towards every aspect of AGV control. However, resource management or battery management is still one aspect which is usually omitted in research. In current industrial AGV systems, AGVs operate until their resource level drops below a certain threshold. Subsequently, they head to a charging station to charge fully. This programmed behaviour may have a negative impact on the manufacturing systems performance. AGVs lose time charging at inconvenient moments while this time loss could be avoided. Using the approach, an AGV can choose independently when it will visit a charging station and how long it will charge there. A general constrained optimization algorithm will be used to solve the problem and the current industrial resource management will be used as a benchmark. We use a simple extension of the Traveling Salesman Problem (TSP) representation to model our approach. The paper follows a decentral approach which is in the interest of the authors. The result of the proposal is a compact and practical method which can be used in today's operative central or decentral controlled AGV systems.     

干扰约束下考虑分组作业面的岸桥AGV联合调度

为解决自动化码头海侧多阶段设备作业的协调问题,加快集装箱在码头内部的周转过程。考虑干扰约束下分组作业面的的岸桥自动导引小车（AGV）联合调度问题。以岸桥、AGV完工时间和AGV等待时间加权总和最小为目标,考虑岸桥实际操作中的干扰约束与AGV堵塞等待等情况,建立岸桥与AGV联合调度优化模型。提出岸桥动态调度与AGV分组作业面调度模式,设计不同规模的算例,并采用遗传算法（GA）进行求解,将计算结果与传统调度模式进行对比。结果表明,该算法能有效提高岸桥与AGV作业效率,降低AGV的等待时间与堵塞次数,为码头实际作业提供依据。     

基于Petri网和人工势场的柔性制造启发式优化方法

制造系统优化调度是NP难组合优化问题,而自动导引车(AGV)路径规划与任务分配紧密耦合,又极大加剧了问题的复杂性.基于此,提出一种基于Petri网和人工势场的启发式优化方法.首先,将制造系统的工艺工序描述为一个任务Petri网,将AGV系统描述为一个路径Petri网,将两个网合成在一起;然后,利用Petri网的拓扑结构,为网络结点设计势能参数,从而为Petri网赋予一个人工势场;接着,利用人工势场设计制造系统加工时间的启发式函数,并构建Petri网人工势场启发式A*算法,其中包括最大势差启发式函数和总体势差启发式函数,并验证最大势差启发式函数是可采纳的;最后,进行两组数值实验,实验结果表明,最大势差A*算法能够得到最优解,且平均计算效率比Dijkstra算法提高57%,但是无法满足大任务量的调度需求,而总体势差A*算法比最大势差A*算法平均计算效率提高至少1个数量级,能够在有限时间内求解AGV任务分配和路径规划的联合问题.     

Flow-shop path planning for multi-automated guided vehicles in intelligent textile spinning cyber-physical production systems dynamic environment

Aiming at the path planning and decision-making problem, multi-automated guided vehicles (AGVs) have played an increasingly important role in the multi-stage industries, e.g., textile spinning. We recast a framework to investigate the improved genetic algorithm (GA) on multi-AGV path optimization within spinning drawing frames to solve the complex multi-AGV maneuvering scheduling decision and path planning problem. The study reported in this paper simplifies the scheduling model to meet the drawing workshop's real-time application requirements. According to the characteristics of decision variables, the model divides into two decision variables: time-independent variables and time-dependent variables. The first step is to use a GA to solve the AGV resource allocation problem based on the AGV resource pool strategy and specify the sliver can's transportation task. The second step is to determine the AGV transportation scheduling problem based on the sliver can-AGV matching information obtained in the first step. One significant advantage of the presented approach is that the fitness function is calculated based on the machine selection strategy, AGV resource pool strategy, and the process constraints, determining the scheduling sequence of the AGVs to deliver can. Moreover, it discovered that double-path decision-making constraints minimize the total path distance of all AGVs, and minimizing single-path distances of each AGVs exerted. By using the improved GA, simulation results show that the total path distance was shortened.     

紧致密集Auto Store系统AGV路径规划与避碰策略

针对新兴紧致密集仓储系统Auto Store具有短途挪库作业多、顶层AGV冲突多、货架结构性角落多等特点,提出一种离线-在线两阶段AGV优化调度方法。离线路径规划阶段,给出改进双层A*算法,在拓扑图建模划分搜索区域基础上,上层通过考虑冲突的启发式函数和考虑转弯的代价函数寻求可行区域,下层在此区域基础上搜索最优路径。在线AGV运行阶段,针对两AGV冲突,扩充了回退策略和路线重规划策略;针对多AGV冲突,提出一种基于贪心算法的区域避碰决策策略,以控制问题规模。最后利用Flexsim仿真进行了验证,结果表明,较于标准A*算法,改进A*算法能在保证搜索效率的同时获得冲突较少的初始路径方案;较于优先级策略,区域避碰策略能减少AGV等待时间;将二者相结合,能缩短整体作业完成时间,且随着AGV数量和作业任务增多,优势越明显。     

Positioning and obstacle avoidance of automatic guided vehicle in partially known environment

This paper presents positioning and obstacle avoidance of Automatic Guidance Vehicle (AGV) in partially known environment. To do this task, the followings are done. Firstly, the system configuration of AGV is described. Secondly, mathematical kinematic modeling of the AGV is presented to understand its characteristics and behavior. Thirdly, the Simultaneous Localization and Mapping (SLAM) algorithm based on the laser measurement system and encoders is proposed. The encoders are used for detecting the motion state of the AGV. In a slippery environment and a high speed AGV condition, encoder positioning method generates big error. Therefore, Extended Kalman Filter (EKF) is used to get the best position estimation of AGV by combining the encoder positioning result and landmark positions obtained from the laser scanner. Fourthly, to achieve the desired coordinate, D* Lite algorithm is used to generate a path from the start point to the goal point for AGV and to avoid unknown obstacles using information obtained from laser scanner. A backstepping controller based on Lyapunov stability is proposed for tracking the desired path generated by D* Lite algorithm. Finally, the effectiveness of the proposed algorithms and controller are verified by using experiment. The experimental results show that the AGV successfully reaches the goal point with an acceptable small error.     

基于改进的蚁群算法（ACO）的混合多目标AGV调度

蚁群算法是一种通过模拟蚁群的寻路行为对现实问题进行优化的现代智能仿生算法。针对实现AGV任务作业调度时,行驶路径最短的实际应用需求,本文将AGV的路径优化模型转化为旅行商问题,分析了多目标AGV优化中出现的冲突问题。在本文中,尝试了一种直接通信机制来进行改进传统算法,改进后的方法能够更好地维持种群的持久性,最终对于AGV调度规划起到积极的作用,有效地提高了AGV调度系统的效率。     

基于动态极大度的极小碰集求解方法

在计算集合簇的碰集时,结合SE-Tree(set enumeration tree)形式化地表达计算过程,逐步生成所有的极小碰集.并在SE-Tree中添加了终止结点,避免了非极小碰集的产生,并且不会因剪枝而丢失正确的解.提出未扩展元素度的概念和结点度的概念,进而在扩展SE-Tree结点时按照未扩展元素度由大到小的顺序扩展,极早地生成集合簇的碰集,减少枚举树生成的结点个数,并且直接根据结点度得出结点对应的集合是否为集合簇的碰集,避免计算集合是否为集合簇的碰集.实验结果表明,该算法程序容易编制且效率较好.     

AGV scheduling for automated material distribution: a case study

We have studied the scheduling of automated guided vehicles (AGVs) for efficient and uniform material distribution from a truck-dock to machining units of the machine shop in an automotive manufacturing plant. The material distribution problem, being a much simpler special case of the more general material transfer problem, is easily amenable to analysis. We have assessed the number of AGVs required to meet the total material requirements of all the machining units in the shop. Proposing innovative dispatch rules, we have evaluated their performance in simulation by monitoring parameters reflecting efficiency and uniformity of material distribution, both for single AGV and multiple AGV case. In multiple AGV case, we introduce the notion of zones having comparable demands for AGV, and assign one AGV to each zone, so that each AGV can operate largely independently—sharing a minimum path with other AGVs using deadlock avoiding protocols. The results of simulation runs and their implications are discussed.     

基于双目定位的离合器下料系统设计

机械臂在生产线上抓取组装完成的离合器后,需要准确地将离合器放置在自动导引车(AGV)的下料位置.针对AGV入库时位置不确定,机械臂无法准确实现下料问题,设计了一种基于机器视觉技术的离合器下料定位系统,并针对AGV尺寸较大、单相机定位存在精度差、定位不稳定的重要问题,采用双目定位提高精度和稳定性.实际使用结果表明:基于VisionPro视觉软件设计的AGV特征识别视觉定位系统具有很好的定位结果,定位精度为±0.5 mm,运行时间小于2 s;能够引导机械臂快速、准确地完成离合器下料任务,满足工业实际应用的需求.     

基于时间窗的AGV动态避碰路径规划方法

为了解决多AGV在动态不稳环境下的无碰撞路径规划和系统效率提升的问题,提出了基于时间窗的AGV无碰撞路径规划方法。首先建立了多AGV的避碰模型,并结合时间窗模型,将多AGV的无碰撞路径规划分为预先规划和实时规划两阶段,预先规划阶段进行多AGV无冲突时间窗的计算和最大化系统中AGV的流通量,实时规划阶段通过改变AGV在避碰模型上的占用优先级和局部重规划的方法进行动态避碰。最后以某智能仓储为应用案例进行仿真实验,证明了该算法能有效避免多AGV的碰撞,提高AGV的流通量,同时在动态环境下具有较好的鲁棒性和柔性。     

基于双种群蚁群算法的AGV路径规划研究

针对蚁群算法存在的收敛速度慢、易陷入局部最优和容易死锁等问题,提出了一种用于自动引导车(Automated Guided Vehicle, AGV)路径规划的双种群蚁群算法。该算法引入差异化信息素初始值,修改启发函数并在信息素更新时对最优及最差路径进行奖惩；以改进策略为基础,引入自适应步长搜索策略,通过具有差异化步长的两个种群相互协作加强算法寻优能力和搜索效率；针对死锁问题,提出了将符合条件的单元格视为障碍物的“填充陷阱”策略。分别进行仿真实验和车间现场实验,结果表明,该算法可以为AGV规划出一条安全且综合性能较好的路径,为AGV路径规划提供了一种可行的方案。     

A study for further exploring the advantages of using multi-load automated guided vehicles

Multi-load Automated Guided Vehicle (AGV) has been proposed for a while, but the advantage of its application is still not fully understood today. In order to fill this knowledge gap, a novel integrated model of a multi-load AGV system is developed in this paper by using an advanced form of Petri Nets, namely Coloured Petri Nets (CPNs), to simulate the operation of the AGV system in various scenarios. The study reported in this paper is focused to answer a few key questions, i.e. whether system performance can be continuously improved by increasing the load capacity of the multi-load AGVs; if not, whether there is an optimal load capacity of the multi-load AGV for a particular system; and whether the multi-load AGVs can still work well in a system with flexible loading and unloading points. The research results have shown that the efficiency of the AGV system can be improved by increasing the load capacity at the beginning, but the effectiveness of such an approach will decrease when the load capacity increases above a certain value. In other words, an AGV system may not perform better after using a larger capacity of multi-load AGV and there must be an optimal load capacity of the multi-load AGV for a specific AGV system. In addition, it is found that a system with flexible loading and unloading points can perform better after using a multi-load AGV.