Decentralized Traffic Control for Non-Holonomic Flexible Automated Guided Vehicles in Industrial Environments

This paper deals with the problem of coordinating flexible automated guided vehicles (AGVs) in real manufacturing systems. The problem consists of ensuring safe and successful task execution while several AGVs operate as a distributed transportation system in real industrial environments. The proposed solution combines different decentralized techniques to increase the flexibility and scalability of the multirobot system. The coordination is addressed by dividing the problem into path planning, obstacle avoidance and traffic control problems. The path planning method takes into account the location of mates for replanning the routes. The obstacle avoidance technique considers the kinematic constraints of the platform for reactive motion control. The traffic control approach makes use of a decentralized control policy that takes into account the capabilities of vehicles. By combining all these techniques and configuring the system properly, we present the successful development of a distributed transportation system composed of a team of flexible AGVs. The proposed solution has been validated using both a set of custom-modified AGVs operating in a real factory and a simulation of several AGVs operating in a virtual scenario.     

A route and speed optimization model to find conflict-free routes for automated guided vehicles in large warehouses based on quick response code technology

This paper focuses on exploring conflict-free routes for automated guided vehicles (AGVs) based on quick response (QR) code technology. Intelligent warehousing has become an industrial development trend, with an increasing number of enterprises deploying AGVs based on QR code technology for transportation tasks, especially in large warehouses. Unlike the early single lane path following navigation systems, AGVs using QR code technology can freely switch tracks at any position. And two AGVs can be accommodated side by side. The conditions of conflict-free routes are analyzed firstly when AGVs use QR code navigation system. After that, a route and speed optimization model (RSOM) that aims at minimizing energy consumption and taking conflict-free routes with time windows as the constraint condition is established. Then a branch-and-bound (B&B) algorithm scheme is proposed by altering AGV routes or speed to arrive at schemes with conflict-free routes to complete transportation tasks. Lastly, the effectiveness and validity of our proposed model and algorithm are verified by numerical examples and a case study. The results show that resolving conflicts during route planning for AGVs guided through QR code have advantages over avoiding collision in the running phase by technology and path following techniques in both aspects of solution time and energy. In addition, the established B&B algorithm has better performance than the rapid-exploring random tree (RRT) algorithm in numerical examples. The gaps between them widen as the size of AGVs increases. The research work has potential for application to meet the requirements of AGV development.     

Network model and effective evolutionary approach for AGV dispatching in manufacturing system

Automated guided vehicles (AGVs), are the state-of-the-art, and are often used to facilitate automatic storage and retrieval systems (AS/RS). In this paper, we focus on the dispatching of AGVs in a flexible manufacturing system (FMS). A FMS environment requires a flexible and adaptable material handling system. We model an AGV system by using network structure. This network model of an AGV dispatching has simplexes decision variables with considering most AGV problem’s constraints, for example capacity of AGVs, precedence constraints among the processes, deadlock control. Furthermore, these problems can be solved by using a lot of heuristic algorithms as network optimization problems. We are also proposed an effective evolutionary approach for solving a kind of AGV’s problems in which minimizing time required to complete all jobs (i.e. makespan) and minimizing the number of AGVs, simultaneously. For applying an evolutionary approach for this multicriteria case of AGV problem, priority-based encoding method and Interactive Adaptive-weight GA (i-awGA) were proposed. Numerical analyses for case study show the effectiveness of proposed approach. Received: June 2005 / Accepted: December 2005     

A distributed routing method for AGVs under motion delay disturbance

In real transportation environments for semi-conductor manufacturing, unexpected disturbances such as motion delays and/or sudden requests for transportation may cause collisions among AGVs increasing total transportation time. It is required to generate a collision-free routing for multiple automated guided vehicles (AGVs) within a few seconds. In this paper, we propose a distributed routing method under motion delay disturbance for multiple AGVs. The proposed method features a characteristic that each AGV subsystem derives its optimal route to minimize the sum of the transportation time and the penalties with respect to collision probability with other AGVs. The penalties reflect the collision probability distribution function for motion delay disturbances. The proposed method is applied to a routing problem for transportation in the semiconductor fabrication bay with 143 nodes and 20 AGVs. The computational results show that the total transportation time obtained by the proposed method is shorter than that of the conventional method. For dynamic transportation environments, an optimal timing for re-routing multiple AGVs under motion delay disturbance is systematically determined by the tradeoff relationship between the total computation time to derive a solution and the uncertainties for re-routings. Markov chain is used to represent uncertainty distribution for re-routings. The proposed method is implemented in an experimental transportation system with 51 nodes and 5 AGVs. The experimental results demonstrate that the proposed method is applicable to real transportation environments.     

An architectural design of control software for automated container terminals

Automation of handling equipment is a current trend in port container terminal. Because a large number of equipment of different types are involved in the handling operation, the structure of control software is highly complicated. This paper introduces an architectural design of control software and a simulation-based test-bed for testing various control rules of the control software. The suggested control system consists of a ship operation manager (SOM) and system controllers for automated guided vehicles (AGVs) and automated yard cranes (AYCs). A SOM is in charge of dispatching tasks to equipment for preplanned load/unload operations. System controllers assign equipment to the SOM and maintain the information about the location and the status of equipment. To test the control system and the operation methods of the suggested container terminal, an object-oriented simulation system was developed using Java. Using the simulation system, several operation-related issues were tested, including the postponement of dispatching decisions and the sequence rearrangement of AGVs entering transfer areas under quay cranes (QCs).     

Distributed route planning for multiple mobile robots using an augmented Lagrangian decomposition and coordination technique

To enable efficient transportation in semiconductor fabrication bays, it is necessary to generate route planning of multiple automated guided vehicles (AGVs) efficiently to minimize the total transportation time without collision among AGVs. In this paper, we propose a distributed route-planning method for multiple mobile robots using an augmented Lagrangian decomposition and coordination technique. The proposed method features a characteristic that each AGV individually creates a near-optimal routing plan through repetitive data exchange among the AGVs and local optimization for each AGV. Dijkstra's algorithm is used for local optimization. The optimality of the solution generated by the proposed method is evaluated by comparing the solution with an optimal solution derived by solving integer linear programming problems. A near-optimal solution, within 3% of the average gap from the optimal solution for an example transportation system consisting of 143 nodes and 14 AGVs, can be derived in less than 5 s of computation time for 100 types of requests. The proposed method is implemented in an experimental system with three AGVs, and the routing plan is derived in the configuration space, taking the motion of the robot into account. It is experimentally demonstrated that the proposed method is effective for various problems, despite the fact that each route for an AGV is created without minimizing the entire objective function.     

A neural network model for the free-ranging AGV route-planning problem

This paper describes the development of a prototype neural network model for the free-ranging AGV route-planning problem. The vehicle planner operates in quasi-real time. A small planning horizon is set and all transport requests existing at the beginning of a planning horizon are examined. A neural network model is proposed to perform dispatching and routing tasks for the AGVs. Its goal is to satisfy the transport requests in the shortest time and in a non-conflicting manner, subject to the global manufacturing objective of maximizing throughputs. Based on Kohonen's self-organizing feature maps, we develop three efficient planning algorithms for the single and multiple AGV problems. The simulation results indicate that the proposed neural network approach gives very efficient solutions.     

An Ant Colony Algorithm (ACA) for solving the new integrated model of job shop scheduling and conflict-free routing of AGVs

This paper concerns with the Job Shop Scheduling Problem (JSSP) considering the transportation times of the jobs from one machine to another. The goal of a basic JSSP is to determine starting and ending times for each job in which the objective function can be optimized. In here, several Automated Guided Vehicles (AGVs) have been employed to transfer the jobs between machines and warehouse located at the production environment. Unlike the advantages of implemented automatic transportation system, if they are not controlled along the routes, it is possible that the production system encounters breakdown. Therefore, the Conflict-Free Routing Problem (CFRP) for AGVs is considered as well as the basic JSSP. Hence, we proposed a mathematical model which is composed of JSSP and CFRP, simultaneously and since the problem under study is NP-hard, a two stage Ant Colony Algorithm (ACA) is also proposed. The objective function is to minimize the total completion time (make-span). Eventually, in order to show the model and algorithm’s efficiency, the computational results of 13 test problems and sensitivity analysis are exhibited. The obtained results show that ACA is an efficient meta-heuristic for this problem, especially for the large-sized problems. In addition, the optimal number of both AGVs and rail-ways in the production environment is determined by economic analysis.     

Optimal coordination in logistics warehousing with sensing and communication limits: A distributed differential game approach

Optimal coordination is essential for multi-automated guided vehicle (AGV) systems, particularly in logistic transportation cases, where the system task completion time needs to be minimized, with the guarantee of safe operation. This is because an optimal coordination strategy (OCS), if achieved, can significantly improve the transportation system's efficiency. In this paper, to deal with the dynamic interaction process among AGVs, and sensing and communication range limits, we formulate the optimal coordination problem into a distributed differential game (DDG) framework, where individual AGVs only use information communicated from nearby AGVs to design their optimal operation trajectories. This helps to significantly reduce the computational and communication requirements for the multi-AGV logistic transportation systems. Targeting operation safety and working efficiency requirements, we incorporate collision avoidance and trajectory optimization objectives into the proposed framework. It is shown that local OCS, obtained by solving the DDG problem for each AGV, will converge to the global Nash equilibrium, which represents the most efficient operating condition for the entire logistic transportation system. Finally, the efficacy of the proposed method is demonstrated, based on simulations and experiments, benchmarked with existing logistic warehousing planning and differential game methods. Compared with conventional methods, the proposed framework successfully helps reduce the task completion time by up to 16%.     

求解存在运输空间约束多单元协作调度问题的拍卖算法

针对存在运输空间约束的多单元协作调度问题,提出合理的运输模式,建立非线性整数规划模型,对问题进行描述.通过拍卖的方式,将设备资源和运输资源分配给每一个工件,得到问题的可行解,并基于改进的(非)连通图对可行解进一步优化.通过拍卖方式,可以恰当地针对不同时间段的资源进行价值评估,从而提高资源的利用率,减少总的生产时间.拍卖过程分为两部分:车辆资源拍卖和设备资源拍卖.在整个拍卖过程中,车辆和设备分别扮演拍卖者,每个工件扮演竞拍者.通过对比实验,验证了所提出运输模式的合理性以及算法的有效性.     

Hybrid cost and time path planning for multiple autonomous guided vehicles

In this paper, simultaneous scheduling and routing problem for autonomous guided vehicles (AGVs) is investigated. At the beginning of the planning horizon list of orders is processed in the manufacturing system. The produced or semi-produced products are carried among stations using AGVs according to the process plan and the earliest delivery time rule. Thus, a network of stations and AGV paths is configured. The guide path is bi-direction and AGVs can only stop at the end of a node. Two kinds of collisions exist namely: AGVs move directly to a same node and AGVs are on a same path. Delay is defined as an order is carried after the earliest delivery time. Therefore, the problem is defined to consider some AGVs and material handling orders available and assign orders to AGVs so that collision free paths as cost attribute and minimal waiting time as time attribute, are obtained. Solving this problem leads to determine: the number of required AGVs for orders fulfillment assign orders to AGVs schedule delivery and material handling and route different AGVs. The problem is formulated as a network mathematical model and optimized using a modified network simplex algorithm. The proposed mathematical formulation is first adapted to a minimum cost flow (MCF) model and then optimized using a modified network simplex algorithm (NSA). Numerical illustrations verify and validate the proposed modelling and optimization. Also, comparative studies guarantee superiority of the proposed MCF-NSA solution approach.     

MLATSO: A method for task scheduling optimization in multi-load AGVs-based systems

In the context of increasingly competitive intelligent manufacturing, the multi-load Automated guided vehicles (AGVs) based Automated Storage and Retrieval System (AS/RS) has been of particular interest, as reductions in the number of AGVs required can significantly decrease potential congestions and increase the system effectiveness. In comparison with the single-load AGVs system, more difficult and critical issue of scheduling multi-load AGVs to automate storage/retrieval missions and to maximize economic benefits remains unresolved. Therefore, we propose a task scheduling optimization method for multi-load AGVs-based systems, with which, the objectives of least number of occupied AGVs, shortest travel time and minimum conflicts can be met simultaneously. The experiments are conducted in various scenarios, and verify that our work can use fewer AGVs to optimize tasks delivery, which enables the AS/RS stakeholders to reach win-win results for system performance and AGVs investment, thus maximizing economic benefit.     

Iterative multimodal processes scheduling

The paper considers the problem of Automated Guided Vehicles (AGVs) fleet scheduling subject to the right match-up of local cyclic acting AGV schedules to given workpiece machining schedules. The main contribution of this work is the solution to a constraint satisfaction problem aimed at AGVs fleet match-up scheduling while taking into consideration assumed itineraries of concurrently manufactured product types. In other words, assuming a given layout of FMS’s material handling and production routes of simultaneously manufactured work orders as well as cyclic schedules of concurrently manufactured product types, the goal is to provide a declarative model enabling multimodal processes, i.e. employing AGVs, hoists, lifts, etc. demand-responsible scheduling of transportation/handling services. An algebra-like driven approach to cyclic scheduling based on step-by-step composition of multimodal transportation network sub-structures is proposed. Results of computational experiments assessing scalability of the method provided are presented as well.     

The simulation design and analysis of a Flexible Manufacturing System with Automated Guided Vehicle System

This paper presents the simulation design and analysis of a Flexible Manufacturing System (FMS) with an Automated Guided Vehicle system (AGVs). To maximize the operating performance of FMS with AGVs, many parameters must be considered, including the number, velocity, and dispatching rule of AGV, part-types, scheduling, and buffer sizes. Of the various critical factors, we consider the following three: (1) minimizing the congestion; (2) minimizing the vehicle utilization; and (3) maximizing the throughput. In this paper, we consider the systematic analysis methods that combine a simulation-based analytic and optimization technique that is Multi-Objective Non-Linear Programming (MONLP) and Evolution Strategy (ES). MONLP determines the design parameters of the system through multi-factorial and regression analyses. ES is used to verify each parameter for simulation-based optimization. A validation test for the two methods is conducted. This method-based approach towards design yields the correct experimental results, ensures confidence in the specification of design parameters and supports a robust framework for analysis.     

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.     

The capacitated multi-AGV scheduling problem with conflicting products: Model and a decentralized multi-agent approach

Automated guided vehicles (AGVs) are a key technology to facilitate flexible production systems in the context of Industry 4.0. This paper investigates an optimization model and a solution using a decentralized multi-agent approach for a new capacitated multi-AGV scheduling problem with conflicting products (CMASPCP) to take full advantage of AGVs. The novelty of the problem and our model lies in the introduction of AGV capacity constraints and constraints arising from conflicting products, i.e. products that cannot be transported together. As the new I4.0 paradigm tends towards decentralized control, we also present a decentralized multi-agent approach in which AGVs autonomously coordinate to solve the task. The performance of the proposed decentralized approach is compared to a mixed-integer linear programming model on a set of 110 problem instances with different sizes and degrees of complexity. The obtained results show that the proposed decentralized multi-agent approach is effective and competitive in terms of the solution quality and computational time.     

Application of reinforcement learning for agent-based production scheduling

Reinforcement learning (RL) has received some attention in recent years from agent-based researchers because it deals with the problem of how an autonomous agent can learn to select proper actions for achieving its goals through interacting with its environment. Although there have been several successful examples demonstrating the usefulness of RL, its application to manufacturing systems has not been fully explored yet. In this paper, Q-learning, a popular RL algorithm, is applied to a single machine dispatching rule selection problem. This paper investigates the application potential of Q-learning, a widely used RL algorithm to a dispatching rule selection problem on a single machine to determine if it can be used to enable a single machine agent to learn commonly accepted dispatching rules for three example cases in which the best dispatching rules have been previously defined. This study provided encouraging results that show the potential of RL for application to agent-based production scheduling.     

A program for dispatching and routing AGVs

This paper describes a program for dispatching and routing of AGVs in real time in a flexible manufacturing environment. The AGVs are assumed to travel on a bidirectional guidepath. Given a set of open jobs, job priorities, their production routings, number of AGVs, AGV travel speeds and their loading/unloading times, the program decides which AGV to dispatch and the path to be followed by the AGV. Continuous monitoring of the system is made possible by a graphical display of AGV movements. The program can also be used to evaluate the design of the AGV system. Some computational results are presented to illustrate this.     

Mission design for a group of autonomous guided vehicles

In this paper, a generic approach for the integration of vehicle routing and scheduling and motion planning for a group of autonomous guided vehicles (AGVs) is proposed. The AGVs are requested to serve all the work stations in a two-dimensional environment while taking into account kinematics constraints and the environment’s geometry during their motion. The problem objective is the simultaneous determination of time-optimum and collision-free paths for all AGVs. The proposed method is investigated and discussed through a number of simulated experiments using a variety of environments and different initial conditions.