Automated generation of roadmaps for automated guided vehicle systems

Automated guided vehicle systems (AGVS) have become indispensable in advanced production facilities. Due to significant progress in the field of AGVS and the increased automation within production plants, the potential applications for AGVS increase. So far the roadmaps for the vehicles are mostly generated manually, which leads to long and laborious planning phases. This thesis examines how system planners' knowledge can be integrated into a pathfinding algorithm in order to combine human logic with mathematical optimization to generate roadmaps for AGVS that are both efficient and applicable.

The combination of mathematical planning and human planning was achieved by combining a fuzzy inference system with a traditional pathfinding algorithm - the A* algorithm. The fuzzy inference system stores the knowledge of the system planners in the form of fuzzy rules and the output of the rules directly influence the path planning of the A* algorithm.

pathfinding, expert system, fuzzy logic, automated guided vehicle system, automated guided vehicle

This paper proposes a method for the automated generation of roadmaps for AGVs. So far the roadmaps are mostly generated manually, which leads to long and laborious planning phases. The presented method incorporates both mathematical roadmap algorithms as well as human knowledge in the form of a fuzzy inference system. The results of the expert system are evaluated in comparisons to the A* algorithm and to manually generated roadmaps on a real production layout. In both cases the expert system performs better.

fuzzy logic, expert system, AGV, roadmap

Automated guided vehicle systems (AGVS) have become indispensable in advanced production facilities. Due to significant progress in the field of automated guided vehicles (AGVs) and the increased automation within production plants, the potential applications for AGVs increase. The design of the roadmap for automated guided vehicle systems is a time-consuming process which is currently performed manually for the most part. Because of the AGVs increasing degree of complexity a manual design of the roadmap becomes more and more difficult and challenging. In the course of the research project “Automatic design of the roadmap for automated guided vehicle systems” (IG 18007) a software demonstrator was developed which allows the automated generation of the roadmaps for AGVS. The software demonstrator was applied to real reference scenarios and it was proven that the automatically generated roadmaps are as reliable as the manually generated ones and in some cases even more efficient.

fuzzy logic, expert system, AGV, roadmap

So far the generation of roadmaps for automated guided vehicles (AGVs) is mostly performed manually. Mathematical path finding algorithms often return results that are mathematically optimal but not applicable to a real production layout. This paper proposes an expert system as a solution that combines traditional path finding algorithms (in the form of a modified version of the A* and the Bellman-Ford algorithm) with a fuzzy inference system that incorporates the human knowledge of AGV system planners. Results that prove the efficiency of the proposed solution are shown in the end.

roadmaps, automated guided vehicles, path planning, fuzzy logic

Automated guided vehicles are well established in modern production environments. Once installed the automated guided vehicles are able to automatically transports good inside the factory. The set up of the system is still a laborous task with high planning costs. Besides the right choice of vehicles and the right controll strategy, the planning of an efficient roadmap is one of the major time consuming tasks. The planning of the roadmaps is mostly performed manually. Experienced system planners decide if a roadmap is feasible or has to be adapted. Aim of this project is to design automatically roadmaps that are both efficient and applicable for AGVs. The planning phase of the roadmaps should be reduced from several weeks to a few hours. In the end a ressource-efficient and standardized planning is aspired.

AGV, road map planning, fuzzy logic

Automated guided vehicles (AGV) are very common in modern production environments nowadays. The planning of roadmaps for the vehicles is still a very time consuming task that includes different optimization goals (e.g. shortest path, collision avoidance, maximizing of the free space). Depending on the scenario different optimization goals have different priorities. The weighting of the goals depends on the experience and judgement of the experts. Automated approaches in form of path planning algorithms give results that are mathematically correct but aren’t applicable in reality. This paper presents an expert system that automatically generates roadmaps for AGVs. The aim is to generate road maps that are both efficient and applicable. For this purpose common path planning algorithms (Lee and A*), that are able to generate mathematical efficient road maps, are combined with expert knowledge in the form of fuzzy logic. The results of the expert system are evaluated using real layouts and are compared to manual generated road maps.

Road map, automated guided vehicles, A*, path planning

Path-finding algorithms (PFA) are successfully used to find the optimal path between two locations. Good results are obtained if they are used in scenarios where the entire environment can be described mathematically. Production environments of automated guided vehicles (AGVs) are not one of those. PFA find solutions that are mathematically correct but miss human expertise that would dismiss solutions of the algorithm that aren’t applicable to a real production layout. This paper presents a hybrid algorithm consisting of an A* algorithm, and a fuzzy logic control in order to generate a fuzzy-enhanced A* algorithm (FEA*) that produces efficient and applicable road maps for AGVs. First computational results are shown.

path-finding algorithm, fuzzy-logic, expert system, agv, road maps

Designing a road layout for automated guided vehicles (AGV) can be a very laborious process that is based in big parts upon the knowledge of experienced system planers. Up until now, it isn’t possible to save that knowledge in a form that makes it usable in an automated layout process for AGV- roadmaps. The research project aims to integrate the knowledge of the system planers in an artificial intelligence, so that in the future an automated process of designing AGV roadmaps is achievable. The knowledge is implemented within a fuzzy-logic and can be used as a controller for the planning process.

automated guided vehicles, expert system, fuzzy logic, fuzzy logic controller

The generation of road maps for automated guided vehicles (AGVs) in the area of logistics is increasing in complexity as a result of the changing production environment. Reasons for that are the shorter product life-cycles and the advances in the field of AGVs. So far, the generation of roadmaps for AGVs is a manual task and is therefore time-consuming and expensive. Information from different sources (e. g. the production layout and the transport matrix) have to be evaluated and be joined. A simplification offers the storage of expert knowledge of system planers in a form of artificial intelligence in a fuzzy-rule base. This concept is going to be realized in a software tool that is going to be able to automatically generate roadmaps for AGVs.

automated guided vehicles, expert system, fuzzy logic, fuzzy logic controller