Navigation using environmental features

Scenario of driverless transport systems in the industry. © Shutterstock / Chesky

However, the greatest technological developments in recent years have been in the field of navigation. This is also one of the most sensitive and important areas for the operation of automated guided vehicles. This is because they can only be used successfully in intralogistics and order picking if the conveyors are able to orientate themselves safely in space and recognize the pick-up and delivery points. Various systems have emerged here in recent years:

Optical, magnetic or inductive guidelines in the floor area:

For a long time, previously available AGVs relied on optical, magnetic or inductive guidelines that were embedded in or glued to the hall floor. Due to the comparatively high installation costs and the very limited flexibility, this technology is virtually no longer used in newer generations of AGVs.

Transponders and magnets for grid navigation:

With grid navigation, the vehicles orient themselves using reflectors, magnets or other reference points that form a grid over the storage or production area. By attaching new reflectors, new routes and changes to the layout can be taken into account relatively quickly and easily.

Laser navigation:

The most commonly used method today is laser navigation, or more precisely laser triangulation. Reflectors on the hall walls and columns can be used to determine the position in the area. Travel paths are adapted and extended exclusively via the corresponding control software.

Navigation using environmental features:

The latest generation of autonomous AGVs (aAGV) uses integrated laser scanners or GPS systems to create a complete 3D image of the hall plan and is therefore able to orient itself independently in space. Instead of moving exclusively along predefined routes, aAGVs search for their own route to get from A to B.

The central prerequisite for this type of "free navigation in space" is a high computing capacity in the device, which enables it to process the data collected by sensors in real time and plan its route accordingly. This optimized navigation enables a much more flexible use of AGVs beyond predefined routes and thus creates the basis for a decentralized or self-organized system of replenishment control.

From central control to decentralized intelligence

Such a system not only allows the AGVs to orient themselves independently in space, but also to intelligently coordinate the allocation and assignment of transport orders in the swarm. At present, this order allocation usually works via a central control station, which distributes the transport orders to the available AGVs in a similar way to a cab control center. Powerful algorithms or software agents ensure that the orders are assigned to the available vehicles that are closest to the object to be transported. Modern systems not only include the next order in the calculation, but also - similar to a good chess computer - other pending orders. Various system settings are possible, depending on whether the focus is on even utilization of the AGVs or a specific maximum duration for the individual transport process. Due to this predictive calculation logic, centralized systems are currently still superior to decentralized order control. This is because in order to apply the same calculation logic to swarm-intelligent systems, the computing capacity in the individual vehicles would have to be considerably higher. In the long term, however, it can be expected that decentralized systems will prevail, as they are superior to centralized systems, particularly in terms of scalability and reliability.

Autonomous replenishment control

The productivity benefits of driverless transport systems compared to conventional forms of warehouse logistics are obvious. This applies both to the degree of utilization of the individual vehicles and to occupational safety. The Continental plant in Regensburg, which received the "Industry 4.0 Award" from ROI in 2015, shows what such a system can look like in practice. There, a completely automated replenishment process was implemented at KLT level, in which the machine independently requests the materials required for the respective order from the warehouse. These are then transported to the production line without human intervention using AGVs.

Keeping an eye on the overall business result

However, in order to implement a comparable solution, the appropriate framework conditions must first be created. Companies often underestimate the effort and costs involved in introducing driverless transport systems. This is because the vehicles themselves are only one part of the overall business calculation. In order to be able to integrate them sensibly into your own logistics processes, an IT connection is just as essential as the adaptation of the relevant material flow interfaces, for example transfer points. The rule of thumb is therefore: vehicle costs multiplied by two equals the total implementation costs.

If mistakes are made in this phase, AGVs often turn from a quick efficiency booster into a technical gimmick. Because in the factory of the future, the flow of goods and the flow of information can no longer be separated. In this sense, driverless transport vehicles must also be part of the information flow in the factory. Only then will they become the nerve pathways of the smart factory.

Prof. Dr.-Ing. Werner Bick, Chief Representative of ROI Management Consulting AG and Professor at OTH Regensburg / ag