Cookie Consent by Free Privacy Policy Generator Effiziente AGV-Integration mit AI-Palettenerkennung - Invanta Blog
November 15th, 2023
5 min read

Pallet detection

Automation and efficiency improvements have been the goal of every manufacturing plant since time immemorial. From sweatshops, we have moved to the possibility of fully automated lines requiring only a handful of operators. However, converting a captive hall to a fully robotic one is very costly and not always logistically possible. So nowadays, we are most often faced with hybrid operations combining partially automated lines and the work of operators. The drive to automate has not escaped the transport of materials within the production area. In recent years, the use of automated guided vehicles (AGVs) has become a growing trend.

AGV systems take care of the transport of materials for processing and the transport of pallets of finished products for dispatch. However, in order to operate, the AGV system must somehow know that it has material/pallets ready for removal. In practice, three solutions are encountered.

1. The first and simplest one is the manual summoning of the trolley by the operator, e.g. using a button. However, this solution is only applicable in some cases and goes against the sense of autonomous operation.

2. The second very common solution is sending the AGV on routine patrols. During this patrol, the truck itself goes around and scans the defined collection points and, if it detects material at these points, creates jobs for subsequent collection. This solution is attractive at first sight because it does not require any additional mounting of possible sensors, but it carries a number of disadvantages:

- Reduced AGV efficiency. While material is waiting at one collection point, the truck is unnecessarily travelling to the other end of the factory to look at other locations before the slot in question arrives.

- Reduced AGV life. Due to the extra distance travelled, AGVs are unnecessarily discharged, which both leads to delays in operation when charging, but more importantly, reduces the overall lifetime as each battery has a limited number of charge cycles.

- Deterioration of safety during saturation of the aisles. An AGV that drives unnecessarily to look at an empty spot as part of a routine is de facto obstructing the path during the run. Which, again, reduces the efficiency of the factory operation, but more importantly increases the risk of collision with, for example, a forklift operator.

3. A third solution is the use of sensors such as a laser beam, limit switch or scanner integrated directly into the production lines. However, with a large number of picking points, a large-scale installation with individual sensors for each location is required. In some cases, it is not practical to install conventional sensors, for example, at mid-warehouse picking locations where laser beams and limit switches require floor intervention and obstruct the floor, while scanners require lowering brackets from the ceiling, which is impractical at higher volumes.

An alternative solution, overcoming the disadvantages of conventional sensors, is offered by our camera sensor. The main advantage of our sensor is:

- The ability to detect material/pallets in several places simultaneously. Thus, only one sensor is needed for several locations side by side, which greatly simplifies the installation.

- Flexibility in the placement of the sensor, which makes installation easier. There is no need to place the sensor in a specific location, e.g. above the area to be monitored. It can be placed, for example, on the wall of a hall, on a pillar nearby, etc.

- Better integration into AGV systems thanks to Ethernet communication

In addition to stand-alone sensors, our company also offers integration with AGV systems along with the development of add-on software to maximize the use of AGVs. Our software enables better scheduling of material transfer jobs, automatic job cancellation when material is removed by the operator, redirection of AGV operation when collection points are full, and more.

Aktuelle Artikel

Alle anzeigen