Forbes Magazine about swarm intelligence

AGILOX Robots rely on Swarm Intelligence by Steve Baker (forbes.com)

Jul 19 2018

AGILOX technology analyzed by Forbes Magazine. Thank you Steve Banker for this great article.

We confirm the author's guess, where he says: "They are poised" - when it comes to IGVs. Read the original article.


Robots with swarm intelligence. That sounds ominous. It also sounds cool! I talked to Dirk Erlacher, the CEO of Agilox, on this topic. Austrian headquartered Agilox designs and manufactures mobile logistics robots that use "swarm intelligence" to intelligently navigate through warehouses and factories, delivering pallets and totes where they are needed.

A mobile logistics robot (MLR) is a more advanced form of an automatic guided vehicle (AGV); AGVs are used to reduce labor by taking over tasks that were traditionally performed with fork lifts.

More complex AGVs have fleet management software. This software makes sure that not too many AGVs are in the same aisles, decides which AGV has the right of way at crossings, and in more complex scenarios, decides which unit will be used to complete a particular task and how it will navigate through the facility. This fleet management software is a form of centralized intelligence. All the AGVs report back to this central control unit, and it makes the decisions.

But in the Agilox solution, there is no central control unit that defines paths; each robot does this itself in a collaborative manner. Each vehicle independently calculates its best route and exchanges this information with the other vehicles of the fleet in real time. This way every vehicle has the latest and most accurate information about its surroundings at any given moment. Because each vehicle is providing constant updates about the current surroundings, the information shared with peer vehicles includes information on potential obstacles on the route. "You can compare this situation to traffic congestion,"​ Mr. Erlacher explained. "The driver inside the traffic jam will always have the most accurate information about what is going on in the streets."​

This sounds like agent technology. Intelligent agents are autonomous, problem-solving computational units often deployed in environments in which they interact and cooperate with other agents. But swarm intelligence does sound a lot cooler than swarm technology.

In the Agilox system, the robots are sent pickup requests and decide among themselves which one will honor the request based on proximity to the pickup location. Eventually, other collaborative rules can be added. For example, Agilox is introducing a bigger mobile logistics robot. If both types of robots are used in the same facility, there may be logic added to the agents about which type of robot is better suited to perform the task.

Mr. Erlacher talked about the various advantages of the Agilox robots which include the ability to make tight turns in constricted environments, a superior battery with opportunistic charging, and a light vehicle that is nevertheless able to pick up heavy loads.

But some of the core differentiators of this solution are based on the traditional advantages of agent-based technology, namely increased flexibility. "That is the reason we chose swarm intelligence," Mr. Erlacher said. While most of their customers are very large companies, Mr. Erlacher sees tremendous opportunities to sell their solution to midsized companies. These companies "don't need a huge fleet of vehicles which have long implementation times. For AGVs, 6 to 12-week implementations are common. We can implement in 8 hours for a facility that only needs 2 to 3 vehicles. We can install the charging station in an hour. And the teaching process takes one and a half to two hours." This is for a system where an operator pushes a button on a tablet or cell phone to call the vehicle. In a system where the missions come down from an ERP, WMS or MES solution, the integration will add time to the implementation.

Part of the reason this is a plug and play environment is that the vehicles routing is based solely on room contours. You bring a robot into a new environment and hit the teach button. The vehicle scans the surrounding environment in a 30-meter radius. You then drive the vehicle another 30 meters, stop it, and it does another scan. Once the entire area is scanned, the robot can share the map with other robots that will operate in the fleet. If there are seasonal surges, and new MLRs need to be added, they also can share the map their partner robots are using.

The autonomous mobile logistics robot market has some highly innovative and interesting young companies that are poised to grow very quickly. Agilox is one of those companies.

Dynamisches Routing und

Präzise Navigation

Navigation (Positionsbestimmung) und autonomes Routing im Schwarm. In diesem Bereich setzen wir uns merklich vom Mitbewerb ab und erfüllen stabil unsere hohen Anforderungen: +/- 2 Millimeter genau, nur an Hand der Raumkontur. Wenn man einerseits ein schnell anpassbares und erweiterbares System haben möchte, gleichzeitig auf enge Fördertechniken übergeben will, ist dies auch notwendig. Magnetpunkte im Boden, aufgeklebte Leitlinien oder eingemessene Spiegel an der Wand waren gestern. Wir nutzen unseren Laserscanner am höchsten Punkt des Gerätes und scannen die Umgebung 360 Grad. Und das funktioniert in den herausfordernsten Umgebungen, auch mit minimalsten bis hin zu gar keinen verfügbaren Landmarken. Routing statt projektierter Fahrspuren, das ist neu am Markt. Die optimalen Fahrspur in Echtzeit bestimmen. Routing im Schwarm bedeutet, für alle Teilnehmer, solidarisch, die bestmögliche Route zu finden, sodass die Gesamtroute aller Fahrzeuge für den Prozess optimal ist. Lineare Optimierung, quasi. Deadlocks werden dabei vermieden und gegenseitige Blockungen effizient aufgelöst. Und wenn ein Operator doch einmal eine fixe Route vorgeben möchte, weil es die Situation erfordert, ist dies ganz einfach über eine virtuell gezeichnete Linie möglich.