AMR design revisited with AI algorithms
Article By : Majeed Ahmad
High-resolution camera takes AMR design to a new path-finding mechanism to ensure greater flexibility and safety on the hall floor.
Mobile robot platforms, commonly known as autonomous mobile robots (AMRs), are seen as the next big thing in warehouses and production floors. However, the recent incidents have highlighted the downside regarding their work alongside humans while handling modern logistics.
The common modus operandi in AMR design has been centered around spatial adaptations to enable robots to travel on specified paths. Here, optical markings help robots orient themselves around specific paths and intersections. In other words, they require a preinstalled guidance system to find the right way. And here comes the rub: irregularities in the hall floor.
A Nuremberg-based startup founded in 2016 by robot experts claims to have a solution based on an entirely new AMR design concept. AMRs from Evocortex simply register irregularities on the floor using a high-resolution camera located on the vehicle’s underside. The camera captures an area of 10 x 10 centimeters and uses the image data to produce the equivalent of a fingerprint of the hall floor.
Figure 1 The wheel module of the EvoRobot is fitted with an IEF3 encoder and a 38/2 S gearhead. Source: Evocortex
Robots move across the hall floor in a grid pattern using complex self-learning artificial intelligence (AI) algorithms that help create a highly precise map based on a pattern of individual points. Here, the controller monitors the vehicle’s own movement. It allows the robot to determine its path in a theoretical area of one square kilometer, precisely positioned within a millimeter.
New scratches on the floor are added to the map, while features that disappear are removed after a certain time. Moreover, robots can be equipped with additional LiDAR sensors on one or two fronts to scan the room in the direction of travel and detect obstacles. The robot stops immediately to ensure the safety of employees.
Figure 2 Barrel-shaped rollers are mounted on the wheel’s rim to give the wheel a continuous rolling surface. Source: Evocortex
The wheel module in Evocortex’s robots is equipped with DC motors from Faulhaber. These wheel motors are also equipped with a brake for fast stops and to ensure that the wheels remain locked in position once stopped. So, wheel modules, equipped with powerful brushless motors, are crucial in AMR’s ability to offer greater flexibility.
This article was originally published on EDN.
Majeed Ahmad, Editor-in-Chief of EDN and Planet Analog, has covered the electronics design industry for more than two decades.
