Choisissez votre emplacement:

Pays

RoboMove 3D

With the RoboMove project, KUKA is conducting research together with InnoSenT and Fraunhofer EMFT on intelligent 3D distance measurement and obstacle detection on robot arms in motion.


Intelligent distance measurement and obstacle detection

We know RADAR-based distance measurements to prevent a rear-end collision in a reliably functioning manner from our passenger car. So why shouldn't we also use this technology for robots to ensure collision-free operation? We are discussing this topic in the research and funding project RoboMove 3D, which is funded by the Bavarian state government for three years (2021 - 2023).

Unfortunately, the direct transfer of the functional principle from the automotive industry to robotics is not so easy. The problem is, first of all, that there are a wide variety of technical and scientific questions to be discussed. These include, for example, how dynamically moving RADAR sensors behave in relation to each other, or how a RADAR sensor can be used to safeguard our robots in safe technology.

To solve these questions, among others, we have brought strong partners to our side. The radar experts at InnoSenT GmbH develop sensors for robotics specifically according to our requirements. Compared to typical automotive sensor technology, our application places different demands on the technology. Among other things, we need significantly faster sensor data processing in the range of a few milliseconds and high sensor resolution in the near-field range. For reliable and safe collision avoidance, we also need to find out how radar detection can be optimally integrated and precisely executed in safe technology.

Logo InnoSenT

The aim of our second partner, the Fraunhofer EMFT, is to integrate the complete RADAR sensor technology on or in bendable or stretchable foils. This would allow these new, thin RADAR modules to be optimally adapted to robot arms and also to be used in a variety of other applications.

Our KUKA corporate research team will simulate and fuse RADAR sensors and implement novel algorithms for environmental modeling and motion prediction based on RADAR data. An important step in the work is to consider scaling from small-scale robotics to larger-scale models that are more capable of carrying loads. Research is needed to determine how and how many RADAR sensors need to be placed on different robot models to ensure perfect environmental sensing. After all, this technology could enhance any robot.