TUM Develops Test Methodology for Robot Sensitivity

Munich researchers create first standardized assessment of industrial robots' physical interaction capabilities – 25 measured values define tactility.

Breakthrough in robotics evaluation at the Technical University of Munich (TUM): Researchers at the Munich Institute of Robotics and Machine Intelligence (MIRMI) have developed a standardized testing system that makes the sensitivity of robots measurable for the first time. The new evaluation scheme is intended to serve as a seal of quality for the dynamic robotics market and help industry make targeted system selections.

Based on Charles Darwin's tree of life, the researchers developed a “tree of robots” to illustrate the fundamentally different adaptations of different robot “species” to their respective areas of application. This was based on the basic skills a robot needs in processes: for example, how well it follows a given path, how well it can take up a position, how gently it can make contact with surfaces, and how safely it can handle collisions between robots and humans.

25 Metrics for Robotic Tactility

For what is known as tactility, the researchers developed 25 metrics that describe how sensitively a robot acts in physical contact with its environment. Among other things, the aim is to find out whether the force applied to a surface is higher than intended or whether a robot can cause injury to humans. The pattern that emerges in a spider diagram based on the 25 measured values shows at a glance how sensitive the respective robot is.

Depending on the performance of the individual systems, the researchers divide the robots into the classes “industrial robots,” “cobots,” “soft robots,” and “tactile robots.” The reason: precision is the most important requirement for a robot used in surgery, while strength and resilience are more important for use in warehouses or production.

Potential for Industrial Testing Standard

“The TUM MIRMI test methodology has the potential to establish itself as an industrial testing standard,” explains Achim Lilienthal, deputy director of MIRMI and professor of perception of intelligent systems. This paves the way for establishing a seal of quality for the highly dynamic robotics market. “Knowing the performance capabilities of a robot system is an enormous help to industry, which wants to use robotic systems as effectively as possible.”

This is a partly automated translation of this german article.