New "Tree of Robots" Taxonomy Helps Evaluate and Compare Robotic Sensitivity for Better Human-Robot Interaction

Researchers at the Munich Institute of Robotics and Machine Intelligence (MIRMI) at the Technical University of Munich (TUM) have developed a new evaluation framework called the “Tree of Robots” to measure the sensitivity of autonomous robots. This taxonomy aims to standardize the assessment of robotic manipulators, which is critical for ensuring safe and effective human-robot interaction. Current industry standards lack a consistent method to evaluate the sensitivity of robots, making it difficult for companies to choose the most suitable systems for their needs. The TUM MIRMI team, led by researchers Kübra Karacan, Robin Kirschner, and Alessandro Melone, tackled this issue by creating a comprehensive set of 25 metrics that describe how well a robot can interact physically with its environment. These metrics cover aspects such as force application, touch detection, and injury prevention during contact with humans. The Tree of Robots is inspired by Charles Darwin’s Tree of Life, visually depicting the diversity and evolution of robotic systems. Each robot is evaluated using a spider diagram, which provides a quick and intuitive overview of its capabilities in tactile and motion performance. This approach helps categorize robots into distinct types, including industrial robots, cobots, soft robots, and tactile robots, based on their performance metrics. Industrial robots are typically designed for strength and resilience, making them ideal for repetitive tasks in manufacturing settings. Cobots, on the other hand, are geared towards sensitive and flexible interaction with humans and are often used in collaborative environments. Soft robots are characterized by their flexibility and adaptability, suitable for applications involving delicate objects or irregular surfaces. Tactile robots excel in tasks requiring precise touch and force control, such as surgical procedures. The new framework combines existing motion metrics with novel tactile metrics to offer a holistic view of a robot’s performance in physical interactions. Lab director Robin Kirschner explained, “We combine existing motion metrics with our new tactile metrics to provide the first-ever overview of the totality of the basic capabilities for physical interactions of a robotic system.” In February 2025, the team published their findings in the prestigious journal Nature in a paper titled “Categorizing robots by performance fitness into the tree of robots.” Prof. Lorenzo Masia, executive director of TUM MIRMI, expressed confidence that the AI Robot Safety & Performance Centre at TUM MIRMI will evolve into a leading national testing center for robotics, further advancing the field. The significance of the Tree of Robots extends beyond academia, with potential practical applications in industry. Companies seeking to integrate robotics into their operations will benefit from a standardized approach to evaluating robot performance. Prof. Achim Lilienthal, deputy director of MIRMI and professor of intelligent systems perception, highlighted the potential impact: “Such a standard could pave the way for a quality seal for the rapidly expanding robotics market. Knowing the performance capabilities of a robot system is an enormous help to industry, where companies want to make the best possible use of robotic systems.” The research at TUM MIRMI is overseen by a five-member board of directors, which includes experts in robotics and machine intelligence. Their roles encompass guiding the institute’s research and innovation strategy, core teaching content, and the establishment of new focus groups. This structured leadership ensures that the Tree of Robots and other innovations from TUM MIRMI are rigorously tested and validated, enhancing their reliability and applicability in real-world scenarios. Industry insiders and experts praise the initiative, noting that it addresses a critical gap in the robotics market. A standardized taxonomy not only aids in the selection and deployment of robotic systems but also drives competition and innovation, as manufacturers strive to meet and exceed the established metrics. This could lead to the development of more advanced, versatile, and user-friendly robots, ultimately accelerating the adoption of robotics across various sectors. With its robust methodology and clear visualization tools, the Tree of Robots is poised to become a cornerstone in the evaluation and certification of robotic systems. As the robotics market continues to grow, TUM MIRMI’s contributions could set a global precedent for quality and safety standards.