The robot encyclopedia organizes robots by design, abilities, and touch-based skills, helping experts compare systems and improve robotics.
Over the years, roboticists have developed many systems with different structures and capabilities. As the number of robots grows, having an easy way to learn about their characteristics, differences, and performance could be highly useful. Researchers at the Technical University of Munich (TUM) have created the “Tree of Robots,” an encyclopedia designed to simplify learning about and comparing robotic systems.
The goal of creating intelligent robots that can understand their environment like humans and perform tasks independently has existed for a long time. The development of tactile robots—systems that perceive their surroundings through touch—began about 20 years ago with the introduction of lightweight designs featuring torque sensors in every joint. Advances in technology, control systems, and response mechanisms have since led to robots that can effectively execute tasks and interact with their environment through touch.
Most existing standards for classifying robots do not consider their ability to adapt to surroundings and interact with objects through touch. This capability is essential for both safety and performance across various real-world applications.
To address the limitations of existing robot classification methods, the research team tested various systems, focusing on features that affect safety, such as detecting contact with objects. They also analyzed robotics tasks in depth, developing metrics to assess capabilities beyond safety, including performance in tactile tasks and human interaction.
The Tree of Robots encyclopedia is designed to be continuously updated, serving as a Wikipedia-like platform for robot information. It covers details from body structures to the motors and sensors used, highlighting key capabilities such as tactility fitness (sensitivity and reliability in physical interactions) and motion fitness (precision in movement).
Unlike many previous robot classification methods, the Tree of Robots encyclopedia clearly defines the specialized capabilities of different systems. It also categorizes robots into three main groups based on tactility fitness, measuring their suitability for specific tasks.
The team’s encyclopedia could support future research by helping computer scientists and roboticists identify the best systems for testing their algorithms. They also plan to expand the Tree of Robots with more robotic systems and additional metrics.
Reference: Robin Jeanne Kirschner et al, Categorizing robots by performance fitness into the tree of robots, Nature Machine Intelligence (2025). DOI: 10.1038/s42256-025-00995-y.
