A recent review in the Chinese journal Robot examines the state of safety testing and evaluation for human-robot interaction (HRI) in non-medical intelligent robots, focusing on collaborative and wearable robots. The authors identify three critical bottlenecks: unclear HRI safety impact mechanisms, lack of test methods and standards, and missing test instruments and calibration methods. These gaps hinder the industrialization and scaling of intelligent robots.
Key takeaways
- HRI safety evaluation parameters for collaborative robots follow ISO/TS 15066, covering physical, information, and other interaction safety, with sub-items for each.
- Wearable robot safety is assessed across three states: general interaction (e.g., mechanical limits, torque monitoring), biomechanical constraints (e.g., force distribution, comfort), and human-robot coupled interaction (e.g., joint angle matching, control precision).
- Contact modes include quasi-static, transient, dynamic continuous, drag-teaching, and indirect contact, each requiring specific force and pressure limits based on biomechanical data.
- Existing test instruments are mostly foreign (e.g., Pilz, GTE, Tekscan), with domestic sensors emerging but no fully localized dedicated HRI safety testers.
The review calls for systematic research into multi-modal interaction mechanisms and complex-scenario risk assessment to build comprehensive safety evaluation systems, supporting broader adoption in manufacturing and service robotics.
Source: 《机器人》期刊 (robot.sia.cn) · Published 2026-01-13 · “非医用智能机器人人机交互安全测评技术现状及趋势综述”