IEEE Trans Neural Syst Rehabil Eng. 2020 Jul;28(7):1573-1583. doi: 10.1109/TNSRE.2020.2989481.
Benchmarks have long been used to verify and compare the readiness level of different technologies in many application domains. In the field of wearable robots, the lack of a recognized benchmarking methodology is one important impediment that may hamper the efficient translation of research prototypes into actual products. At the same time, an exponentially growing number of research studies are addressing the problem of quantifying the performance of robotic exoskeletons, resulting in a rich and highly heterogeneous picture of methods, variables and protocols. This review aims to organize this information, and identify the most promising performance indicators that can be converted into practical benchmarks. We focus our analysis on lower limb functions, including a wide spectrum of motor skills and performance indicators. We found that, in general, the evaluation of lower limb exoskeletons is still largely focused on straight walking, with poor coverage of most of the basic motor skills that make up the activities of daily life. Our analysis also reveals a clear bias towards generic kinematics and kinetic indicators, in spite of the metrics of human-robot interaction. Based on these results, we identify and discuss a number of promising research directions that may help the community to attain a comprehensive benchmarking methodology for robot-assisted locomotion more efficiently.
基准一直被用于验证和比较许多应用领域中不同技术的就绪水平。在可穿戴机器人领域，缺乏公认的基准测试方法是一个重要的障碍，可能会阻碍研究原型向实际产品的高效转化。同时，越来越多的研究正在解决量化机器人外骨骼性能的问题，导致方法、变量和协议呈现出丰富而高度异构的图景。本综述旨在组织这些信息，并确定最有前途的性能指标，可以转化为实际的基准。我们的分析重点是下肢功能，包括广泛的运动技能和性能指标。我们发现，一般来说，下肢外骨骼的评估仍然主要集中在直线行走上，日常生活活动中大多数基本运动技能的评估都很差。我们的分析还揭示了尽管存在人机交互指标，但对通用运动学和动力学指标仍存在明显的偏见。基于这些结果，我们确定并讨论了一些有前途的研究方向，这可能有助于社区更有效地为机器人辅助运动建立全面的基准测试方法。
