A Multi-Indicator evaluation method for Human-Machine effectiveness of lower limb wearable exoskeleton

The human-machine effectiveness evaluation is crucial in designing and optimizing lower limb exoskeletons. However, the current single-indicator evaluation method lacks a unified standard and it is one-sided. Therefore, a more objective and comprehensive multi-indicator evaluation method was proposed in this study. Firstly, some single indicators of human-machine effectiveness evaluation were integrated, and a comprehensive evaluation system was constructed accordingly. Then, multi-task gait experiments were designed to collect the evaluation indicators, and the dataset was constructed by processing the collected indicators. Finally, the dataset was clustered and analyzed based on the self-organizing map (SOM) neural network. The results demonstrated that evaluating human-machine effectiveness by multi-indicator was more reliable and comprehensive than singleindicator and this superiority was verified by abnormal and fuzzy samples. It compensates for the one-sided of single-indicator and provides a more reliable theoretical reference for the design and optimization of the lower limb exoskeleton.