Human-in-the-loop optimization of wearable robots to reduce the human metabolic energy cost in physical movements

Abstract

Most designs of wearable robots are based on human biomechanical statistics, engineering experience or individual experiments. Despite great successes, few of them consider the human–robot integration and individual differences between users. Additionally, the design periods, cost and safety also need to be further improved. Learning from the natural driving mechanism of human body, we propose a general human-in-the-loop (HIL) optimization designing approach for this kind of wearable robots. Firstly, the human–robot coupling model of the personalized wearable robot and the human musculoskeletal model are established. Then, the Computed Muscle Control (CMC) tool embedded in software OpenSim and the Bayesian optimization used in machine learning are combined to find the optimal design scheme for the personalized wearable robots to reduce the human metabolic energy cost in specific physical movement. The HIL approach could not only optimize the control parameters of wearable robots, but also optimize their geometry, material and any other design parameters flexibly and effectively. An application example for the HIL approach is also provided to help designers better understand and use the HIL method proposed in this paper.