Lumbar Exoskeleton, as an important instance of wearable exoskeleton, has broad application prospects in logistics, construction, and other industries. Specifically, in the working scenarios that require long-term and repeated bending and rising movements, active lumbar exoskeleton (ALE) can provide effective protection and flexible assistance to wear’s waist muscles and bones, which will significantly reduce the risk of lumbar muscle strain. How to improve the human-machine coupling and enhance the assistance performance are the main challenges for ALE’s development. Based on the biomechanical analysis of the movement of lifting heavy objects from bottom up, this paper proposes a lightweight but powerful ALE, named as SIAT-WEXv2, which can output maximum assistive force of 28 N. Additionally, we use robust fuzzy adaptive algorithm to improve SIAT-WEXv2’s antidisturbance ability, so that it can provide continuous and supple assistance for wearer. Electromyography (EMG) signals of the lumbar erector spinae (LES) from ten subjects in two experimental cases (with or without SIAT-WEXv2) were collected to evaluate the effectiveness of our new ALE. The experimental results indicate that the reduction of iEMG signal at LES decreased monotonically from 60% ± 5.5% to 40.5% ± 6.5% as the weight of lifting load increased from 0 to 25 kg.
A New Fuzzy Adaptive Algorithm to Classify Imbalanced Data
