A Multi-terrain Adaptive Feed-Forward Control Walking Algorithm for Lifting Exoskeleton

Lift up exoskeletons have been used in industrial and military area. However, these exoskeletons often do not have the walking assistance algorithm. In factories, warehouses or common outdoor environments, workers usually not work in fixed position. The need of workers moving or carrying load is clear, but this need can not be achieved with most of exoskeletons. In this paper we introduce a multi-terrain adaptation algorithm which can identify the required assistance interval according to different terrains and generate corresponding assistance curves. The algorithm can automatically adapt to all usage scenarios of lift up exoskeletons, and can smoothly switch between different terrains. The core idea of the algorithm is that human body make corresponding actions based on the terrain, by monitoring human movement, the algorithm can generate the desire curve for each step.