Terrain recognition using neuromorphic haptic feedback

Recent years have witnessed relevant advancements in the quality of life of people with lower limb amputations thanks to technological developments in prosthetics. However, prostheses providing information about the foot-ground interaction, in particular about irregularities in terrain structures are still missing on the market. Lacking tactile feedback from the foot surface, subjects might step into uneven terrain without noticing, increasing the risk of falling. Here, this issue is addressed by evaluating in intact subjects a biomimetic unilateral haptic vibrotactile feedback conveying information about discrete gait events and terrain features relying on the readings of an integrated insole. After shortly experiencing both even and uneven terrains, subjects discriminated them with an accuracy of 87.5%, solely relying on the replay of the vibrotactile feedback. Via a machine learning approach, we estimated that the subjects achieved such performance taking into account a temporal resolution of 45 ms. This work is a leap forward in bringing lower-limb amputees to appreciate the floor conditions while walking, to allow adapting the gait and promoting a more confident use of the artificial limb.