Learning the min-max gait comfort region when wearing shoes

Shoe-embedded sensors are a recent, common and convenient choice for devices intended for locomotion-related applications. They are advantageous, compared to other wearable devices, since they allow performing gait assessment in real-world environments, recognizing the user walking pattern in real-time and directly on the feet level. However, shoe with embedded sensors, while providing affore-mentioned advantages, must in no way disturb the walk. In this paper, we present a simple, reliable and cheap method to check whether, and in which manner, the shoe affects the gait. The ultimate goal of the developed methodology is to provide a way to assess the performance of an innovative sole adaptable to changes in the external environment and in the dynamic state of the user. The proposed method is based on using a 2D camera to monitor three geometrical angles (knee, ankle, shoe) which reflect, in a simple but exhaustive manner, the lower limb behavior during the gait cycle. The ability of the shoe not to affect the normal gait is tested by comparison between the three angles trends of a person wearing the shoe and the person's min-max gait comfort regions, with respect to the three angles. Min-max comfort regions are learned through real experiments on different subjects and shoes.