A low-cost human locomotion speed recognition for augmented virtual environments exploration

This paper presents a novel and low-cost interface designed for real-time human locomotion speed recognition, which fits with the exploration of kinesthetic virtual environments (VE). According to the interface paradigm, the human locomotion recognition feeds VE navigation control. An experimental session has been organized in order to acquire acceleration data related to locomotion of 10 healthy subjects (men and women) aging between 23 and 35 years. A treadmill has been used to capture the velocity at which subjects were moving. Our system was designed to optimize classification performances in human locomotion speed recognition in real-time. The recognized human speed locomotion has been shown to enhance users' sensation of presence in the virtual environment. A simple scenario has been developed to assess the system functionality. The experiments carried out show that our system is excellent at classifying a wide range of human locomotion and can be used both in virtual and augmented reality (VR) environments for improved interaction.