KUMITRON: Artificial Intelligence System to Monitor Karate Fights that Synchronize Aerial Images with Physiological and Inertial Signals

Abstract

New technologies make it possible to develop tools that allow more efficient and personalized interaction in unsuspected areas such as martial arts. From the point of view of the modelling of human movement in relation to the learning of complex motor skills, martial arts are of interest because they are articulated around a system of movements that are predefined -or at least, bounded- and governed by the Laws of Physics. Their execution must be learned after continuous practice over time. Artificial Intelligence algorithms can be used to obtain motion patterns that can be used to compare a learners’ practice against the execution of an expert, as well as to analyse its temporal evolution during learning. In this paper we introduce KUMITRON, which collects motion data from wearable sensors and integrates computer vision and machine learning algorithms to help karate practitioners improve their skills in combat. The current version focuses on using the computer vision algorithms to identify the anticipation of the opponent's movements. This information is computed in real time and can be communicated to the learner together with a recommendation of the type of strategy to use in the combat.