Wearable Sensor Optimized Layout based on Swarm Intelligence for Activity Recognition
Chengshuo Xia, Yuta SugiuraWearable intelligent systems that recognize the daily activities of humans have significantly contributed to many useful applications, such as health monitoring, rehabilitation training, etc. However, in practical wearable applications, the target individuals may have different body conditions and demands. Thus, the conventional fixed-position interface is not suitable for all potential users. Moreover, the sensors may return different information depending on their placement, which determines the quality of a recognition system to some extent. This paper investigates the influences that different numbers and placements of wearable accelerometers have on a human activity recognition system. Seventeen different body placements and 10 human activities are examined in this research. Using the support vector machine classifier, the designed multistage and multiswarm discrete particle swarm optimization algorithm is applied to explore the best sensor combinations and accuracy trends corresponding to the different requirements of sensor numbers. With the proposed optimization scheme, a relevant recognition system can be designed based on the demands and physical condition of the subject.