Indoor Positioning Method using Time-Series Data of Multiple Human Sensors

In fry tuna farming, positioning of indoor workers around the cage is required to investigate the effects of workers' behavior on fly tuna. In the indoor positioning around the fry tuna cage, the following points need to be considered: (1) Obstacles around the cage make it difficult to use cameras; (2) mobile devices for positioning should not be used because fry might be stressed if the device carried by the worker falls into the cage. This paper proposed the indoor positioning method that only uses data from human sensors which are commercially available. Concretely, our method realize indoor positioning considering the following conditions: (i) whether the estimated point is inside or outside of the detection range of each sensor; (ii) for the time when the sensor data changes from 0 to 1 or 1 to 0, the estimated point is near the limit of detection range of the sensor; and (iii) when the worker is moving at a constant speed, the moving distance of the worker is proportional to the number of sensor data acquired. From the simulation, it is revealed that the solution candidates obtained by the proposed method is close to the actual movement path.