We are developing a method to acquire position information of a cow outdoors using Received Signal Strength Indicator (RSSI) of Bluetooth Low Energy (BLE). As existing research, there is a localization method using fingerprint database as learning data in deep learning. However, that method has the problem that it costs to create a database by measurement in a vast outdoor environment. Therefore, we considered to build a part of the fingerprint database using virtual space modeling received power measurement environment in a pasture. Experimental results showed that an average distance error to GPS data is about 6 m by training DNN using the database and additionally training DNN using actual GPS data.

In order to detect estrus and abnormalities from the interactions of grazing cattle, we are developing a position estimation method using low-power wireless devices. In this paper, in order to obtain a natural trail of cow's locations, we propose a localization method based on long short term memory. Our evaluations show that the proposed method suppresses unnatural trail and achieves the average location error of 5.25 m for the cow that is used for learning and about 6 m for the other cows.

No abstract