Automatic milking: an experiment to inspect teats using an electronic nose

Abstract

Methods are required for automatically detecting the hygienic condition of cows' teats before teat cups are applied by an automatic milking system. The dynamic characteristics and the odour discrimination capability of organic conducting polymers were investigated as a potential method for examining teats before milking. A controlled air flow system, an array of sensors and automated data acquisition were used to detect odours emitted by wet muck, dry muck, blood and water placed on the surface of an artificial teat. Canonical variate analysis of the output of 20 sensor elements subjected to a total of 28 random replications of the experimental procedure was able to discriminate between four treatments and the control (P = 0.1). Further work is required to adapt this laboratory system to detect teat contamination in field conditions.