Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Challenges in modern automated Feeding Systems

Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan

Citation:  2018 ASABE Annual International Meeting  1801101.(doi:10.13031/aim.201801101)
Authors:   Matthias Reger, Johannes Friedrich, Sascha Wörz, Jörn Stumpenhausen, Anton Sieghart, Heinz Bernhardt
Keywords:   AFS, Automation, Autonomous Navigation, Dairy Farming

Abstract. Over 7.800 automatic milking systems (AMS) are installed in Germany, with an upward trend [1, 2]. About 50 to 70 % of new installed milking systems in Germany are AMS [3]. Based on this development a similar development for AFS (automated Feeding Systems) seems likely or can be expected. Especially time and labour savings for the farmer and nutritionally aspects on cows are driving forces behind this automation step.

Every commercially available AFS is a partly automated feeding system. These are usually not self-propelled and by now not capable to autonomously take fodder in. This limits the operating range of these systems, since human intervention is necessary in relatively short intervals. Some manufacturers already presented studies of their fully automated feeding systems emphasizing the future trend towards self-propelled AFS.

In case of self-driving vehicles free navigation and safety are essential. Especially the various environments on farms can be crucial and challenging for machine operations. For the technical implementation durable sensing is required. Under practical conditions a novel radar scanner and a laser 2D-scanner were tested at a dairy farm. The generated data has been pre- and post-processed for accurate and precise mapping and navigating. Therefore a high-complex physical system has been designed which is used in an Extended Kalman filter application. The sensory proved to be a suitable solution for automation tasks, as well as safety matters. Likewise the physical system performed well and was capable to improve already Monte Carlo filtered data. These preliminary tests are essential ingredients in a self-driving AFS.

(Download PDF)    (Export to EndNotes)