Automated milking systems make dairy operations more efficient and less stressful for animals and farmers alike. However when a problem occurs, a suspect animal must be found and diagnosed quickly. This effort aims to evaluate the feasibility of a Q-Track Milk Operations Optimization (Q-MOO) system. The proposed Q-MOO system will employ a novel, low-frequency real-time location system (RTLS) that exploits Near-Field Electromagnetic Ranging (NFER) technology. NFER RTLS has the potential to provide high accuracy location data of animals at a reasonable cost even in complicated RF propagation environments such as when animals are clustered in close proximity or when line-of-sight to a receiver is obscured by metal fences, shelters, or structures. This effort also aims to take a more comprehensive look at the value of location data in optimizing dairy operations. Teaming with dairy experts at Auburn University, Q-Track will explore whether location awareness can further enhance the safety and productivity of dairy operations by correlating animal locations, associations, activity levels, and other innovative ways location data can add value. Our commercialization partners anticipate the proposed system will be economically viable by reducing the labor required to find a suspect cow and speeding intervention in the context of automated milking systems. We will further explore additional value and benefits in a trial at Auburn University's E.V. Smith Dairy Research Center. OBJECTIVES: The ultimate objective of this project is to develop, deploy, and evaluate a prototype Q-Track Milk Operations Optimization (Q-MOO) System in an extended trial of about one month's duration. Q-Track will work closely with academic researchers at Auburn University and representatives of Lely to evaluate the value and commercial potential of the proposed system. The questions needed to establish the technical feasibility of the proposed approach include: What accuracy can the system obtain and what accuracy is needed What range can be obtained and how much infrastructure will be needed What will be the cost of the system What update rate can the system obtain, and what update rate is needed What value can dairy cow location data offer APPROACH: TASK #1: DEVELOP PRELIMINARY REQUIREMENTS AND SPECIFICATIONS Our initial expectation is that a Q-Track Milk Operations Optimization (Q-MOO) system must meet the following specifications: Accuracy Goal: 1m so the location of an individual animal may be determined with respect to another animal, or watering, feeding, or milking station. Cost Goal: We estimate the system will add $12 to the cost of existing cattle monitoring tags and will require about $5000-$10,000 in infrastructure for a typical 200 cow dairy. Durability: The system must be able to withstand extremes of outdoor weather and be mechanically ruggedized to handle rough treatment by tagged animals. The system must further be reliable enough to operate with minimal maintenance and intervention. Compactness: The system must be small and non-intrusive so as not to stress tagged animals. Flexibility: The system must allow for easy access to data and analysis of the data collected. Q-Track will collaborate with Auburn and Lely to finalize preliminary requirements and specifications. TASK #2: MODIFY HARDWARE Q-Track has implemented a weatherized receiver. We will review and implement any special arrangements necessary for installing the receiver. We will develop a suitable weatherized tag enclosure for our existing compact tag and devise a better means of securing it to a cow collar. TASK #3: INTEGRATE SYSTEM When the individual components are ready, we will integrate them with a custom software application. We will perform an extended outdoor test of the system at our Huntsville office to ensure that it is ready for deployment. TASK #4: DEPLOY PROOF-OF-CONCEPT SYSTEM We will take the system developed in Tasks#2 and #3 and deploy it at Auburn University's E.V. Smith Dairy Research Center. This approximately 25,000 sqft facility currently houses about 120 cows and has a capacity of about 200. TASK #5: EVALUATE ACCURACY Concurrent with the deployment of Task #4, we will evaluate the accuracy of the location system. As part of our proof-of-concept test, we will generate histograms of the error distribution as well as plots of the cumulative location error. TASK #6: PERFORM ONE-MONTH TRIAL We will deploy the system tracking at least twelve cows for a period of a month. In this longer term trial of the prototype Q-MOO system, we will similarly assess the accuracy of the location tracks by examining known choke points - gates, water or feed troughs, and the milking parlor to confirm that the tracking system is yielding reliable and repeatable results. TASK #7: EVALUATE RESULTS We will collaborate with Auburn lead, Prof. Keith Cummins, to analyze the results. Prof. Cummins is particularly interested in looking at the accuracy of the data, reproductive issues with cows that spend time together, and how dominance order affects the cows' responses to management of the herd. TASK #8: FINAL REPORT Q-Track will produce a final report that describes the performance of the prototype Q-MOO and the results of the Auburn study
