Progress 07/01/23 to 06/30/24
Outputs
Target Audience:During this funding year, our target audience included the public, researchers and practitioners in the dairy industry, academics and students in electrical engineering, computer science, and agricultural engineering, as well as state legislators. We proactively engaged with Wisconsin legislators, state leaders, UW alumni, and the public at the Wisconsin State Capitol during an exclusive, invitation-only session of the annual Research in the Rotunda event. At this event, one of our students presented the current research progress and its real-world application to Wisconsin's dairy industry. The presentation detailed the development of a wireless wearable localization system designed to monitor real-time behaviors related to heat stress in dairy cattle. It effectively highlighted the research's goal of addressing the increasing challenge of heat stress in the dairy industry through methods that can be easily implemented by dairy producers. The University of Wisconsin's media highlighted a news article focused on this project, emphasizing that the newly developed ear tags are an effective tool for measuring temperature in real time. The article acknowledged our work on creating low power, embedded, implanted, and wearable devices, which have potential applications across various fields. It was crafted for an audience that includes current students, alumni, faculty, and staff. We enhanced the curriculum for the Heat and Mass Transfer course (BSE 464) in the Department of Biological Systems Engineering to include comprehensive theoretical foundations of modern computational methods and a series of computational fluid dynamics projects. The course now incorporates the latest findings from our current project, particularly the use of wireless sensors developed during this research and the significant role of big data derived from them. These data sets are utilized for modeling computational simulations of advanced barn designs to improve airflow patterns, cool dairy cows during heat waves to mitigate heat stress, and develop control systems using variable speed fans with neural-network-based smart decision-making tools. Additionally, we introduced the latest research outcomes in a freshman- and sophomore-level course on computer-aided design and engineering simulations (BSE 270) to help agricultural engineering students understand how advanced computing technologies can transform the future of animal well-being and the nation's food supply. ?Aiming at a highly technical audience comprising both academic and industry professionals, we meticulously prepared and presented our findings at prestigious national and international conferences, including the Annual International Meeting of the American Society of Agricultural and Biological Engineers, the International Conference on Mobile Computing and Networking (MobiCom), and the Conference on Neural Information Processing Systems (NeurIPS). Our peer-reviewed articles were also published in competitive journals such as Biosystems Engineering and the proceedings of MobiCom. Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided?We have created a series of benchmark projects for both graduate and undergraduate students that utilize computational fluid dynamics (CFD). These ten projects encompass simulations of laminar and turbulent flows, along with heat and mass transfer. Incoming graduate students are expected to learn about mesh generation, various turbulent flow models, and flow dynamics over complex shapes through training sessions centered around these benchmark problems. At UW-Madison, the BSE 464 course covers the fundamentals and advanced theories of heat and mass transfer, incorporating these projects throughout the semester. Additionally, BSE 270 introduces the principles of computational fluid dynamics alongside computer-aided design (CAD). Both classes explore CFD- and sensor-based engineering solutions for animal housing, global warming, and the mitigation of heat stress in dairy cows, drawing on current research. Both graduate and undergraduate students are encouraged to showcase their research activities at national and international conferences, graduate seminars, and weekly group meetings. As a result, they presented both computational and experimental findings at the 2023 International Conference on Mobile Computing and Networking (MobiCom), the 2024 ASABE Annual International Conference, and the BSE 900 Graduate Seminar. Additionally, a presentation has been accepted at the highly competitive International Conference on Neural Information Processing Systems (NeurIPS). How have the results been disseminated to communities of interest?Together with a group of Wisconsin's dairy extension educators, we initiated a series of outreach activities funded by state resources. We will introduce an innovative ventilation system designed to effectively alleviate the heat stress commonly experienced in dairy buildings, which has been partially developed through our current research efforts. We will highlight the importance of these research initiatives, given the substantial economic impacts that heat stress has on dairy farmers and the welfare of cows. Without appropriate measures grounded in engineering and science, we anticipate that significant losses will continue to escalate as extreme weather events become more frequent and average summer temperatures rise. What do you plan to do during the next reporting period to accomplish the goals?In Year 4, we will create an ultra-lightweight, low-power ear tag designed to measure the 3D location of cows using Ultra-Wideband (UWB) technology. Our objective is to showcase the effectiveness of non-invasive behavior monitoring through real-time location tracking for detecting and mitigating heat stress. We will also carry out a field deployment, leveraging the insights and lessons learned from Years 2 and 3 to determine the optimal configuration and algorithm for the indoor localization system, while further refining the installation locations of the UWB anchors. Furthermore, we will create and validate a new model to accurately detect heat stress by analyzing changes in cattle behavior, delivering actionable insights to improve our CFD-optimized cooling system. Our emphasis will be on precision cooling systems that have recently been proposed and designed to reduce greenhouse gas emissions and energy consumption while effectively managing animal housing during heat waves. We will also implement a fully functional closed-loop control system that includes behavior monitoring for heat stress detection and precision cooling for heat stress mitigation. To prepare for this deployment, we are designing and constructing a cooling system featuring three plenums that will deliver independently controlled fresh air flows to cows in various areas of the pen. We will use the heat stress data from our sensing models alongside the location information from the UWB system to optimally direct cooling airflow to the cows experiencing heat stress. The cooling system will undergo thorough testing prior to its deployment in the summer of Year 4. We anticipate that this system will significantly reduce energy and water consumption while effectively mitigating heat stress.
Impacts
What was accomplished under these goals?
In Year 3, we concentrated on analyzing the large-scale multimodal data collected in Year 2 to gain insights into how heat stress impacts cattle behavior. We employed various computational and machine learning techniques to process the data and extract valuable information that could help us understand changes in cattle behavior. We identified seven behaviors that are well-known indicators of cattle well-being: walking, standing, feeding with their heads up, feeding with their heads down, drinking, licking mineral blocks, and lying down. Initially, we extracted information from the raw data, including 3D location, head direction, and leg orientation. We also developed a comprehensive set of ground truth data for cattle behaviors, cow identities in visual data, and their body positions within the pen. Subsequently, we examined the statistical relationships between changes in cattle behavior and ambient conditions, particularly during instances of heat stress. Our findings revealed strong correlations between various behavioral changes and the ambient temperature and humidity levels inside the barn. We have advanced the development of a cutting-edge ventilation system for dairy barns by utilizing Computational Fluid Dynamics (CFD) and IoT sensors. This innovative design solution tackles common issues found in traditional mechanical ventilation methods and improves cooling efficiency through precision cooling. Our approach combines computational techniques with wireless monitoring to optimize design, supported by experimentally validated CFD simulations. These simulations provide valuable insights into the precision air supply system under various real-world conditions. The findings and methodology of this study are crucial for the widespread adoption of future barn systems, offering a novel design path through cost-effective virtual experimentation. This can significantly influence the future development of dairy barn ventilation systems, addressing both economic and animal welfare concerns.
Publications
- Type:
Conference Papers and Presentations
Status:
Accepted
Year Published:
2024
Citation:
Trey Standiford, Hien Vu, Dimuth Panditharatne, Hanwook Chung, Younghyun Kim, Christopher Choi. (2024) Development of a wireless, multi-modal wearable IoT system to monitor wellbeing of dairy cows, ASABE Annual International Conference.
- Type:
Conference Papers and Presentations
Status:
Accepted
Year Published:
2024
Citation:
Hanwook Chung, Hein Vu, Dimuth Panditharatne, Trey Standiford, Younghyun Kim, Christopher Choi. (2024) Monitoring Responses to Heat Stress and Corresponding Resting Behavior of Dairy Cows using Ultra-wideband (UWB) Sensors, ASABE Annual International Conference.
- Type:
Journal Articles
Status:
Published
Year Published:
2024
Citation:
Hien Vu, Hanwook Chung, Christopher Choi, and Younghyun Kim. 2024. eTag: An Energy-Neutral Ear Tag for Real-Time Body Temperature Monitoring of Dairy Cattle, Proceedings of International Conference on Mobile Computing and Networking (MobiCom).
- Type:
Conference Papers and Presentations
Status:
Accepted
Year Published:
2024
Citation:
Hein Vu, Omkar Prabhune, Unmesh Raskar, Dimuth Panditharatne, Hanwook Chung, Christopher Choi, Younghyun Kim. (2024) MmCows: A Multimodal Dataset for Dairy Cattle Monitoring, the Conference on Neural Information Processing Systems (NeurIPS).
|
Progress 07/01/22 to 06/30/23
Outputs
Target Audience: We have reached out to U.S. dairy industry personnel who serve as supporting advisors to dairy farmers on topics relating to heat stress, heat abatement, and ventilation, including dairy consultants, nutritionists, veterinarians, extension educators, and fan/ventilation-system manufacturers and distributors. These audiences were reached through both in-person workshops and webinars delivered by Co-PI Van Os. A senior capstone design team in the Biological Systems Engineering (BSE) Department developed and tested a controlled environmental monitoring system based on the IoT-based wireless data acquisition unit. We have been reaching out to the producers via the BSE extension specialists and the veterinarian school faculty and staff. Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided?In Year 2, three PhD students, one MS student, and four undergraduate students participated in the project. Graduate students were closely co-advised by PI Kim and Co-PI Choi. ECE graduate students: Embedded system hardware and software design. Computer vision and machine learning. Complete UWB system planning and setup. Camera setup and power supply wiring. BSE graduate students: CFD modeling. Preparation of wearable and cow in-body sensor devices. Microclimate monitoring. Wearable device design, testing, and construction. Field/ barn experiment planning and setup. ECE undergraduate students: Sensor system design, deployment, and evaluation. BSE undergraduate students: Assist field experiment with dairy cows, data acquisition, and analysis. How have the results been disseminated to communities of interest?In addition to the publications reported, Hanwook Chung, a participating PhD student, gave a talk on the current research works under the title "Development of a Continuous, Low-Power, Wireless Core Body Temperature Monitoring System to Detect Heat Stress of Dairy Cows in Real Time" at 2023 American Society of Agricultural and Biological Engineers (ASABE) Annual International Meeting (AIM). What do you plan to do during the next reporting period to accomplish the goals?In Year 3, we will focus on analyzing the data collected in summer 2023 to understand how their physiological and behavioral signatures change under heat stress. The multimodal dataset newly collected in Year 2 will further improve the model. Various machine learning techniques, including computer vision and time-series prediction, will be used to understand and model the data. Specific goals include: Optimize the UWB system in terms of both anchor location and deployment, ear tag size wearable device, efficient and accurate position analysis from the raw data. Develop a ventilation control system that can integrate the multimodal dataset that we are building. Continue to develop a compact, lightweight ear tag unit. Conduct field studies with dairy cows with the developed and laboratory-tested system. Carry out computational simulations to mitigate heat stress. We will also continue investigating cattle cooling system optimization using computational fluid dynamics (CFD).
Impacts
What was accomplished under these goals?
In Year 2 of the project, we deployed and tested the sensor system developed in Year 1. In summer 2022, we deployed wearable temperature sensors for temperature monitoring at the Dairy Cattle Center (DCC) at UW-Madison to collect subcutaneous and vaginal temperatures. The design of the sensor has been published at the ACM International Conference on Mobile Computing and Networking (MobiCom), and the comparison results are accepted for publication in the Biosystems Engineering Journal. In summer 2023, we developed a new sensor system that includes core body temperature sensing, localization, direction and elevation sensing. We also installed multiple time-lapse cameras for collecting visual data. The system was deployed at the Dairy Research Farm at UW Madison. These data will be used to understand and model heat stress-related signatures in dairy cattle.
Publications
- Type:
Journal Articles
Status:
Published
Year Published:
2023
Citation:
Seunghyeon Jung, Hanwook Chung, and Christopher Y. Choi. 2023. Assessment of Airflow Patterns Induced by a Retractable Baffle to Mitigate Heat Stress in a Large-Scale Mechanically Ventilated Barn, Agriculture, 13, no. 10.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2023
Citation:
Hien Vu, Hanwook Chung, Christopher Choi, and Younghyun Kim. 2023. eTag: An Energy-Neutral Ear Tag for Real-Time Body Temperature Monitoring of Dairy Cattle, International Conference on Mobile Computing and Networking (MobiCom).
- Type:
Journal Articles
Status:
Accepted
Year Published:
2023
Citation:
Hanwook Chung, Hien Vu, Younghyun Kim, and Christopher Y. Choi. 2023. "Subcutaneous Temperature Monitoring through Ear Tag for Heat Stress Detection in Dairy Cows," Biosystems Engineering.
|
Progress 07/01/21 to 06/30/22
Outputs
Target Audience: We have reached out to U.S. dairy industry personnel who serve as supporting advisors to dairy farmers on topics relating to heat stress, heat abatement, and ventilation, including dairy consultants, nutritionists, veterinarians, extension educators, and fan/ventilation-system manufacturers and distributors. These audiences were reached through both in-person workshops and webinars delivered by Co-PI Van Os. We have interacted with a local IT entrepreneur who is interested in integrating dairy welfare with their block-chain based dairy product supply chain management. This audience was reached through both in-person and online meetings by PI Kim and Co-PI Van Os. Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided?In Year 1, three PhD students, one MS student, and four undergraduate students participated in the project. Graduate students were closely co-advised by PI Kim and Co-PI Choi. ECE graduate students:Embedded system hardware and software design. RFID and wireless power transfer circuit design.Computer vision and machine learning. BSE graduate students:CFD modeling. Wearable device design for field experiments with dairy cows.Microclimate monitoring. Embedded system software design for field experiments with dairy cows. ECE undergraduate students:Cloud computing. Embedded software design. BSE undergraduate students:Field experiment with dairy cows, data acquisition, and analysis. How have the results been disseminated to communities of interest? Co-PI Van Os delivered talks at in-person workshops and webinars. Other products are under preparation for dissemination through conference and journal papers in Year 2. What do you plan to do during the next reporting period to accomplish the goals?In Year 2, we will complete the evaluation part of Objective 1, and focus on Objective 2 based on the data collected through Objective 1. More specifically, the following research tasks will be performed. Analyze cattle behavior under heat stress: In our experiment planned for this summer, we will make a video recording of the cattle's behavior as we measure CBT using the developed sensors. The video will capture individual behavior of the cattle (such as standing/lying time and eating/drinking amount) as well as their herd behavior (such as location of and distance between the cows). We will analyze the correlation between CBT and the behavioral indications that can be used to infer heat stress without directly measuring CBT of all cows. Analyze response to subcutaneous sensor: The recorded video will also reveal how the cows respond to the subcutaneous temperature sensors and the wearable reader. Although the injection of subcutaneous devices for animal identification, the reader will be slightly heavier than normal ear tags, and its acceptance by the cows will be systematically studied by analyzing their behaviors. Design heat stress assessment models: Based on the collected data and CFD simulations, we will build microclimate and heat stress assessment models that can help the decision making of the farmers during the design and operation of barns. This will also be used for automated CFD-based cooling optimization in Objective 3.
Impacts
What was accomplished under these goals?
Measuring the cow's CBT in real time is the key to enable the proposed cyber-physical system (CPS) for heat stress management. In Year 1, we focused on the development of the implementable temperature sensor, which is Objective 1 of the project. Specifically, we have achieved the goal of designing a CBT sensor and behavioral and environmental sensors. The wearable CBT sensor is the first wearable RFID reader device with wirelessly charging capability. The details of the design is described in the "Other Products" section of this report. We are currently at the stage of validation and optimization of the sensors. We also have made progress in Objective 3 where the goal is to optimally design dairy barns based on CFD. In Year 1, we developed a CFD model of a dairy barn, which will be utilized for the optimal management of dairy barns in the later part of this project.
Publications
|