Source: NORTH CAROLINA STATE UNIV submitted to NRP
INTEGRATING ROBOTICS, AUTOMATION, AND INFORMATION TECHNOLOGY WITH DOMAIN KNOWLEDGE TO ADVANCE SENSING SYSTEMS IN BIOLOGICAL AND AGRICULTURAL APPLICATIONS
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
COMPLETE
Funding Source
Reporting Frequency
Annual
Accession No.
1021499
Grant No.
(N/A)
Cumulative Award Amt.
(N/A)
Proposal No.
(N/A)
Multistate No.
(N/A)
Project Start Date
Dec 6, 2019
Project End Date
Sep 30, 2024
Grant Year
(N/A)
Program Code
[(N/A)]- (N/A)
Recipient Organization
NORTH CAROLINA STATE UNIV
(N/A)
RALEIGH,NC 27695
Performing Department
Biological & Agr Engineering
Non Technical Summary
Today, the development of new technologies and sensors to automate the collection of data across a wide range of agricultural and biological processes is widespread. The potential for integrating robotics, automated sensing systems, and computational tools in agricultural and biological applications is increasing as costs reduce, capabilities are enhanced, and barriers are overcome. The rapid and continuous integration of these technologies and tools into the entire agricultural value chain has led to what is now known as the "digital agriculture revolution", which is currently ongoing. Though exciting and groundbreaking, these new technological advances bring new challenges. In particular, what has been lacking is a wholistic approach to developing new sensor systems that incorporates technical innovation, engineering approaches, data analytics, and human factors to develop valuable technologies that are scientifically robust yet maintain usability. Moving forward, there is a need to integrate deep technical knowledge about these new technologies with domain knowledge about the application system so their full utility can be achieved.This project will bridge the fields of robotics, automation, and sensing (various combinations of computer science, electrical engineering, and mechanical engineering) with biological and agricultural systems analysis to generate case studies that demonstrate how researchers and practitioners in this field can extract useful information and solve problems using technology. Systems to be utilized in this project include small unmanned systems (ground, aerial, underwater, and surface systems), image and spectral sensors (hyperspectral, multispectral, RGB), human-technology interaction technologies (robot interfaces), and sensors and control systems (environmental sensors). Example application systems include water resources, plant breeding and phenotyping, production agriculture, aquaculture systems, among others. This project will involve many interdisciplinary applications, primarily of those who have significant domain expertise of the application system. Through these collaborations, a community of researchers, stakeholders, students, and practitioners will become more aware of the benefits of integrating automated sensing systems into current practices and how value can be generated in the form of new data, insights, cost savings, reduced environmental impact, among many others.
Animal Health Component
50%
Research Effort Categories
Basic
50%
Applied
50%
Developmental
0%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
4047410202050%
4027410202050%
Goals / Objectives
The primary goal of this project is to advance agricultural and biological sensing systems by integrating expert-level technical knowledge of robotics and controls with domain knowledge about the application and/or research problem. To meet this goal, new sensing and robotic systems will be developed for solving well-defined biological and agricultural problems (estimated between 5-10 different problems or opportunities) in order to present a series of case studies that demonstrate the positive benefits of integrating novel technologies into current agricultural and biological applications. These case studies will include collaborators who have deep domain knowledge about the application or problem, but who may have not been successful in integrating technology into their application, thereby growing the network of researchers and practitioners who have an understanding and appreciation of these technology-driven approaches.
Project Methods
Methods for developing new systems will include applying principles of control theory (linear and nonlinear), circuit design, mechanical fabrication and testing, and software development (utilizing C++, Python, ROS, and others). In addition to system development, off-the-shelf sensors may be used to enhance or augment the system capabilities. These new systems will be tested in the field to generate data sets focused on the specific case study; these data sets may be supplemented with existing data sets from in situ sensors, surveys, previous studies, among others. This work will be carried out with collaborators (university and federal researchers, stakeholders, practitioners) who likely have not consistently utilized automated technology in their work, thereby increasing and broadening the adoption of new sensing systems. Additionally, students will be trained through formal classroom and laboratory instruction, and updated curriculum will be developed as technology advances to ensure curriculum is up to date. A smaller subset of these students will be intensively trained in the use of technology development, deployment, and evaluation through the development of MS thesis and PhD dissertation projects that integrate deep technical knowledge with domain knowledge about biological and agricultural systems. To evaluate the impact on the intended audiences, progress towards developing the outlined products (e.g., journal articles, presentations, extension materials) will be monitored.

Progress 12/06/19 to 09/30/20

Outputs
Target Audience:Target audiences reached during the reporting period include Loblolly pine tree improvement programs and breeders; industry stakeholders conducting research on cotton production; pork producers and researchers in NC; aquaculture researchers and staff at NC State; and university collaborators and researchers in the fields of data analytics, aquaculture engineering, computer science, animal science, soil science, and plant breeding. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided? Provided professional development opportunities to students (1PhD, 2MS, 1undergraduate, 1 postdoc) through research mentoring; Attended the Virtual Annual International Meeting of the American Society of Agricultural and Biological Engineers where the PI, postdoc, and graduate students presented; Had the opportunity to attend the ASCE ExCEED teaching workshop, unfortunately, this was canceled due to COVID-19; Has the opportunity to present and attendthe ASCE EWRI Congress conference where the PI and one student would have presented; unfortunately, this was also canceled due to COVID-19; How have the results been disseminated to communities of interest? One manuscript has been published as part of this work to engineering and precision agriculture communities; additional manuscripts are in review and/or in preparation for dissemination. Results have been presented at two meetings: ASABE AIM. and ASCE EWRI Congress. What do you plan to do during the next reporting period to accomplish the goals? Continue to meet with case study collaborators and stakeholders who would benefit from new precision agriculture sensing and robotics technology; Continue to mentor graduate and undergraduate students on case study projects; Create new publications for dissemination of research results to the scientific community; Continue to present in national and international conferences to dissemination results to the scientific community; Collect and analyze data, and develop new sensor and robotic systems for case studies important for agricultural and biological engineering; Continue to prepare and submit research proposals to secure funding for these efforts.

Impacts
What was accomplished under these goals? This reporting period focused on developing new collaborations, developing proposals, and collecting preliminary data. The following main goals were accomplished: AFRI Foundational proposal submitted focused on developing improved and optimized methods for aerial cropspraying (Hatch PI as PI); USDA-NSF National Robotics Initiative proposal was submitted focused on developing a fleet of aerial and surface vehicles for monitoring water quality for coastalaquaculture (Hatch PI as PI); A project was secured and pursued focused on collecting images of weeds in cotton fields for developing deep learning models for autonomous weed detection, this was the primary project for the postdoc supported by this Hatch; Hyperspectral data were analyzed of stressed loblolly pine seedlings (cold stress and disease stress) as an illustrative case study; two manuscripts were prepared and submitted focused on this work with graduate students and postdoc supported by this Hatch in collaboration with tree breeders; A new unpiloted aerial vehicle (UAV) system is being developed to improve soil moisture sensing capabilities for precision agriculture applications as another case study being pursued in collaboration with soil scientists; UAV imagery and in situ sampling data were collected for monitoring water quality with collaborators in aquaculture production; PI gave 5 Extension talksabout new technologies being developed for a variety of precision agriculture applications.

Publications

  • Type: Journal Articles Status: Awaiting Publication Year Published: 2020 Citation: P. Pandey, H. Dakshinamurthy*, and S. Youngy, Autonomy in detection, actuation, and planning for robotic weeding systems, Transactions of the ASABE (in press), 2021. doi: 10.13031/trans.14085.
  • Type: Journal Articles Status: Published Year Published: 2020 Citation: Y. Lu and S. Young, A survey of public datasets for computer vision tasks in precision agriculture,Computers and Electronics in Agriculture, vol. 178, p. 105 760, 2020. doi: 10.1016/j.compag.2020.105760.
  • Type: Journal Articles Status: Under Review Year Published: 2020 Citation: Y. Lu, K. Payn, P. Pandey*, J. Acosta, A. Heine, T. Walker, and S. Young, Hyperspectral imaging-enabled high-throughput screening of loblolly pine (Pinus taeda L.) seedlings for freeze tolerance, Biosystems Engineering (in review), 2021.
  • Type: Journal Articles Status: Under Review Year Published: 2020 Citation: S. Kronberg, F. Provenza, S. van Vliet, and S. Young, Closing nutrient cycles for animal productioncurrent and future agroecological and socio-economic issues, Animal (in review).