Source: Case Western Reserve University submitted to NRP
COLLABORATIVE RESEARCH: CPS: MEDIUM: GREENER PASTURES: A PASTURE SANITATION CYBER PHYSICAL SYSTEM FOR ENVIRONMENTAL ENHANCEMENT AND ANIMAL MONITORING
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
ACTIVE
Funding Source
Reporting Frequency
Annual
Accession No.
1025882
Grant No.
2021-67021-34459
Cumulative Award Amt.
$540,000.00
Proposal No.
2020-11363
Multistate No.
(N/A)
Project Start Date
Jul 1, 2021
Project End Date
Jun 30, 2025
Grant Year
2021
Program Code
[A7302]- Cyber-Physical Systems
Recipient Organization
Case Western Reserve University
10900 Euclid Avenue
CLEVELAND,OH 44106
Performing Department
Mechanical & Aerospace Engineering
Non Technical Summary
1) Responsible livestock manure management is a responsibility of livestock owners worldwide. When properly managed, livestock manure can be leveraged as a high-quality fertilizer in cropping or pasture production systems but failures in appropriate management have negative consequences on water quality through runoff into surface waters. Currently, manure management and runoff mitigation are laborious and time intensive manual tasks that cause significant burden to livestock owners.2) Case Western Reserve University (CWRU) investigators will collaborate with Virginia Tech (VT) investigators to leverage an existing cyber physical system (CPS) framework that captures and reports livestock position and behavior data in real time. The existing CPS provides information about the location of manure and livestock stress levels within a pasture necessary to produce an autonomous system for manure mitigation capable of operating effectively within populated pastures. An existing general purpose robot platform will be customized for the VT collaborators to study animal behavior and pasture properties. Robot control algorithms and intelligence will be developed by the CWRU investigators to allow for autonomous operation within the pasture that will integrate the findings of the VT investigators to minimize stress induced on livestock by the presence of the robot in the pasture. A commercially available subcompact tractor will be converted to be capable of computer-controlled, robotic operation using this software. Effectiveness of the new CPS will be tested in its ability to autonomously mitigate manure in a pasture without disrupting the normal behavior of the pasture livestock.3) The goals of this project are to collaborate with Virginia Tech to automate a small tractor to locate, collect and transport livestock manure away from streams, rivers and lakes. Thus, livestock owners will more easily be able to capture the full value of the high-quality fertilizer produced by their livestock and protect surface waters from contamination from manure runoff.
Animal Health Component
80%
Research Effort Categories
Basic
20%
Applied
80%
Developmental
(N/A)
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
4025310202080%
4030790202020%
Goals / Objectives
Work with Virginia Tech collaborators to extend an existing Cyber Physical System for monitoring grazing animal habits to develop a robot to automate manure management with the aim of reducing environmental water contamination. The project has four goals, each with objectives to pursue to realize it.1. Develop CPS algorithms and control for automated manure mitigation 1.1 Update and extend existing mobile robot software necessary for pasture sanitation robot 1 (PSR1) to measure animal behavior via teleoperation in-pasture1.2 Create robot navigation methods, procedures, and algorithms to achieve co-livestock autonomy in various elements of pasture maintenance1.3 Develop robot co-livestock control algorithms for loading/moving manure1.4 Implement algorithms for livestock stress minimizing movement plans2. Develop, implement and assess autonomous robot for automated manure mitigation2.1 Upgrade and extend the existing robot platform (PSR1) with additional sensors to collect data with respect to robot-livestock interactions to determine flight zones, time to habituation, effective habituation strategies, and management strategies to minimize livestock anxiety and destructive behaviors.2.2 Automate commercially available subcompact tractor by integrating sensors and actuators to develop an autonomous pasture sanitation pasture sanitation robot 2 (PSR2).2.3 Implement pasture locomotion and manure loading in PSR2.2.4 Assess autonomous operation of the robot for manure management in livestock occupied pastures.3. Dissemination3.1 Publish in scholarly journals and conference proceedings3.2 Publish open source code4. Education and Outreach4.1 K-12 and underrepresented communities engagement4.2 Mentor undergraduate students4.3 Advise MS theses and graduations4.4 Advise PhD dissertations and graduations.
Project Methods
To realize the stated goals, our efforts will be focused on the development of software and hardware.The results of these efforts will be evaluated to determine metrics of performance. These efforts will be directed at the previously identified objectives stated for the project's goals.Efforts:1. Develop robot algorithms and control for automated manure mitigation1.1 Update and extend existing mobile robot programming for robot (pasture sanitation robot 1, PSR1) to facilitate in-pasture measurements1.2 Create robot navigation methods, procedures, and algorithms to achieve the elements of pasture maintenance1.3 Develop robot control algorithms for loading/moving manure1.4 Implement animal stress minimizing movement plans2. Develop robot platform for automated manure mitigation2.1 Upgrade and extend the existing robot platform (PSR1) with additional sensors to collect data with respect to robot-livestock interactions to determine flight zones, time to habituation, effective habituation strategies, and management to minimize destructive behaviors2.2 Automate a commercially available subcompact tractor and integrate sensors and actuators to develop an autonomous pasture sanitation CPS (PSR2)2.3 Implement pasture locomotion and manure loading in PSR22.4 Assess autonomous operation of the robot for manure management in an occupied pasture3. Dissemination3.1 Publication in scholarly journals and conference proceedings3.2 Publication of open source code4. Education and Outreach4.1 Engagement with K-12 and underrepresented communities4.2 Mentoring undergraduate senior projects4.3 Advising MS students4.4 Advising PhD studentsEvaluation:1. Develop robot algorithms and control for automated manure mitigation1.1 Length of time of teleoperation within a pasture1.2 Conduct repeated trials for navigation between selected points within pasture measuring time/distance and RMS deviation from path plans1.3 Conduct repeated trials at various locations for autonomous loading and depositing of manure measuring time and accuracy, respectively, in simulation1.4 Monitor animal stress measurements from existing CPS animal behavior monitoring system to determine amount of induced stress2. Develop robot platform for automated manure mitigation2.1 Availability of system for data collection as percentage of time in pasture collecting data versus total time2.2 Repeatability and precision of autonomous movement2.3 Conduct repeated trials at various locations for autonomous loading, transporting and depositing of manure measuring time and accuracy, respectively, in pasture2.4 Percentage of manure produced during trial that is correctly mitigated3. Dissemination3.1 Number of publications produced and number of students contributing to publications3.2 Number of software packages published and number of downloads, external contribution back to code base4. Education and Outreach4.1 Number of classrooms visited, number of robot demonstrations, number of individuals engaged4.2 Number of undergraduate senior projects4.3 Number of MS students graduated4.4 Number of PhD students graduated

Progress 07/01/22 to 06/30/23

Outputs
Target Audience:Two senior projects for undergraduate students have been supported via this grant. An interface for the simulated environment was improved, and a means to increase GPS precision were both developed by undergraduate students in the last year. Four undergraduate students have been involved in these projects. Graduate students have been involved in the project. One master's student completed a preliminary design of a superstructure that will allow the automation elements of the Pasture Sanitation Robot 2 (PSR2) to be easily attached and detached from the tractor platform. Another master's student helped develop a mechanism for collecting the manure from the pasture without damaging the soil and vegetation in the pasture. A doctoral student has been working with Virginia Tech in the initial data collection of the PSR in a pasture with animals. The student is also facilitating the design of the PSR2 platform and has begun working with the PSR2 platform. This student is also deeply engaged in guiding both master's level graduate students, as well as being the primary advisor for the undergraduate student projects. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?One masters student and two doctoral students are being trained based on this project. There have also been several engineering senior projects for students in the Electrical, Computer, & Systems Engineering, and the Mechanical & Aerospace Engineering Departments. Undergraduate students exploring opportunities in research laboratories have also contributed to the project. How have the results been disseminated to communities of interest?Members of our research group ran a robotics workshop for K-12 students as part of the AdvantageCLE Tennis and Education camp in summer 2022. Five sessions were held with students divided by age group, ranging from 6 to 16. In each of the sessions, 3-4 graduate students spent one hour with groups of 15-20 students who were primarily African-American. For the robotics workshops, students used Lego Mindstorms kits and software. The activity focused on students building cranes to lift objects of various shapes and sizes. The students adjusted the gear ratio on the crane winch, then participated in a friendly competition to see whose crane could lift the most weight. This led the groups to focus on the impact of gear ratios on torque given the finite power of the motor and battery, as well as an interactive design process. The program moderators were pleased and said they look forward to us coming back next year. What do you plan to do during the next reporting period to accomplish the goals?The project is expected to produce specific outcomes for each objective identified for our four stated goals. 1. Develop algorithms and control for automated manure mitigation using wheeled robots 1.1 Software for supporting in-pasture operation (change in condition) Software for connecting and interpreting the existing sensor network that exists on the pasture animals and in the field will be developed. 1.2 Software to support autonomous locomotion operation (change in condition) New algorithms for in-pasture locomotion will be developed that will allow the PSR2 to safely negotiate the pasture terrain. 1.3 Algorithms to autonomously load manure (change in condition) This will be addressed in year 3. 1.4 Implement path planning algorithms that use available information about animal location and behavior (change in condition) This will be addressed in year 3. 2. Develop robot platform for automated manure mitigation 2.1 A robot platform updated to support measuring various pasture and livestock parameters (change in condition) Further studies of animal behavior and habituation will be conducted. Additional sensors will be added to the PSR1 platform to enhance data collection. 2.2 A commercially available subcompact tractor integrated with sensors, actuators, and computation to support autonomous operation (change in condition) The automation package will be installed and tested on the PSR2 platform. 2.3 The autonomous tractor will be commissioned for locomotion and manure loading using new algorithms and programming resulting in a reduction in water contamination (change in condition) This will be addressed in year 3. 2.4 The autonomous tractor will be evaluated using measurements of various metrics (time, distance, completion percentage, RMSE, livestock behavioral data, etc.) for modes of operation in-pasture (change in knowledge) This will be addressed in year 3. 3. Dissemination 3.1 Publication in scholarly journals and conference proceedings (change in knowledge) Further scholarly articles will be submitted for publication. 3.2 Publication of open source code (change in knowledge) Software developed for the platform will be publicly disseminated. 4. Education and Outreach 4.1 K-12 and underrepresented communities engagement (90 students)(change in knowledge) Several K-12 activities are planned for the second year, including Engineering Week, Lego Mindstorm sessions at the Thurgood Marshall Recreation Center, Cleveland, OH, classroom visits, etc. 4.2 Undergraduate senior projects (4)(change in knowledge) Senior engineering design projects will continue to be supported. 4.3 MS theses and graduations (2)(change in knowledge) An MS student is anticipated to graduate and other MS students will be working on various aspects of the project. 4.4 PhD dissertations and graduations (2)(change in knowledge) A PhD student is engaged and a second began in year 2.

Impacts
What was accomplished under these goals? The Greener Pastures project intends to address the need for improved pasture management by developing an autonomous robot capable of removing/redistributing livestock waste to prevent the contamination of surface waters and redistribute nutrients to improve pasture conditions. Ranchers and livestock farmers will be able to conserve valuable human work hours for tasks other than collecting animal waste in pastures. Accomplishments include: 1. Develop CPS algorithms and control for automated manure mitigation 1.1 Update and extend existing mobile robot software necessary for pasture sanitation robot 1 (PSR1) to measure animal behavior via teleoperation in-pasture An existing robot platform was adapted to collect data from a pasture. This platform was taken to the Virginia Tech livestock pastures for operations in those pastures. Data from sensors on the robot was collected, as well as data about the behavior of the pasture occupants. 1.2 Create robot navigation methods, procedures, and algorithms to achieve co-livestock autonomy in various elements of pasture maintenance Robot sensor data was collected during pasture operations. This data is being used to form the basis of path-planning algorithms adapted to the unique environment inside occupied pastures. 1.3 Develop robot co-livestock control algorithms for loading/moving manure A mechanism for the collection of manure was developed. The raking mechanism was tested in the pasture on real manure. A synthetic manure stand-in was developed for testing the mechanism outside of the pasture. 1.4 Implement algorithms for livestock stress minimizing movement plans Data was collected on animal behavior during the operation of PSR1 within occupied pastures. 2. Develop, implement and assess autonomous robot for automated manure mitigation 2.1 Upgrade and extend the existing robot platform (PSR1) with additional sensors to collect data with respect to robot-livestock interactions to determine flight zones, time to habituation, effective habituation strategies, and management strategies to minimize livestock anxiety and destructive behaviors. The PSR1 was upgraded with new sensors and has been transported to Virgina Tech on several occasions for data collection. 2.2 Automate commercially available subcompact tractor by integrating sensors and actuators to develop an autonomous pasture sanitation Pasture Sanitation Robot 2 (PSR2). The platform named in the proposal was reviewed thoroughly once the project began. The same John Deere 1025R sub-compact tractor with a front loader from the proposal was confirmed, and has been purchased and delivered The platform has had sensors and actuators added to it. A computer system has also been added to the platform. The computer is capable of received data from the sensors and of commanding movements of the actuators. Preliminary teleoperation of the platform via the onboard computer is anticipated soon. 2.3 Implement pasture locomotion and manure loading in PSR2. A mechanism to collect manure in the pasture was developed. A prototype has been tested in the pasture, and a synthetic manure alternative developed to allow for more standardized testing. The mechanism is being adapted to easily be added to the new PSR2 platform. 2.4 Assess autonomous operation of the robot for manure management in livestock occupied pastures. Assessment of autonomous operation of the PSR2, etc., will take place in year 3. 3. Dissemination 3.1 Publication in scholarly journals and conference proceedings A conference paper has been submitted for the proceedings of Living Machines in July, 2023. Additionally, posters describing the work of undergraduate students working on the senior projects based on aspects of the project have been presented in the university's undergraduate poster session. 3.2 Publish open source code Several source code packages for the robot and the physical and virtual pastures have been created. They are being hosted on GitHub, which will maximize their dissemination when released. The virtual environments for the ROS/Gazebo simulation of the robot have been publicly released. These virtual environments are geo-referenced to the actual environments where the robot operates using USGS elevation and satellite data. 4. Education and Outreach 4.1 K-12 and underrepresented communities engagement Engineering Week in the spring of 2022 at CWRU An event with students from the project used Lego Mindstorms to demonstrate gear ratios and mechanical design. This event served K-12 and underrepresented communities at Thurgood Marshall Recreation Center, Cleveland, OH. An event with students both from CWRU and Virginia Tech used PSR1, and a horse for an interactive demonstration. The event showcased robot sensing, motion control, and the robot's effects on animal behavior at the Alphin-Stuart livestock arena, Blacksburg VA for K-12 4H students from the local community. 4.2 Undergraduate mentoring Three undergraduate senior projects have been supported by this project. One has led to a poster being presented at the university's undergraduate poster session. 4.3 Advise MS theses and graduations While no MS these or graduations have been produced by this project, one master level student has a project/thesis based on the project. 4.4 Advise Ph.D. dissertations and graduations While no Ph.D. dissertations or graduations have been produced by this project, one doctoral level student is actively working on the project for their dissertation. A second PhD student joined the project in the second year.

Publications

  • Type: Conference Papers and Presentations Status: Awaiting Publication Year Published: 2023 Citation: Adams, I., Quinn, R., Lee, G., Kroeger, A., and Feuerbacher, E., Autonomous versus Manual Control of a Pasture Sanitation Robot, The 12th International Conference on Biomimetic and Biohybrid Systems (Living Machines 2023), July, 2023. (Submitted)


Progress 07/01/21 to 06/30/22

Outputs
Target Audience:Two senior projects for undergraduate students were supported via this grant. An interface for the simulated environment was improved and a means to increase GPS precision were both developed by undergraduate students during this reporting period. Three undergraduate students have been involved in these projects. Graduate students have also been involved in the project. One master's student completed a preliminary design of a superstructure that will allow the automation elements of the Pasture Sanitation Robot 2 (PSR2) to be easily attached and detached from the tractor platform. Another master's student helped develop a mechanism for collecting the manure from the pasture without damaging the soil and vegetation in the pasture. A doctoral student has been working with Virginia Tech in the initial data collection of the PSR1 robot operating in a pasture with animals. The student also began facilitating the automation of the PSR2 robot platform. This student is also deeply engaged in guiding both master's level graduate students, as well as being the primary advisor for the undergraduate student projects. Members of this project organized and assisted in STEM outreach activities for approximately 90 primarily African-American children ranging in age from 6-16. The children built and tested robots using Lego Mindstorms kits. For every 4-5 children, there was one graduate student helper. The students learned about engineering principles in a flipped classroom model and enjoyed building and testing their robotic vehicles. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?One master student and one doctoral student are being trained based on this project. There have also been several engineering senior projects for students in the Electrical, Computer, & Systems Engineering, and the Mechanical & Aerospace Engineering Departments. Undergraduate students exploring opportunities in research laboratories have also contributed to the project. How have the results been disseminated to communities of interest?The K-12 educational activities conducted at the Thurgood Marshall Recreation Center in Cleveland, OH, allowed the project to reach approximately 90 African-American children. What do you plan to do during the next reporting period to accomplish the goals?The project is expected to produce specific outcomes for each objective identified for our four stated goals. 1. Develop algorithms and control for automated manure mitigation using wheeled robots 1.1 Software for supporting in-pasture operation (change in condition) Software for connecting and interpreting the existing sensor network that exists on the pasture animals and in the field will be developed. 1.2 Software to support autonomous locomotion operation (change in condition) New algorithms for in-pasture locomotion will be developed that will allow the PSR2 to safely negotiate the pasture terrain. 1.3 Algorithms to autonomously load manure (change in condition) This will be addressed in year 3. 1.4 Implement path planning algorithms that use available information about animal location and behavior (change in condition) This will be addressed in year 3. 2. Develop robot platform for automated manure mitigation 2.1 A robot platform updated to support measuring various pasture and livestock parameters (change in condition) Further studies of animal behavior and habituation will be conducted. Additional sensors will be added to the PSR1 platform to enhance data collection. 2.2 A commercially available subcompact tractor integrated with sensors, actuators, and computation to support autonomous operation (change in condition) The automation package will be installed and tested on the PSR2 platform. 2.3 The autonomous tractor will be commissioned for locomotion and manure loading using new algorithms and programming resulting in a reduction in water contamination (change in condition) This will be addressed in year 3. 2.4 The autonomous tractor will be evaluated using measurements of various metrics (time, distance, completion percentage, RMSE, livestock behavioral data, etc.) for modes of operation in-pasture (change in knowledge) This will be addressed in year 3. 3. Dissemination 3.1 Publication in scholarly journals and conference proceedings (change in knowledge) Further scholarly articles will be submitted for publication. 3.2 Publication of open source code (change in knowledge) Software developed for the platform will be publicly disseminated. 4. Education and Outreach 4.1 K-12 and underrepresented communities engagement (30 students)(change in knowledge) Several K-12 activities are planned for the second year, including Engineering Week, Lego Mindstorm sessions at the Thurgood Marshal Recreation Center, Cleveland, OH, classroom visits, etc. 4.2 Undergraduate senior projects (4)(change in knowledge) Senior engineering design projects will continue to be supported. 4.3 MS theses and graduations (2)(change in knowledge) An MS student is anticipated to graduate and other MS students will be working on various aspects of the project. 4.4 PhD dissertations and graduations (2)(change in knowledge) A PhD student is engaged and a second should begin in year 2.

Impacts
What was accomplished under these goals? The Greener Pastures project intends to address the need for improved pasture management by developing an autonomous robot capable of removing/redistributing livestock waste to prevent the contamination of surface waters and redistribute nutrients to improve pasture conditions. Ranchers and livestock farmers will be able to conserve valuable human work hours for tasks other than collecting animal waste in pastures. Accomplishments include: 1. Develop CPS algorithms and control for automated manure mitigation 1.1 Update and extend existing mobile robot software necessary for pasture sanitation robot 1 (PSR1) to measure animal behavior via teleoperation in-pasture An existing robot platform was adapted to collect data from a pasture. This platform was taken to the Virginia Tech livestock pastures for operations in those pastures. Data from sensors on the robot was collected, as well as data about the behavior of the pasture occupants. 1.2 Create robot navigation methods, procedures, and algorithms to achieve co-livestock autonomy in various elements of pasture maintenance Robot sensor data was collected during pasture operations. This data is being used to form the basis of path-planning algorithms adapted to the unique environment inside occupied pastures. 1.3 Develop robot co-livestock control algorithms for loading/moving manure A mechanism for the collection of manure was developed. The raking mechanism was tested in the pasture on real manure. A synthetic manure stand-in was developed for testing the mechanism outside of the pasture. 1.4 Implement algorithms for livestock stress minimizing movement plans Data was collected on animal behavior during the operation of PSR1 within occupied pastures. 2. Develop, implement and assess autonomous robot for automated manure mitigation 2.1 Upgrade and extend the existing robot platform (PSR1) with additional sensors to collect data with respect to robot-livestock interactions to determine flight zones, time to habituation, effective habituation strategies, and management strategies to minimize livestock anxiety and destructive behaviors. The PSR1 was upgraded with new sensors and has been transported to Virgina Tech on several occasions for data collection. 2.2 Automate commercially available subcompact tractor by integrating sensors and actuators to develop an autonomous pasture sanitation Pasture Sanitation Robot 2 (PSR2). The platform named in the proposal was reviewed thoroughly once the project began. The same John Deere 1025R sub-compact tractor with a front loader from the proposal was confirmed, and has been purchased and delivered. The global supply-chain problems, as well as a strike at John Deere threatened to delay this by as much as a year, but the delays proved shorter than expected. 2.3 Implement pasture locomotion and manure loading in PSR2. A mechanism to collect manure in the pasture was developed. A prototype has been tested in the pasture, and a synthetic manure alternative developed to allow for more standardized testing. 2.4 Assess autonomous operation of the robot for manure management in livestock occupied pastures. Assessment of autonomous operation of the PSR2, etc., will take place in year 3. 3. Dissemination 3.1 Publication in scholarly journals and conference proceedings A conference paper has been accepted for the Living Machines Conference and proceedings in July, 2022. Additionally, posters describing the work of undergraduate students working on the senior projects based on aspects of the project have been presented in the university's undergraduate poster session. 3.2 Publish open source code Several source code packages for the robot and the physical and virtual pastures have been created. They are being hosted on GitHub, which will maximize their dissemination. They are not yet mature enough to be released publicly, but publicly releasing the code in the current structure will be trivial when it is ready. 4. Education and Outreach 4.1 K-12 and underrepresented communities engagement An event with students from the project used Lego Mindstorms to demonstrate gear ratios and mechanical design. This event served K-12 and underrepresented communities at Thurgood Marshall Recreation Center, Cleveland, OH. 4.2 Undergraduate mentoring Three undergraduate senior projects have been supported by this project. One has led to a poster (video alternative) being presented at the university's undergraduate poster session. 4.3 Advise MS theses and graduations While no MS theses or graduations have been produced by this project, one master level student has a project/thesis based on the project. 4.4 Advise Ph.D. dissertations and graduations While no Ph.D. dissertations or graduations have been produced by this project, one doctoral level student is actively working on the project for their dissertation.

Publications

  • Type: Conference Papers and Presentations Status: Accepted Year Published: 2022 Citation: Ian Adams, Roger Quinn, Greg Lee, Alexandra Kroeger, Rebecca Thompson, and Erica Feuerbacher (2022) Animal Acceptance of an Autonomous Pasture Sanitation Robot, Living Machines: Conference on Biomimetic and Biohybrid Systems, pp. 366-377, July 2022.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2022 Citation: Frank, E., Shannon, C., and Davis, R., Mobile RTK-GPS Base Station, advisor: Adams, I., Case Western Reserve Intersections Undergraduate Poster Session, May, 2022.