Progress 03/01/19 to 02/29/24

Outputs
Target Audience:Since this is a fundamental research developing sensors and robots for RUMEN understanding, target audience are the scientifc community working in this field. To increase the experiential learing opportunities and outreach activity, students and postdocs have visited Virginia Tech University working with the colloborators and collecting data in the field. In this program, we are developing a robot that can operate autonomously for monitoring the rumen health. This robot will need to move through the complex and changing environment of the rumen while taking measurements of chemicals, pH, gases, and temperature. The approach we have taken uses research about fish locomotion. By utilizing a bio-inspired robot design, we can leverage the solutions nature has developed to enable unique behaviors in the operation of the robot. The initial design of the bio-inspired fish robot was based on segmented architecture. The robot consists of multiple segments that each contain one actuator made up of two magnets and a single coil. This actuator allows each segment to oscillate +/- 20 degrees by simply changing the polarity of the current flowing through the coil. In our previous work, we have developed a magnetic, modular, undulatory robot (µBot). µBot has a compact size which is suitable for rumen monitoring. However, in the original design, µBot was tethered to external control circuit and power supply using magnetic wires. To achieve remotecommunication, we included microcontroller and customized printed circuit board (PCB) inside the robot. The goal this year was to make the robot fully autonomous. We will also add Inertial Measurement Unit (IMU) in the robot to reject heading disturbance, which lays the foundation for robot navigation. µBot was a 2D swimmer swimming beneath the water surface. We will add pectoral fins on the robot to achieve 3D swimming. As part of the overall goal of developing biosensors and gas sensors to improve productivity and overall health of ruminants to meet the growing demand for livestock production, sensors to monitor UV exposure, and greenhouse gas emissions have been developed utilizing inorganic semiconducting zinc oxide (ZnO). Previously a UV sensor was fabricated and tested as a collar device on ruminants, which could be used to monitor the exposure to UV light and prevent/mitigate effects of both primary and secondary photosensitization. The goal this year was to modify the ZnO UV photodetector to detect specific greenhouse gases at lower operating temperatures. The detection of methane and carbon dioxide emissions from ruminants can aid in the early identification and treatment of health and production issues for the animals. The detection of these greenhouse gases can also provide invaluable information on the relative emissions from different feed sources, enabling farmers to make adjustments to reduce these emissions as well. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?Four graduate students were trained in this reporting period. Training in professional research practices is being provided in the context of the research through the weekly meetings. During these weekly meeting, PI discusses: (a) the fundamental knowledge needed to support the tasks in the proposed effort, (b) the strategies for maintaining depth and breadth in the relevant literature, including active service as a peer reviewer, (c) the technical presentation skills, and (d) the skills and best practices in technical writing for publications. Students were trained on design of microrobots, sensor integration, piezoelectric actuation and sensors, and remote-control operation of robot. One female Ph.D. student is being trained on biosensor fabrication and testing. Hergoals for this project wereto establish a sensing platform that can be modified to accurately and reliably sense individual organic gases, volatile fatty acids and UV exposure. She has developed an ultraviolet (UV) photodetector with this sensing platform subsequntelyfabricating this into a wearable collar photodetector. The wearable photodetectorcould be utilized to monitor ruminants' exposure to UV lightto mitigate the effects of primary and secondary photosensitization. In addition, her methane and carbon dixode gas sensors could be utilized for complimentary sensing of ruminants' exhaled gases to aid in determination of overall health and productivity of ruminant.One MS student was trained to design remotely operated modular robot for rumen monitoring. The Ph.D. student is being trained on design of the u-bots including autonomous controls and integrated sensing and power delivery. Extensive focus is on shrinking the size of the robot while maintaining high degree of freedom to achieve advanced locomotion in tight environments. Additionally 3 capstone projects were overseen (comprising of 20 undergraduate students), and2 undergraduate students were mentored and produced honors theses, with their research focused on fish-like robotic system. How have the results been disseminated to communities of interest?Graduate students working on the program helped to organize and execute youth summer camp programs that taught students about materials science and engineering, with a focus on smart sensors and applications of materials research. The graduate students also had the opportunity to present their research at conferences to disseminate the research to a scientific audience, as well as at various other colleigate, university and international exhibitions, symposia, and meetings. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? Sensor Design, Fabrication and Testing Various morphologies of zinc oxide (ZnO) films were synthesized and studied for gaseous and aqueous sensing applications. Tetrapod ZnO (T-ZnO) was of great interest due to its unique properties from the three-dimensional geometry. T-ZnO is composed of four rod-like arms connected to a spherical core via tetrahedral angles. This geometry provides a larger surface to volume ratio in addition to forming a highly porous networked structure conducive for sensing of multiple different analytes. The T-ZnO was hypothesized to be advantageous over other microstructures, including, ZnO nanorods, microspheres and nanospheres as it possesses an increased number of active sites and a faster charge carrier path compared to other microstructures and the bulk ZnO. ZnO-based films were studied as potential candidates for ultraviolet light detection, greenhouse gas sensing, namely carbon dioxide and methane gases and for volatile fatty acid sensing in a composite structure composed of an inorganic ZnO layer and an organic polymeric layer to enhance selectivity and sensitivity in aqueous media. T-ZnO was grown via a direct flame transport synthesis method resulting in arm lengths of approximately 40-60 mm. Carbon dioxide sensors were fabricated with a direct ink writing extrusion-based 3D printing method and the sensitivity was studied as a function of surface decoration with carbon nanotubes (CNT) and gold nanoparticles (AuNPs). The addition of CNT or AuNPs was found to enhance the sensing response and the response/recovery times of the sensor until an upper loading threshold was surpassed, which resulted in the formation of a short circuit or a decrease in the sensor performance. A manuscript is in preparation based on this work. In addition, an acetate sensor is under development based on a composite T-ZnO/molecularly imprinted polymer film design. Two polymerization routes have been investigated: a bulk polymerization of acrylamide and an electropolymerization of either pyrrole and thiophene. The sensing response in controlled media is in progress focusing on a comparison between films composed of T-ZnO/molecularly imprinted polymer, T-ZnO/non-imprinted polymer, molecularly imprinted polymer, and non-imprinted polymer. The findings of this study will also be submitted as a manuscript in the near future with the goal of an integrative sensor network to be deployed on the robotic system. Robot Design, Manufacturing and Testing A fully autonomous swimming robot platform was developed. This robot, named µBot, features onboard computing, sensing, and power. Its compact size and modularity render the robot an ideal platform for swimming inside the rumen of cows as rumen ROV. The robot is equipped with a microcontroller in its head that communicates with external computers through Bluetooth Low Energy (BLE) and sends motor commands to the body segments via Inter-Integrated Circuit (I2C) protocol. Each body segment has a customized printed circuit board (PCB) that receives commands and controls the electromagnetic actuator for generating body movements. The robot head is also equipped with an Inertial Measurement Unit (IMU) to measure its heading and a battery for power. The robot actuators were activated via rhythmic motor input from a central pattern generator (CPG). Experimental results showed that the swimming speed was highly sensitive to the frequency of the motor input, with a maximum swimming speed of 130 mm/s (equivalent to 0.7 body length per second) at 6 Hz. The robot also had the capability to correct its heading with IMU feedback and follow desired paths using a line-of-sight (LOS) guidance law with an overhead camera. An experimental reinforcement learning setup was built to optimize the swimming performance of the robot platform, which could also be used for gaits optimization of rumen ROV. The reinforcement learning was performed with µBot swimming in a water tank (58 cm width × 56 cm height × 305 cm length). µBot motion was captured by an overhead monochrome camera (acA2000-165umNIR, Basler AG Inc, Ahrensburg, Germany) equipped with a 760nm filter and IR light sources. The µBot motor control system was implemented using a laptop, an Arduino Mega board, three L293B motor controllers, and a DC power supply. To track the µBot motion, reflective markers were positioned on µBot's dorsal side at every joint connecting the adjacent segments. The positions of the markers were captured by the overhead camera for swimming gaits extraction. The overhead camera was calibrated with MATLAB camera calibration toolbox (MATLAB R2020a, MathWorks, Natick, MA, USA) with an accuracy of ±1.5 mm when measuring 100 mm distance. At each time step in the experiment, the laptop relayed voltage signals to the robot via the Arduino and the motor controllers, while simultaneously receiving µBot markers' data from the camera. Serial communication for robot operation and motion capture was facilitated within MATLAB (Serial Port Interface and Image Acquisition Toolbox), running at a frequency of 50 Hz. The relationship between the swimming performance and body morphology of the robot platform was investigated. The results built the foundation for robot morphology selection for rumen ROV applications. 3.1Number of body actuators: For forward swimming, the optimized swimming speed increased as the body Degrees-of-Freedoms (µBot-DoFs) increased, while the swimming speed in Body Length per second (BL/s) decreased from 1.3 BL/s (µBot-2) to 1.0 BL/s (µBot-4), and to 0.8 BL/s (µBot-6). Removing the caudal-fin decreased forward speed from 1 BL/s (173mm/s) to 0.37 BL/s (52 mm/s). For backward swimming, µBot-2 was unable to swim backward, while µBot-4 and µBot-6 reached maximal backward speed at 0.16 BL/s (27 mm/s) and 0.25 BL/s (55 mm/s), respectively. 3.2Caudal fin stiffness: In forward swimming, the variations in caudal fin stiffness gave rise to three modes of optimized motor frequencies and swimming gaits including no caudal fin (4.6 Hz), stiffness<10-4 Pa·m4 (~10.6 Hz) and stiffness>10-4 Pa·m4 (~8.4 Hz). Swimming speed, however, varied independently with the modes of swimming gaits, and reached maximal at stiffness of 0.23×10-4 Pa·m4, with the mBot without caudal fin achieving the lowest speed. In turning maneuver, caudal fin stiffness had considerable effects on the amplitudes of both initial head steering and subsequent recoil, as well as the final heading change. It had relatively minor effect on the turning motor program except for the mBots without caudal fin. Optimized forward swimming and turning maneuver shared an identical caudal fin stiffness and similar patterns of peduncle and caudal fin motion, suggesting simplicity in the form and function relationship in mBot swimming.

Publications

• Type: Journal Articles Status: Accepted Year Published: 2023 Citation: Knoepfel, A., Poudel, B., Gupta, S. Surface-Catalyzed Zinc Oxide Nanorods and Interconnected Tetrapod Zinc Oxide as Efficient Methane Gas Sensing Platforms Chemosensors, 2023.
• Type: Journal Articles Status: Accepted Year Published: 2023 Citation: Gupta, S., Knoepfel, A., Zou, H., Ding, Y. Investigations of methane gas sensor based on biasing operation of n-ZnO nanorods/p-Si assembled diode and Pd functionalized Schottky junctions Sensors and Actuators B: Chemical, 2023, 392, 134030.
• Type: Conference Papers and Presentations Status: Other Year Published: 2023 Citation: Knoepfel, A., Poudel, B., Gupta, S. Development of ZnO-based UV and Gas Sensing Platforms to Improve Precision Livestock Farming Materials Challenges in Alternative & Renewable Energy (MCARE) (2023). Seattle, WA (Poster Presentation)
• Type: Journal Articles Status: Accepted Year Published: 2024 Citation: Hankun Deng, Donghao Li, Colin Nitroy, Andrew Wertz, Shashank Priya, and Bo Cheng, Robot motor learning shows emergence of frequency-modulated, robust swimming with an invariant Strouhal number. Journal of The Royal Society Interface
• Type: Journal Articles Status: Accepted Year Published: 2024 Citation: Hankun Deng, Donghao Li, Kundan Panta, Andrew Wertz, Shashank Priya, and Bo Cheng, Effects of caudal fin stiffness on optimized forward swimming and turning maneuver in a robotic swimmer. Bioinspiration & Biomimetics.
• Type: Conference Papers and Presentations Status: Published Year Published: 2023 Citation: Hankun Deng, Colin Nitroy, Kundan Panta, Donghao Li, Shashank Priya, and Bo Cheng, Development of an autonomous modular swimming robot with disturbance rejection and path tracking, in 2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 6645-6651.
• Type: Other Status: Accepted Year Published: 2023 Citation: Magnetic, modular, undulatory robots as robophysical models for exploration of fish-inspired swimming, in APS March Meeting 2023, American Physical Society, Las Vegas, NV. (abstract)

Progress 03/01/22 to 02/28/23

Outputs
Target Audience:Since this is a fundamental research developing sensors and robots for RUMEN understanding, target audience are the scientifc community working in this field. To increase the experiential learing opportunities and outreach activity, students and postdocs have visited Virginia Tech University working with the colloborators and collecting data in the field. In this program, we are developing a robot that can operate autonomously for monitoring the rumen health. This robot will need to move through the complex and changing environment of the rumen while taking measurements of chemicals, pH, gases, and temperature. The approach we have taken uses research about fish locomotion. By utilizing a bio-inspired robot design, we can leverage the solutions nature has developed to enable unique behaviors in the operation of the robot. The initial design of the bio-inspired fish robot was based on segmented architecture. The robot consists of multiple segments that each contain one actuator made up of two magnets and a single coil. This actuator allows each segment to oscillate +/- 20 degrees by simply changing the polarity of the current flowing through the coil. In our previous work, we have developed a magnetic, modular, undulatory robot (µBot). µBot has a compact size which is suitable for rumen monitoring. However, in the original design, µBot was tethered to external control circuit and power supply using magnetic wires. To achieve remote communication, we included microcontroller and customized printed circuit board (PCB) inside the robot. The goal this year was to make the robot fully autonomous. We will also add Inertial Measurement Unit (IMU) in the robot to reject heading disturbance, which lays the foundation for robot navigation. µBot was a 2D swimmer swimming beneath the water surface. We will add pectoral fins on the robot to achieve 3D swimming. As part of the overall goal of developing biosensors and gas sensors to improve productivity and overall health of ruminants to meet the growing demand for livestock production, sensors to monitor UV exposure, and greenhouse gas emissions have been developed utilizing inorganic semiconducting zinc oxide (ZnO). Previously a UV sensor was fabricated and tested as a collar device on ruminants, which could be used to monitor the exposure to UV light and prevent/mitigate effects of both primary and secondary photosensitization. The goal this year was to modify the ZnO UV photodetector to detect specific greenhouse gases at lower operating temperatures. The detection of methane and carbon dioxide emissions from ruminants can aid in the early identification and treatment of health and production issues for the animals. The detection of these greenhouse gases can also provide invaluable information on the relative emissions from different feed sources, enabling farmers to make adjustments to reduce these emissions as well. Changes/Problems:There was change in the PI of the program. Original PI has accepted the position at University of Minnesota but he will remain engaged as Co-PI on the program (unpaid position). He will continue to co-advise the Ph.D. student and ensure that student can graduate on time with excellent track record. The new PI on the program has been involved as a collaborator in the biosensinglaboratory so there will be no impact on the approach. What opportunities for training and professional development has the project provided?Three graduate students were trained in this reporting period. Training in professional research practices is being provided in the context of the research through the weekly meetings. During these weekly meeting, PI discusses: (a) the fundamental knowledge needed to support the tasks in the proposed effort, (b) the strategies for maintaining depth and breadth in the relevant literature, including active service as a peer reviewer, (c) the technical presentation skills, and (d) the skills and best practices in technical writing for publications. Students were trained on design of microrobots, sensor integration, piezoelectric actuation and sensors, and remote-control operation of robot. One female Ph.D. student is being trained on biosensor fabrication and testing. Here goal in this year was to establish a sensing platform that can be modified to accurately and reliably sense individual organic gases and UV exposure. She is focusing on developing an ultraviolet (UV) photodetector with this sensing platform and fabricated a wearable collar photodetector that could be utilized to monitor ruminants' exposure to UV light and to mitigate the effects of primary and secondary photosensitization. One MS student was trained to design remotely operated modular robot for rumen monitoring. The Ph.D. student is being trained on design of the u-bots including autonomous controls and integrated sensing and power delivery. Extensive focus is on shrinking the size of the robot while maintaining high degree of freedom to achieve advanced locomotion in tight environments. How have the results been disseminated to communities of interest?Graduate students working on the program have submitted manuscripts for publication and are in the process of completing few more papers (listed in the products). Also there are conference presentations to disseminate the information to the scienfic audience. What do you plan to do during the next reporting period to accomplish the goals?A fully 3D-printed T-ZnO sensor has been fabricated utilizing the direct ink writing method for both a T-ZnO thin film and carbon nanotube (CNT) interdigitated electrodes (IDE). The T-ZnO films were decorated with 0.2, 0.5, 1.0, and 2.0 wt% of gold nanoparticles (AuNP) synthesized from an AuCl4 precursor solution or 0.2 or 1.0 wt% CNT, both resuspended in deionized water. The 3D-printed T-ZnO sensors with Au or CNT surface decoration will be thoroughly evaluated as CO2 sensors. The response, sensitivity and response/recovery times will be measured for both the Au/T-ZnO and CNT/T-ZnO sensors with 3D-printed CNT interdigitated electrodes as the electrical contacts. This will be done in a custom-built isolated gas chamber in dark and UV-illuminated conditions. Aqueous biosensors for volatile fatty acids, including acetate and propionate, will also be developed which will work in tandem with the developed methane and carbon dioxide exhaled gas sensors to quantify volatile fatty acid concentrations more precisely. The next step in the design of robot is to achieve 3D swimming in a controlled fashion. Pressure sensor will be added inside the robot head to measure the distance between the robot and fluid surface (by measuring the hydrostatic pressure). Then a feedback control system will be designed to control the robot motion in vertical direction. Another challenge for rumen monitoring is to achieve remote charging of the battery. When a robot is inside the rumen of a cow, it needs to remain on standby for control command. In the future, we will design and fabricate compact remote charging module to make the robot truly autonomous and stay operated (swimming or standby) for longer time.

Impacts
What was accomplished under these goals? Sensor Design, Fabrication and Testing Two morphologies of zinc oxide (ZnO) thin films were synthesized for use as sensing layers due to the attractive properties of ZnO for sensing. ZnO is a wide bandgap (~3.37 eV) semiconductor with UV photosensitivity and high carrier mobility, as well as its Tetrapod zinc oxide (T-ZnO) and nanorod zinc oxide (ZnO-NR) thin films were fabricated and tested as methane gas sensors in a custom-built isolated gas chamber. T-ZnO possess a unique three-dimensional (3D) geometry composed of four rod-like arms connected to a core via tetrahedral angles. This unique geometry results in high porosity, a larger surface-to-volume ratio, and improved carrier mobility compared to zero-dimensional (0D) and one-dimensional (1D) morphologies. The ZnO-NR possess a 1D geometry which has previously been reported as good candidate for gas sensing due to the high number of surface atoms present compared to the bulk and the high length to diameter ratio which allows for faster charge transfer, which decreases the response time and the operating temperature. The sensing performance of the two morphologies were compared as a function of temperature and gas concentration, as well as the effect of UV illumination and palladium nanoparticle decoration as surface catalysts on the sensor's response, sensitivity and response and recovery times. Pt interdigitated electrodes (Pt IDE) were deposited via photolithography onto p-type Si substrates prior to deposition of either film for electrical contacts. The T-ZnO and ZnO-NR morphologies, crystallinity, and optical properties were characterized in addition to the sensing performance to validate the sensing mechanisms observed. ZnO-NR films, grown via a hydrothermal synthesis procedure, first were coated with a zinc acetate in ethanol seed solution containing 40 wt% Au nanoparticles prior to suspension in an equimolar zinc nitrate and hexamethylenetetramine ZnO growth solution and placed vertically in sealed containers in a furnace at 86 oC for 8 hr.The ZnO-NR had a small size distribution with the diameter ranging from 50-150 nm and the length of the nanorods ranging from 1-1.5 μm. The T-ZnO films displayed connected arms, which had a length of 40-50 μm. The T-ZnO were grown via a direct flame transport synthesis route.1 The SEM images of the Pd decorated ZnO-NR and T-ZnO films showed a uniform distribution of Pd nanoparticles (PdNPs) along the exposed surface facets. The PdNPs were synthesized via a UV reduction method from a PdCl2 precursor solution. The samples are tested for sensitivity and responsivity. Robot Design, Manufacturing and Testing Robot mechanical design: The anatomy of the swimming robot is composed of a larger head segment connected to several identical body segments and lastly, a caudal fin. Actuators are located within the head and body segments. The modular nature of the robot's design allows for a varying number of segments to be utilized for swimming. In current version, the robot consists of two body segments, thus employing three actuators. Since the robot is upgraded from µBot, we named it µBot 2.0. Head segment mechanical design The robot's head segment houses a Beetle BLE microcontroller, an IMU, a custom five wire-to-ribbon cable adapter, a battery set, and a power switch. The battery set is placed at the lowest point in the head to ensure a low center of gravity. 32 American Wire Gauge (AWG) magnetic wires connects the electrical components inside the robot head. The 5-pin wire adaptor with ribbon cable connector connects the magnetic wires with the ribbon cable which connects the PCBs in body segments. Body segments design µBot 2.0 uses electromagnetic actuators. As sent from the motor driver on the PCB, an electric signal is driven through the coil and creates an electromagnetic field, which generates a force pointing to one of the magnets. Inverting the electric signal changes the direction of the electromagnetic field and generates the force in the opposite direction. Therefore, the oscillatory motion of the coil and the coil clamp around the pivot is achieved by generating oscillatory motor input. The range of the rotation angle is plus/minus of 20 degrees. The body skeleton is designed to house the PCB and the actuator. The PCB is light (1g) and is placed at the upside of the segment, while a 5g tungsten weight is placed at the bottom side. Two holes are added at the upside to reduce the weight. This design lowers the center of gravity of the segment. The transverse plane of the segment is nearly an ellipse, while the upside is wider than the bottom side, which moves the center of geometry slightly to the upside. Therefore, the center of gravity is set below the center of geometry to achieve roll stability. The skeleton of a body segment is composed of two pieces fixed together with screws. The screw holes are placed at the bottom side. The ribbon cable path is reserved for electrical connections between the segments. The coil clamp of each segment is mechanically connected to the square mount of the next segment using easily removable screws to enable fast assembly and further increase the modularity of the design. Two O-ring grooves are reserved for O-ring placement during assembly. Caudal fin segment mechanical design The caudal fin of µBot 2.0 is inspired by pikes. It is made of 0.25 mm thick clear polyvinyl chloride sheeting. This thin caudal fin is secured to the 3D printed segment using thin wire threaded through matching holes. This design allows for quick manufacturing of caudal fins with different shapes and stiffness, and caudal fin replacement can be achieved easily. Robot assembly The robot's structural components were 3D printed in Rigid 4000 photopolymer resin with a Form 2 printer (Formlabs, MA, USA). The average density of µBot 2.0 is slightly less than water so that buoyancy equilibrium is achieved with a minor part of the body (less than 10% in depth) above the water surface, therefore achieving 2D swimming. The assembled µBot 2.0 is 19 cm in length, 2.4 cm in width, and 3.2 cm in depth. The overall weight is 85g. Pectoral fin for 3D swimming: To enable the movement in vertical direction, we are adding pectoral fins on both side of the robot. The pectoral fins help the initial pitching up/down. Then the robot can swim up/down with thrust generated by the caudal fin. Since the robot has limited space inside the head, we used a bio-inspired shape memory alloy composite (BISMAC) actuator.Two BISMAC actuators are equipped at the leading edge and trailing edge, respectively. When one of the actuators is activated, the pectoral fin tilts, generating force in vertical direction when the robot swim forward. Activating the other actuator will change the force direction. Therefore, the movement in vertical direction can be controlled.

Publications

• Type: Journal Articles Status: Published Year Published: 2022 Citation: Upinder Kaur, Rammohan Sriramdas, Xiaotian Li, Xin Ma, Arunashish Datta, Barbara Roqueto dos Reis, Shreyas Sen, Kristy Daniels, Robin White, Richard M. Voyles, Shashank Priya, Indwelling robots for ruminant health monitoring: A review of elements, Smart Agricultural Technology, Volume 3, 2023, 100109.
• Type: Journal Articles Status: Published Year Published: 2022 Citation: Chan Su Han, Upinder Kaur, Huiwen Bai, Barbara Roqueto dos Reis, Robin White, Robert A. Nawrocki, Richard M. Voyles, Min Gyu Kang, Shashank Priya, Invited review: Sensor technologies for real-time monitoring of the rumen environment, Journal of Dairy Science, Volume 105, Issue 8, 2022, Pages 6379-6404.
• Type: Journal Articles Status: Under Review Year Published: 2023 Citation: Hankun Deng, Donghao Li, Colin Nitroy, Andrew Wertz, Shashank Priya, Bo Cheng, "Motor learning reveals rhythm-controlled, robust swimming at an invariant Strouhal number", Nature Communications, under review.
• Type: Journal Articles Status: Other Year Published: 2023 Citation: 6. Hankun Deng, Donghao Li, Kundan Panta, Andrew Wertz, Shashank Priya, Bo Cheng, "Optimized forward swimming and zero-radius turning maneuver in a robotic swimmer share an identical caudal fin stiffness", Journal of the royal society interface, to be submitted
• Type: Journal Articles Status: Other Year Published: 2023 Citation: Knoepfel, S. Gupta, and B. Poudel. Investigations of Surface-catalyzed Zinc Oxide Nanorod Arrays and Interconnected Tetrapod Films for Efficient Methane Sensing. To be submitted
• Type: Journal Articles Status: Other Year Published: 2023 Citation: Knoepfel, P. Schadte, L. Siebert, R. Adelung, and S. Priya. A Surface-enhanced 3D-printed T-ZnO sensor for detection of CO2. To be submitted
• Type: Journal Articles Status: Published Year Published: 2022 Citation: Kai Wang, Sumanta Kumar Karan, Mohan Sanghadasa, Congcong Wu, Shashank Priya, Implantable photoelectronic charging (I-PEC) for medical implants, Energy Reviews, Volume 1, Issue 2, 2022, 100006.
• Type: Conference Papers and Presentations Status: Other Year Published: 2023 Citation: 1. Hankun Deng, Colin Nitroy, Kundan Panta, Donghao Li, Shashank Priya, Bo Cheng, "Development of an autonomous modular swimming robot with disturbance rejection and path following", in 2023 IROS IEEE/RSJ International Conference on Intelligent Robots (IROS), to be submitted.
• Type: Conference Papers and Presentations Status: Other Year Published: 2023 Citation: 2. Hankun Deng, Andrew Wertz, Na Liu, Kundan Panta, Shashank Priya, Bo Cheng, "Development of an autonomous 3D swimming robot with pectoral fin activated by shape memory alloy", in 2023 IROS IEEE/RSJ International Conference on Intelligent Robots (IROS), to be submitted.
• Type: Journal Articles Status: Published Year Published: 2022 Citation: Knoepfel, A., Liu, N., Hou, Y., Sujani, S., Dos Reis, B. R., White, R., Wang, K., Poudel, B., Gupta, S., & Priya, S. (2022). Development of Tetrapod Zinc Oxide-Based UV Sensor for Precision Livestock Farming and Productivity. Biosensors, 12(10), 837.

Progress 03/01/21 to 02/28/22

Outputs
Target Audience:We are developing a bio-inspired robot to operate autonomously to monitor the rumen health of a cow. This robot will need to move through the complex and changing environment of the rumen while taking measurements of volatile fatty acids, pH, gases, and temperature. The approach we have taken uses research about fish locomotion. By utilizing a bio-inspired robot design, we can leverage the solutions nature has developed to enable unique behaviors in the operation of the robot. The initial design the bio-inspired fish robot was based on segmented architecture. The robot consists of multiple segments that each contain one actuator made up of two magnets and a single coil. This actuator allows each segment to oscillate +/- 20 degrees by simply changing the polarity of the current flowing through the coil. In the previous design, all circuitry was completely external of the robot, leading to 2 wires being connected directly to each of the coils of the robot, leading to frequent wire failures. Similarly, there was no on-board processor or sensor on board to perform any tasks that would be required for the eventual autonomous operation of the robot. The goals this year were overall to redesign this base-model fish robot to help enable the eventual autonomous operation of the robot inside the rumen environment. This redesign consisted of multiple changes to increase the efficiency of assembly and operation, as well as incorporate on-board processors and a wireless communication chip for remote operation and data collection. Biosensors have been developed to monitor organic compounds found in the rumen to improve health monitoring of ruminants. Enzymatic-based biosensors exhibited problems with differentiation of target compounds in rumen fluid, due to the complex chemical composition present. To develop biosensors that are able to differentiate various targets, our research has shifted from enzymatic-based sensors to inorganic nanoparticle-based sensors, which can be used as catalysts for biochemical reactions. The goal this year was to establish a sensing platform that can be modified to accurately and reliably sense individual organic gases and volatile fatty acids found in the rumen. Piezoelectric materials such as Zinc Oxide or sodium potassium niobate can be used for designing sensors and actuators. A theoretical model was developed to understand the electromechanical behavior of piezoelectric materials as a function of applied electric field. One of the possibilities for powering the robots and sensors is to utilize solar energy. A preliminary investigation was conducted in modeling the solar and thermal energy based power sources. A novel concept was developed that combines these various forms of energies into one harvesting system. Changes/Problems:There is no major change to the project. COVID presented several challenges in the year 2021 due to reduced access to laboratory, changes in the personnel due to family emergencies, and visa delays for international postdoc and research faculty. However, we continued to perform our research tasks within the given environment. What opportunities for training and professional development has the project provided?Two young research faculty, two graduate students, one postdoctoral scholar and one undergraduate student were trained. Training in professional research practices is being provided in the context of the research through the weekly meetings. During these weekly meeting, PI discusses: (a) the fundamental knowledge needed to support the tasks in the proposed effort, (b) the strategies for maintaining depth and breadth in the relevant literature, including active service as a peer reviewer, (c) the technical presentation skills, and (d) the skills and best practices in technical writing for publications and proposals. PI also actively involves the research faculty in the discussions and decision-making with collaborative partners. Research faculty were trained on design of microrobots, sensor integration, piezoelectric actuation and sensors, and remote control operation of robot. One female Ph.D. student is being trained on biosensor fabrication and testing. Here goal this year was to establish a sensing platform that can be modified to accurately and reliably sense individual organic gases and volatile fatty acids found in the rumen. She is also focusing on developing an ultraviolet (UV) photodetector with this sensing platform and fabricated a wearable collar photodetector that could be utilized to monitor ruminants' exposure to UV light and to mitigate the effects of primary and secondary photosensitization. One undergraduate student is being trained on design, fabrication and characterization of thermal energy harvesting mechanisms. These thermal harvesters can be embedded on the collar or in a miniature form on the robot. One MS student is being trained to design remotely operated modular robot for rumen monitoring. One postdoc is assisting the project with fabrication of thermal energy harvesters that can fit the form factor required for the collar and robot integration. The PhD student and MS student have visited Virginia Tech in this past year to conduct testing on the dairy cows. This data is now being summarized for a journal publication. How have the results been disseminated to communities of interest?Results from various aspects of the project have been reported in journal publications. Graduate students have also participated in presenting their research results at the national workshops. Jungjin Yoon, Yuchen Hou, Abbey Marie Knoepfel, Dong Yang, Tao Ye, Luyao Zheng, Neela Yennawar, Mohan Sanghadasa, Shashank Priya, Kai Wang, "Bio-inspired strategies for next-generation perovskite solar mobile power sources", Chemical Society Reviews (2021). Rammohan Sriramdas, Dong Yang, Min-Gyu Kang, Mohan Sanghadasa, Shashank Priya, "Universal Multienergy Harvester Architecture", ACS Applied Materials & Interfaces 13 (1), 324-331 (2021). Xiaotian Li, Rammohan Sriramdas, Yongke Yan, Mohan Sanghadasa, Shashank Priya, "A new method for evaluation of the complex material coefficients of piezoelectric ceramics in the radial vibration modes", IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control 68, 3446 - 3460 (2021). Tao Ye, Ke Wang, Yuchen Hou, Dong Yang, Nicholas Smith, Brenden Magill, Jungjin Yoon, Rathsara RHH Mudiyanselage, Giti A Khodaparast, Kai Wang, Shashank Priya, "Ambient-Air-Stable Lead-Free CsSnI3 Solar Cells with Greater than 7.5% Efficiency", Journal of the American Chemical Society 143 (11), 4319-4328 (2021). Yuchen Hou, Congcong Wu, Xu Huang, Dong Yang, Tao Ye, Jungjin Yoon, Rammohan Sriramdas, Kai Wang, Shashank Priya, "Self?Powered Red/UV Narrowband Photodetector by Unbalanced Charge Carrier Transport Strategy", Advanced Functional Materials 31 (7), 2007016 (2021). Sumanta Kumar Karan, Rammohan Sriramdas, Min-Gyu Kang, Yongke Yan, Shashank Priya, "Small-Scale Energy Harvesting Devices for Smart Electronics", Elsevier 2021, in Encyclopedia of Materials: Technical Ceramics and Glasses (0-12-822233-6, 978-0-12-822233-1), (p. 391). What do you plan to do during the next reporting period to accomplish the goals?The next goals for the development of the robot involve further work to make the robot operate autonomously in an aquatic environment. This means testing the robot in water to learn effective swimming gaits to achieve basic navigation controls such as forward, turning, and reverse. With these basic capabilities, the robot can choose what state to perform when encountered in the field. After this, coordinating movements with an IMU sensor will help to allow localization to further enhance its autonomous swimming abilities. Another task to be pursued is changing the design slightly to enable 3-dimensional swimming by means of a pseudo-pectoral fin to change the pitch of the body as robot swims, allowing complete navigation of its entire environment. The next hurdle to overcome in the design of this robot is to incorporate a self-contained power source with wireless recharging abilities. The unique inter-connected design of the robot can enable a series of batteries to be distributed in each of the segments to help distribute the weight and space taken up. Using the head of the robot to store lithium-ion batteries is currently being explored as well. Next, additional sensors can be added onto the robot to further enhance its capabilities while operating in the rumen environment. pH sensors, and volatile fatty acid(VFA) sensors are currently under development to incorporate into the robot to allow reporting for the purpose of monitoring and maintaining the ruminant's health. Lastly, we will address the challenge with operating a robot within an environment as biologically active as the rumen is that a layer of bio-film is actively created on anything placed in the fluid. This will likely need to be prevented, possibly through the use of substances such as copper or silver nanoparticles that could help to inhibit the growth of such films as demonstrated by Lange et al.

Impacts
What was accomplished under these goals? Development of a remotely operated modular robot for rumen monitoring Body Design: The anatomy of the swimming robot is composed of a larger head segment connected to two identical modular body segments and a tail. The large size of the robot's head with respect to the other body segments is primarily needed in order to house the necessary electronics to generate locomotion, read sensors, and enable communication. One actuator is located within the head and each of the two body segments. The robot's structural components were designed using the computer aided design (CAD) software SolidWorks. These designs were then printed using a Formlabs 2 3D-printer. Formlab's Rigid 4000 photopolymer resin was used to print the robot's head, body segments, tail, and actuator armatures due to its ability to remain rigid in a wide range of temperatures. Waterproof Suit: The waterproof suit encasing the robot is made up of multiple segments of thin silicone rubber. These segments were created using a 3D-printed clam-shell mold. The thin silicone casing has a durometer rating of 00A to allow the suit to stretch while impeding the undulatory movement of the robot's segments. Silicone also has the added benefit of being bio-compatible, enabling the robot to be safely placed into the livestock's rumen. Suit Retention System: In order to enhance the modular design of the robot and make assembly and disassembly quicker and easier, a new silicone suit retention system was developed. Two silicone O-rings wrap around each of the robot's segments and rest in grooves on the exterior of the body. These O-rings hold the waterproof silicone suit in place and create a tight seal to prevent any fluids from entering into the robot's electronic control system. Flat Ribbon Cable Connections: Each of the robot's body-segments need to be connected electrically to control their respective actuation. These connections need to be made while exerting minimal flexural forces on the connecting segments and exhibit a high-cycle life to withstand the repeated bending from the segment's actuations. Previously, these connections were simply achieved using five separate 38 AWG wires that needed to be soldered between each of the PCB's. This method required substantial time to solder the 10 total pads between each circuit board and had low reliability due to the repeated bending on the thin wires. Each of these aspects were improved in the final prototype through the use of flat printed ribbon cables. These cables connect each of the PCB's by simply inserting the cable into the one of the two quick locking connectors soldered onto the circuit boards. Using these flat ribbon cables, the overall design of the robot is improved in several ways. First, the quick connectors enhance the modular design of the robot by eliminating any soldering between each of the actuator's PCB. Second, the reliability is improved by decreasing the number of separate connections spanning the gap between each of the body-sections of the robot, reducing the number of connections that could break. The lifespan of the robot is also improved by utilizing the flat ribbon cables. Overview of Robot's Communication Protocol: When deployed into the field, data will need to be exchanged with the robot remotely from within the animal. To achieve this, remote control and data acquisition are performed using the Bluetooth Low Energy (BLE) communication protocol. The microprocessor that is used to interface with BLE is a commercially available board called the DFRobot Bluno Beetle. The robot is sent commands using a BLE-capable device, such as a Raspberry Pi, and those commands are received using the Bluno Beetle. The Beetle then transmits the commands as bytes over I2C(Inter-Integrated Circuit) to the ATTiny816 microprocessor contained in each segment. Each of those bytes are used to send a pulse-width modulation (PWM) signal to the motor driver also contained in each segment. The output from the motor driver connects directly to the coil in that respective segment that then drives the motion of the robot. External Controller: The robot is configured to act as a Bluetooth peripheral device that can accept movement commands while also sending data back to the controller regarding on-board sensor data. For testing purposes in the laboratory setting, a Raspberry Pi 4 was used to communicate with the robot. Using Python's pyBluez and bluepy libraries, the robot can be commanded to execute specific movements through different swimming parameters. As the robot executes its swimming patterns, data obtained from any connected sensor will be sent serially through the BLE connection, and will be available to read and log onto the Raspberry Pi. Robot Brain: The Bluno Beetle acts as the central brain of the robot, receiving instructions from the Bluetooth controller and sending the equivalent commands to each of the robot's actuators. The chipset used in the Bluno Beetle is the ATmega328P, the same as an Arduino Uno, along with a CC2540 Bluetooth System-on-Chip device to achieve BLE operations. Being an ATmega328P also allows the Bluno Beetle to be programmed using the Arduino IDE and makes use of the readily available Arduino libraries. The Bluno Beetle's I2C bus and analog pins enable the connection of a wide range of sensors that can be connected anywhere along the communication bus in the robot's head or any of the body segments. This capability enables the robot to therefore possess as many sensors as there are body segments, space permitting. This sensor data can then read by the Buno Beetle and transmitted via Bluetooth back to the server for data acquisition. Robot Segments: Each robot segment houses an identical custom printed circuit board (PCB) and actuator. This PCB is stored above the magnet-coil actuator in each segment. The circuit board is primarily composed of an 8-bit ATTiny816 VQFN microprocessor and a Texas Instruments DRV8872 DC motor driver. The microprocessor enables communication via I2C, and transmits the actuation data from the I2C master to the motor driver in the form of two bytes that set the pulse width modulation(PWM) duty cycle signals. The ratio of these two PWM signals are used to control the direction and power level of the output sent to the coil actuator. ATTiny816 Microprocessor: The ATTiny816 is a microcontroller with an 8-bit AVR® processor. Once each PCB was fully assembled with each component soldered in place, the ATTiny816 microprocessor needed to be programmed to function as needed in the robot. This was achieved using Microchip's Atmel-ICE programmer. DRV8872 Motor Driver: Texas Instrument's DRV8872 chip is a bidirectional H-bridge motor driver that can be controlled with just two PWM signals. It can operate between the voltage ranges of 6.5V to 45V, enabling its continued use when the robot is transitioned to a 7.4V lithium-ion battery power source. Bulk capacitance was needed in parallel with the motor voltage power supply and ground to overcome voltage fluctuations and parasitic inductance. The two PWM signals are received from the microprocessor and are used to determine the output voltage sent to the coil actuators. The output's power and direction are defined by the ratio of the two signals received. Actuation via Central Pattern Generator: Different types of swimming behaviors are achieved in the robot through the implementation of central pattern generators(CPG). Central pattern generators exist in an organism's nervous system to produce rhythmic motor patterns to allow walking, breathing, flying, and swimming. The CPG utilized for this robot was based on the model proposed by Matsuoka. This algorithm describes two neurons in each actuator that work by inhibiting each other.

Publications

• Type: Book Chapters Status: Published Year Published: 2021 Citation: Sumanta Kumar Karan, Rammohan Sriramdas, Min-Gyu Kang, Yongke Yan, Shashank Priya, Small-Scale Energy Harvesting Devices for Smart Electronics, Elsevier 2021, in Encyclopedia of Materials: Technical Ceramics and Glasses (0-12-822233-6, 978-0-12-822233-1), (p. 391).
• Type: Journal Articles Status: Published Year Published: 2021 Citation: Xiaotian Li, Rammohan Sriramdas, Yongke Yan, Mohan Sanghadasa, Shashank Priya, A new method for evaluation of the complex material coefficients of piezoelectric ceramics in the radial vibration modes, IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control 68, 3446  3460 (2021).
• Type: Journal Articles Status: Published Year Published: 2021 Citation: Jungjin Yoon, Yuchen Hou, Abbey Marie Knoepfel, Dong Yang, Tao Ye, Luyao Zheng, Neela Yennawar, Mohan Sanghadasa, Shashank Priya, Kai Wang, Bio-inspired strategies for next-generation perovskite solar mobile power sources, Chemical Society Reviews (2021).
• Type: Journal Articles Status: Published Year Published: 2021 Citation: Tao Ye, Ke Wang, Yuchen Hou, Dong Yang, Nicholas Smith, Brenden Magill, Jungjin Yoon, Rathsara RHH Mudiyanselage, Giti A Khodaparast, Kai Wang, Shashank Priya, Ambient-Air-Stable Lead-Free CsSnI3 Solar Cells with Greater than 7.5% Efficiency, Journal of the American Chemical Society 143 (11), 4319-4328 (2021).
• Type: Journal Articles Status: Published Year Published: 2021 Citation: Yuchen Hou, Congcong Wu, Xu Huang, Dong Yang, Tao Ye, Jungjin Yoon, Rammohan Sriramdas, Kai Wang, Shashank Priya, Self?Powered Red/UV Narrowband Photodetector by Unbalanced Charge Carrier Transport Strategy, Advanced Functional Materials 31 (7), 2007016 (2021).
• Type: Journal Articles Status: Published Year Published: 2021 Citation: Rammohan Sriramdas, Dong Yang, Min-Gyu Kang, Mohan Sanghadasa, Shashank Priya, Universal Multienergy Harvester Architecture, ACS Applied Materials & Interfaces 13 (1), 324-331 (2021).

Progress 03/01/20 to 02/28/21

Outputs
Target Audience:Maximizing the health and productivity of ruminant animalsis an essential societal goal for a variety of reasons including food security, agricultural sustainability, and global health; however,monitoring the health of ruminant animals is currently a time-consuming and imprecise process.An indwelling robot residing within the ruminant gut would likelyrevolutionize health monitoring inruminantsby providingnew monitoring capacity withinthe complex and stratifiedfermentationoccurring.In order forindwelling robotics to make a meaningful contribution to rumen monitoring above existing bolus-based technologies, there are four critical needs. First, robotics must be mobile to allow precise navigation within the rumen, and eventually within the downstream gastrointestinal tract. Second, managers or users must have the ability to identify the location of the robot within the gut for precise sampling or diagnostic procedures.Additionally, robots will need to have reliable and low-power methods for wireless data transfer from within the body cavity to external devices.Finally, it is unrealistic for a single battery charge to power indwelling robotics indefinitely and as such, there is a need for wireless power transmission to facilitatethelong-termfunctionalityof indwelling robotics. The overarching goal in the program is to demonstrate remotely operated vehicle (ROV) that travels through the viscous rumen environment and performs the measurement of volatile fatty acid concentration and pH level inside the rumen. Three-dimensional control of ROV is being developed in order to position the ROV in a desired location within rumen. A novel concept for locomotion is being developed based on the surface travelling wave concepts. Demonstration of integrated rumen sensors on the ROV will allow real-time monitoring of the pH and VFA concentration and composition in the cow's rumen. In parallel to ROV, we are developing highly sensitive VFA sensor units utilizing electrochemical platform and miniaturized pH sensing units with three-dimensional structures for high sensitivity, outstanding linearity and low power consumption. In ruminants, the volatile fatty acid (VFA), acetate, propionate, and butyrate, constitute the major source of energy, providing at least 50 % of the total amount of digested energy. Therefore, both the yield of total VFA and relative quantity of each type of VFA formed in the rumen can significantly affect the utilization of absorbed nutrients in dairy cows and, thus, can affect milk volume and composition to a considerable extent. In addition to the VFAs, ruminal pH is a major determinant of the profile of nutrients available for absorption. It affects ruminal bacterial species and fiber degradation, which determine the feed intake, microbial metabolism, and feed digestion, and are related to inflammation, diarrhea, and milk fat depression. The second-generation earthworm inspired ROV has been developed that houses different sensors, onboard power source, and communicates wirelessly. The upgraded design also offers higher force for locomotion. An articulating joint mechanism has been implemented in the robot to achieve in plane or out of plane steering. The joint offers an articulating torque of 88 Nmm and can steer up to ±44° in yaw and 36° upward and 20° downward angles. The power consumed by servos is estimated to be 0.45 W for carrying a load of 100 g at 36° pitch angle. The force exerted by the segment on different surfaces is also measured. Peak forces of up to 8 N are exerted by the segment. The segment is observed to perform 81 mJ work by pushing a load of 200 g. Master and slave microcontrollers, infrared command communication and 433 MHz data communication links are individually verified for their integration with worm robot.The project has provided an opportunity to understand fundamental working mechanisms of the enzymatic electrochemical sensor and the extended-gate field-effect transistor (EGFET) based pH sensors. The project has provided an opportunity to realize an earthworm mimicking ROV that houses onboard power source, and wireless data communications module. Critical understanding of static and dynamic charging of electronic devices, near-field and far-field WPT technologies, powering of wearable and implantable devices, and WPT systems integration and packaging has been achieved. Changes/Problems:Due to COVID, the laboratory access has been limited but we have continued to utilize the time effectively to conduct experiments and modeling. We were able to access all the modeling softwares required for developing the robots, controls, and communication electronics. There is no major effect on our progress, except for delay in conducting joint experiments with Purdue and Virginia Tech. Once the travels start, we will visit Virginia Tech with our robotic and sensory devices to conduct testing on animals. What opportunities for training and professional development has the project provided?The project provided excellent opportunity for engineering researchers to learn about fundamental working mechanisms of the crawling and worm robots. Research training was provided on understanding and implementing bioinspired locomotion for robotic platforms encountering various terrains. The morphology and locomotion of next-generation robots will be influenced by the locomotory capabilities of small insects and other animals that live in the natural environments whose anatomy is adapted for efficiency at various desired scales. This topic provides excellent learning opportunity for students and postdoctoral fellows. FEM modeling in conjunction with analytical models (Matlab/Mathematica) is being developed to design an earthworm mimicking robot. Novel piezoelectric micromotors are being designed and fabricated including the piezoelectric material required to provide large actuation stroke. Control system design and modeling techniques are being developed and implemented. Critical understanding of static and dynamic charging of electronic devices, including near-field and far-field WPT technologies for powering of implantable devices is being developed. All these topical areas provide excellent platform for training next generation of workforce having knowledge in the areas of implantable devices, robotics, wireless sensors, wireless power transfer, and animal health monitoring. Nutrition is a key part of any ruminant production business, therefore, real-time monitoring of the pH and VFA in the rumen will provide valuable information to balance the feeding a diet with good quality and long fiber forage resulting in a healthy rumen as well as allowing ruminant to make the best use of food. Through the development of efficient WPT devices for powering bioimplants effective management of animal as well as human health is possible. These convergent research areas provide unique opportunity to educate students and postdocs about economy, nutrition, animal health, along with engineering topics. How have the results been disseminated to communities of interest?Two journal papers are in the process of being submitted: (1) ChanSuHan, UpinderKaur, HuiwenBai,BarbaraRoquetodosReis,Robin White,RobertANawrocki,Richard M.Voyles, Min Gyu Kang, and Shashank Priya, "Frontiers in Precision Dairy Technology I: Sensor technologiesforreal timemonitoringoftherumenenvironment", Journal of Dairy Science (to be submitted). (2) RammohanSriramdas, Xiaotian Li,UpinderKaur,Xin Ma,ArunashishDatta,BarbaraRoquetodos Reis,Shreyas Sen,KristyDaniels,Robin White, Richard M.Voyles, Shashank Priya, "Frontiers in Cyber-Animal Systems 2: Indwelling Robotics for Active Health Monitoring in Ruminants", Journal of Dairy Science (to be submitted). Two journal papers have been published: (1) Anthony Garcia, Gregory Krummel and Shashank Priya, Bioinspir. Biomim. 16 (2021) 026003. (2) Rammohan Sriramdas, Dong Yang, Min-Gyu Kang, Mohan Sanghadasa, and Shashank Priya, ACS Appl. Mater. Interfaces 13 (2021) 324−331. What do you plan to do during the next reporting period to accomplish the goals?The wireless data transmission from the sensors installed on the worm robot to an external base station is being developed. The functioning of individual sensors is being verified and control systems are being integrated into the robot. A fully autonomous system, including wireless power transmission to the robot, is being developed. The worm robot will feature certain performance tasks such as target seeking, obstacle avoidance, and autonomous decision making. The next generation of the robot will be improved by miniaturizing the existing design and using a combination of soft and hard surfaces for implementing other bioinspired gait mechanisms. The sensors will be further replaced with the pH and VFA sensors developed in-house. Piezoelectric actuators and motors will be implemented to reduce the energy consumption. Magnetic field based localization and energy harvesting techniques are being developed to provide power for the robot. Collaborations with Purdue University is allowing us to obtain networking and communication support. Collaborations with Virginia Tech is providing us testing platform to confirm the performance of our sensors and robots. Integration of the sensors - All the developed sensors will be integrated into a single chip. The chip will consist of enzymatic acetate and propionate sensors, nanostructured pH sensor, and a temperature sensor constructed using micro-patterned gold thin films. Rumen ROV - This second-generation worm robot inspired by earthworm is realized with an articulating joint and an onboard power source. This robot integrated with the onboard sensors can transmit the measured data wirelessly. The robot will be designed to have a wireless power transfer module to charge the onboard batteries wirelessly. The next generation of the indwelling robot will be consisting of a combination of hard and soft robotic elements that facilitate mimicking other bioinspired gait mechanisms in a more compact manner including inhouse developed VFA and pH sensors, and all the features of the existing design.

Impacts
What was accomplished under these goals? Design of the Remotely Operated Worm Robot: We finished development of a bioinspired robot that can navigate through the rumen environment. The robot can travel through the complex and viscous environment to take measurements such as volatile fatty acids (VFA), pH, temperature, etc. The conventional bolus does not have locomotion capability. When equipped with locomotion capability, the bolus can navigate through the rumen and obtain data at different locations. An indwelling robot equipped with the ability to continuously monitor different parameters will be an invaluable tool to assess animal health. The limbless locomotion is the preferred locomotion for the indwelling robot moving through the rumen. The peristaltic crawling observed in earthworms is a favorable, highly desirable, and safe form of locomotion for the ruminal robot as the limbless mechanism is harmless to the internal organs and causes less disturbance to the digesta. The large surface of contact during locomotion and small space requirements are the key advantages of the mechanism. Considering the merits of the crawling locomotion observed in earthworms, the same is adopted to locomote the robot. A proof-of-concept prototype of the mechanism is verified by building a small worm robot. However, to increase the payload capacity and enable autonomy, the volume of the worm robot is increased. The final version of the robot has an external diameter of 51 mm and a total length of 300 mm. The robot consists of three segments and one articulating joint. The individual segments are connected by a common rack and an inbuilt driving mechanism. The first and the second segments are combined through an articulating joint. The articulating joint imparts the steering capability to the robot. In addition to the in-plane turning, the joint also permits out of plane pitch motion to the robot. The worm robot consists of four onboard sensors including methane, CO2, alcohol, and total volatile organic compound (TVOC) sensors. Additionally, the head also consists of an IR receiver that communicates with an external controller or the user. The head segment consists of all the sensors and the IR receiver module. Additionally, the head segment also houses a micro controller unit (MCU) that acts as a master MCU responsible for the primary functions such as communicating with the sensors, transferring the measured data wirelessly and receive IR communication for decision making. A slave MCU is attached to the central segment that takes care of the locomotion mechanism. The individual segments are controlled by the slave MCU to realize the locomotion mechanism, while the command to initiate the locomotion is given by the master MCU. The individual segments also house a Li-polymer battery for onboard power supply to the MCUs, sensors and motors. A 433 MHz data communication module is installed in the rear segment of the module to wirelessly transmit the data from the sensors. Force Measurements: The worm robot moves forward by moving one segment at a time in a specific pattern. The use of three segments offers less drag when any one segment is moved forward, and this is the key concept in this type of locomotion. The surface of the segments is designed with a draft of 2 mm increasing towards the rear of the segment. This type of draft ensures that the resistance to the forward motion is lower than the resistance to the backward motion. This difference in the surface resistance on the segments ensures that the force on the moving segment is lower than the reaction force on the other stationary segments. In order to quantify the difference in the forces between the segments, the actual force acting on the moving segment has to be obtained. The force acting on the segment can be obtained by completely absorbing the force acting on the stationary segments. The experiments are conducted on the head segment spanning over a length of 90 mm moving on different kinds of surfaces. When the segment moves forward, the reaction force offered by the other segments is simulated by attaching the second segment to a load cell and the segment itself resting on a roller support. This support is essential to make sure that the entire reaction force is felt only by the load cell and none of the segment drafts offer any resistance. This ensures that the force measured by the load cell is the true reaction to the head segment motion. The force on the load cell emulates the result of having two segments; thus, the force recorded by the load cell represents the sum of reaction forces on the other two segments. The force acting on the segment on different surfaces is estimated by changing the roughness of the surface over which the head segment moves. The reaction force measured by the load cell can be approximated as the drag force acting on the segment as no other reaction is present in the segments except the one at the load cell. It may be noted that a peak force of 1.9 N is exerted by the segment when moving through Al plate and a force of 2.3 N is exerted when moving through wooden plank. A nominal force of 0.64 N on metal and 1.01 N on wood are recorded between the forward and reverse motion of the head segment. The forces are large when the segment is made to move on a silicone rubber of durometer 20A. The soft rubber sheet offers more resistance to the segment compared with hard surfaces. As the actual working environment is a combination of soft and hard surfaces, experiments are performed using the same setup to determine the force distribution with different loads. It was found that peak forces up to 10 N are briefly achieved during pushing 200 g mass on the silicone rubber. An average force of 3.5 N is exerted by the segment when moving 100 g and 4.5 N when pushing 200 g mass. The difference of 1 N force further corroborates the consistency in the measured data. The distance over which the segment moves is 18 mm and it is covered in 1 s time. This the energy expended in moving the segment by pushing 200 g is thus 81 mJ. These numbers are promising for the locomotion needs of the worm robot. The articulating joint enables the yaw and pitch motion to the head segment. The joint consists of two servo motors in two complementary directions to permit universal joint motion. The overall length of the joint is 60 mm, and the total weight of the joint is 43 g. The servos are designed to move in steps of 4.5° in each direction. The maximum angle for yaw is set to , while the maximum angle in pitch is limited to 36° in upward direction and 20° in the downward direction due to the mechanism limitations. In order to determine the capability of the joint, a load is applied to the one end of the joint at 45 mm away from the pitch servo and at the extreme position of 36° angle. Three different loads are considered for the experiments starting with 50 g, a 100 g and a maximum load of 200 g. The resulting torque on the servo is 22.1 Nmm, 44.1 Nmm and 88.3 Nmm. The corresponding power consumption is estimated by monitoring the current flowing through both the servo motors. The current drawn by the servo motors at 36° pitch angular displacement for the three loads is 0.11 A, 0.12 A and 0.34 A. The voltage input is set to 4V and the current is measured by noting the voltage across a 1.1 ? series resistance in the circuit. It may be noted that the power consumed by the servos for torque below 44.1 Nmm is nearly constant at 0.45 W while that at 88.3 Nmm, it is 1.4 W. The joint is capable of taking a torque load of up to 88 Nmm. However, the power drawn increases with the load above 44 Nmm.

Publications

• Type: Journal Articles Status: Published Year Published: 2021 Citation: A. Garcia, G. Krummel,and S. Priya, Bioinspir. Biomim. 16 (2021) 026003
• Type: Journal Articles Status: Published Year Published: 2021 Citation: Rammohan Sriramdas,Dong Yang, Min-Gyu Kang, Mohan Sanghadasa, and Shashank Priya, ACS Appl. Mater. Interfaces 13 (2021) 324?331
• Type: Journal Articles Status: Other Year Published: 2021 Citation: Chan Su Han, Upinder Kaur, Huiwen Bai, Barbara Roqueto dos Reis, Robin White, Robert A Nawrocki, Richard M. Voyles, Min Gyu Kang, and Shashank Priya, Frontiers in Precision Dairy Technology I: Sensor technologies for real-time monitoring of the rumen environment, Journal of Dairy Science (to be submitted).
• Type: Journal Articles Status: Other Year Published: 2021 Citation: Rammohan Sriramdas, Xiaotian Li, Upinder Kaur, Xin Ma, Arunashish Datta, Barbara Roqueto dos Reis, Shreyas Sen, Kristy Daniels, Robin White, Richard M. Voyles, Shashank Priya, Frontiers in Cyber-Animal Systems 2: Indwelling Robotics for Active Health Monitoring in Ruminants, Journal of Dairy Science (to be submitted).

Progress 03/01/19 to 02/29/20

Outputs
Target Audience:The focus in this period was on development, manufacturing, and testing of remotely operated agricultural robots capable of navigating difficult fluidic environments. Current inability to sample the rumen environment non-disruptively limits our understanding of the interplay between diet, rumen tissues, and rumen microorganisms. Poor sample collection also precludes identification of microbial species present in unique microclimates within the rumen which may be critical to our overall understanding of ruminant metabolism and efficiency of fiber fermentation. Micro-robots being developed in the program can navigate through the rumen environment while measuring and sampling both rumen tissue and rumen content. Rumen remotely operated vehicle (ROV) comprises of both electromagnetic and piezoelectric actuators to provide both large distance locomotion with fine positioning. Rumen ROV utilizes the traveling wave locomotion concept to move towards the station-keeping location in reticulum while taking advantage of the compression currents. When desired, the rumen ROV will "wake up" and navigate to the location of interest by combining magnetic localization techniques, image guidance, inertial measurement unit, and piezoelectric actuated traveling wave locomotion. Data transfer from the robot to a wearable neck collar on the cow will be achieved through a body area network. Battery power of the robot is being supplemented with magnetoelectric energy mechanism where external magnetic fields can be converted into electricity. Close collaboration with Virginia Tech and Purdue is being developed to ensure rumen ROV field testing equipped with wireless sensing and communication technologies. Changes/Problems:There are no major changes in the project. We are building new test stations and manufacturing processes to ensure high reliability in the rumen ROV. Thus in this period there was some additional time allocated towards measurement setup design and fabrication. What opportunities for training and professional development has the project provided?The project provided excellent opportunity for engineering researchers to learn about fundamental working mechanisms of the enzymatic electrochemical sensor and the extended-gate field-effect transistor (EGFET) based pH sensors. Further, it provided an opportunity to realize a flapping fin locomotion similar to that observed in pectoral fish swimming and develop manufacturing process to achieve robots with dimensions meeting the rumen ROV requirement. FEM modeling in conjunction with analytical models (Matlab/Mathematica) is being developed to design an earthworm mimicking robot. Novel piezoelectric micromotors are being designed and fabricated including the piezoelectric material required to provide large actuation stroke. Critical understanding of static and dynamic charging of electronic devices, including near-field and far-field WPT technologies for powering of implantable devices is being developed. All these topical areas provide excellent platform for training next generation of workforce having knowledge in the areas of implantable devices, robotics, wireless sensors, wireless power transfer, and animal health monitoring. How have the results been disseminated to communities of interest?We maintain close collaboration with animal agriculture scientists at Virginia Tech and Purdue University. Regular discussions are being held to incorporate all the suggested design requirements in rumen ROV and wireless sensor nodes. As we complete, collecting all the data and analysis, we plan to submit following three papers: Ram et al., "Systematic rumen monitoring in cows using autonomous sensing module for health and disease prevention", PNAS -in preparation Kang et al., "Integrated enzymatic biosensors for in-situ monitoring of volatile fatty acids in rumen", Biosensors and Bioelectronics -in preparation Han et al., "Novel three-dimensional nanostructures for implantable pH sensing layers", Advanced Functional Materials - in preparation Further, we are planning to make presentations at following conferences: IEEE/ASME international conference on advanced intelligent mechatronics, MRS Fall Meeting & Exhibit, and International Workshop on Acoustic Transduction Materials and Devices. What do you plan to do during the next reporting period to accomplish the goals?pH sensors- A novel material for pH sensing units will be explored to improve sensitivity, reduce response time, and enhance long-term stability in a rumen environment. Synthesis of the homogeneous three-dimensional nanostructures will be optimized to enhance the sensitivity and linearity of the pH sensing units. VFA sensors- The tri-enzyme acetate sensor will be optimized to achieve required sensitivity, selectivity and long-time stability. In addition to the acetate sensor, a co-enzyme sensor comprising of propionate CoA-transferase and short-chain acyl-CoA oxidase enzyme mixture will be developed for the propionate detection. Integration of the sensors- All the developed sensors will be integrated into a single chip. The chip will consist of enzymatic acetate and propionate sensors, nanostructured pH sensor, and a temperature sensor constructed using micro-patterned gold thin films. Rumen ROV - The proof of concept prototypes of the rumen ROV based on pectoral fin and earthworm locomotion will be tested in realistic operating environments. The next version of the robot would include the capability to steer in 3 dimensions and built-in intelligence to perform specific tasks. The drag forces will be reduced by using surface travelling wave concepts for locomotion limiting or eliminating the electromagnetic motors by replacing them with multilayer piezoelectric motors. Wireless power transfer - New approaches for designing compact and miniaturized WPT devices with high power density and efficiency are being pursued. Effects of size parameters such as thickness, volume, and dimensional gradient on the ME coupling and corresponding power density of the WPT device will be studied. Tests will be conducted to assess the power density and efficiency of the WPT device in different mediums such as water, oil, animal body tissue. Optimization of parameters influencing the performance of WPT device will be carried out.

Impacts
What was accomplished under these goals? In ruminants, the volatile fatty acid (VFA), acetate, propionate, and butyrate, constitute the major source of energy, providing at least 50 % of the total amount of digested energy. Therefore, both the yield of total VFA and the type of VFA formed in the rumen can significantly affect the utilization of absorbed nutrients in dairy cows and, thus, can affect milk volume and composition to a considerable extent. In addition to the VFAs, ruminal pH is a major determinant of the profile of nutrients available for absorption. It affects ruminal bacterial species and fiber degradation, which determine the feed intake, microbial metabolism, and feed digestion, and are related to inflammation, diarrhea, and milk fat depression. Thus, our approach in this period was: Demonstration of integrated rumen sensors on the rumen robot, which allow real-time monitoring of the pH and VFA concentration and composition in the cow's rumen. Development of highly sensitive VFA sensor units utilizing electrochemical platform. Development of miniaturized pH sensing units with three-dimensional structures for high sensitivity, outstanding linearity and low power consumption. The remotely operated vehicle (ROV) will have the ability to travel through the viscous rumen environment and perform the measurement of volatile fatty acid concentration and pH level inside the rumen. Three-dimensional control of ROV is being developed in order to position the ROV in a desired location within rumen. We are implementing surface travelling wave concept for locomotion through the rumen environment. In order to facilitate continuous monitoring and diagnostics of cow health, power supply is need for prolonged periods. As an alternative to powering the robot and sensors with bulky batteries having limited lifetime, wireless power transfer (WPT) technology is being developed. Although WPT systems have been demonstrated for charging of consumer mobile devices, their development and implementation for implantable devices is still in early stages. Using magnetoelectric approach, we are designing an efficient WPT system for powering rumen ROV and sensors. Major Activities Development of a tri-enzyme sensor composed of immobilizing acetate kinase (AK), pyruvate kinase (PK), and pyruvate oxidase (PyOx) for acetate the detection. Understanding electrochemical sensing mechanism of the tri-enzyme equipped working electrode. Optimization of the enzyme mixture composition to maximize enzymatic activity to generate H2O2. Development of extended-gate field-effect transistor (EGFET) based pH sensors with excellent linearity Design of a flapping fin-based locomotion and earthworm motion for the rumen ROV and implementation in a lab scale prototype. A leakproof design for overcoming any potential leak of the surrounding medium into the body of rumen ROV is developed and experimentally verified. Piezoelectric multilayer motors are developed for aiding locomotion by using travelling waves. Evaluation of different technologies for powering rumen ROV and sensors. Identification of technological barriers associated with the design of WPT systems Designing magnetoelectric composites for WPT systems

Publications