Progress 03/01/24 to 01/31/25

Outputs
Target Audience:Onda Vision Technologies focuses on farm owners who plant, harvest, and distribute specialty crops (e.g., fruits, vegetables, or nursery products). This niche agricultural sector holds particular significance in the Southern and Western regions of the United States, boasting a multi-billion-dollar industry. Specialty crops represents one of the most labor-intensive agricultural products because they are hand-picked, hand-cut, or hand-packed. Within the Specialty Crops Industry, farmworkers, the majority of whom are Latino, are at least 20 times more likely than other workers to suffer fatalities from heat-related illness. Our research project addresses the occupational safety needs of this vulnerable population who are at a higher risk for workplace injury and illness due to social, economic, and environmental disadvantages. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?Ongoing heat-related injuries and fatalities among farmworkers, the majority of whom are Hispanic, underscore the need to identify new technologies that anticipate and reliably prevent hazards and risks in the agricultural workplace through design (PtD). As a low-cost, marketable technology, hydration sensors have the potential to overcome social and economic safety disincentives in agriculture. Because the sensor data can be monitored by the user as well as a designated third party, such as an agricultural safety manager in a crop production setting, the technology could help validate water intake strategies and reduce morbidity and mortality, as well as economic costs, associated with heat-related illnesses (HRI). Yet understanding the impacts of agricultural workplace dynamics on prevention behaviors is key to its adoption. This study collected survey and biological data from a crew of 15 field crop workers to test the applicability of a wearable hydration sensor under agricultural conditions. How have the results been disseminated to communities of interest? Onda Vision Technologies participated in company panel discussion sponsored by the First Flight Venture Center (Research Triangle Park, NC). The company panel enabled Onda Vision to present our USDA-funded research to Ag Experient Station Directors from over fifty land grant institutions including representatives from the 1890 (HBCUs) and 1994 (Tribal Colleges) Land Grant Universities. We promoted our research efforts with hydration monitoring of specialty crop farm workers and announced opportunities for potential collaborations for our USDA SBIR Phase II proposal pursuit. Our technical manuscript is being prepared (work-in-progress)for submission to a scholarly journal covering environmental health and agricultural safety. The manuscript reports the results of biological and qualitative research conducted to test the applicability of the wearable hydration sensor under agricultural conditions. The target publications journals include: Journal of AgroMedicine and the International Journal of Environmental Research and Public Health. Onda Vision Technologies submitted a provisional patent application: U.S. Provisional Patent Application No. 63/693,440 "METHODS, DEVICES, AND SYSTEMS FOR MONITORING HYDRATION USING WEARABLE TECHNOLOGIES. Onda Vision also filed the provisional patent under the iEdison portal. What do you plan to do during the next reporting period to accomplish the goals?Onda Vision Technoloiges completed the Phase I feasibility study. The goal of this site study was to evaluate the usability of a wearable hydration sensor for on-farm safety surveillance. Investigating the impacts of agricultural workplace dynamics on HRI-prevention behaviors was a key objective. Conclusions that can be drawn from the study's findings: Research suggests ad libitum strategies did not adequately replace fluids lost during the agricultural workday. Results of urinalysis revealed study participants were hypohydrated, on average, for the duration of the fieldwork. Survey responses confirmed that few drank water hourly. Nonetheless, all were trained in heat safety and familiar with protective practices. Apparent discrepancy between USG values and perceived susceptibility to HRI implies that technological monitoring of hydration status in real time could improve detection of water deficits and recognition of heat hazards. Because the wearable hydration sensor posed no issues with agricultural work tasks, the accuracy of its readings should predict its potential to enhance agricultural workplace safety surveillance. Demonstrated the feasibility of collecting wearable bioimpedance data in a dynamic farming operation (e.g., preparing green peppers for harvest). For example, our wearable sensors did not impede or prevent the participant from performing the required arm movement and work-related tasks throughout the day. Demonstrated the ability to extract hydration-related bioimpedance parametersfrom noisy wearable sensor data. Phase II development look-ahead: Wearable Data Analysis Improve accuracy of hydration-related bioimpedance parameters by correcting the raw measurements against sweat and other physiological effects. For example, during work in hot/humid conditions, sweating leads to increased hydration of the skin and underlying tissues. The elevated skin moisture decreases the resistance across the outer layers, further lowering the total resistive component.Subtracting the impact of sweat on bioimpedance measurements involves isolating its influence on the readings and accounting for the impact on the bioimpedance measurements. Onda Vision developed deep learning neural network models to estimate the sweat-induced contribution and subtract it from the total bioimpedance measurement. To develop the initial models, we curated the data collected from Day 1 and Day 2 to construct training, validation, and tesing datasets. We designeda single deep learning model for each person, but trained the model using the person-specific data sourcedue to the unique physical characteristics that manifest the bioimpedance measusurements. Initial results indicate a 10-20% compensation in sweat-induced impedance contribution. Smart Safefy IoT Infrastructure Our initial assumptions related to installing our Smart Safety hardward (e.g., IoT Gateway) into existing farm infrastructure presents several challenges. Farm vehicles and trucks are not always located in a fixed location, can potentially leave the farm area to deliver crops during field operation, or are located several meters (> 75 m) from the targeted farmworkes. Our field experiments show this causes gaps and spuriousdata collections. In addition, these fixed locations prevents covering large acres (> 40)of crops. Thus, making continuous measurements difficult. Second, and more importantly, farmers do want foreign infrastructure placed throughouttheir crop fields. We plan to revise our Smart Safety IoT system using an aerial (drone) based approach. This enable large acre (> 40) coverage, AI-powered route planning, and continuous data collection. We received inital feedback on this approach during our field data collection experiments(Corbett Brothers Farms). We are in the inital planing stages to select an appropriate drone platform with the necessary payload capacity and flight endurance.

Impacts
What was accomplished under these goals? The goal of this site study was to evaluate the feasibility of a farm-level safety surveillance system that continuously monitors the hydration status of farmworkers. The goal was met by assessing the usability of a wearable hydration sensor in an agricultural setting and gauging farmworker reception to the technology. Major Activitesaccomplished under these objectives: In cooperation labor contractor (Trejo Management) and specialty crop grower (Corbett Brothers Farms), a small on-farm sensor network was set up near the state line between Florida and Valdosta, Georgia to evaluate the performance of the wearable hydration sensors.Trejo Management agreed to provide Onda Vision Technologies with access to their worker facilities and farming operations during the experiments. Their harvesting crew started the day at 7 am and ended the day at 7 pm. Provided a project introduction, obtained informed consent, and collected baseline survey and biological data from a crew of 15 agricultural workers during the study's orientation session in preparation for field research. Using audience response technology, the crew completed a 30-item questionnaire that assessed routine work experiences, fluid-intake practices, familiarity with heat safety, and demographics. The crew also provided a selfcollected urine sample and weight and height measurements. Baseline urine specific gravity, body mass index, and perceived workday thirst levels were established. Collected pre- and post-workday urine samples and weight measurements from the crew during three days of field research. Urine samples were analyzed via refractometry to assess hydration status at the start and end of each workday. Post-workday survey research data were also collected over the three-day field research period using audience response technology. The crew completed a 20-item questionnaire each evening that assessed self-reported beverage intake, exertion levels, cooling practices, perceived thirst, and response to the wearable technology. Conducted an exit interview on the last day of field research. Study participants were provided supplemental heat safety materials. Final feedback regarding the wearable hydration sensor technology was obtained.we completed the day by collecting end of day urine samples, bioimpedance measurements, and body mass measurements. Data Collection and Analysis Baseline data were collected during the orientation session held on the first day of the study. At the end of the session, study participants were fitted with an armband containing the hydration sensor and further instructed on its use during work hours. During the remaining three days of the study, biological data were collected at the housing complex before and at the end of the workday. Each morning, workers dropped off self-collected urine samples and were weighed as they departed for work. The same procedure was used upon their return. Care was taken to ensure the same clothing was worn during both weigh-ins. Once all biological data were collected each evening, the post-workday survey was collectively administered using audience response technology. All biological data collection and urinalysis procedures were performed by a single researcher for consistency. Wearable Data Collection Wearablesensors were outfitted for each person's upper bicep (left arm) to collect local bioimpedance measurements. The workers wore the devices and provided pre and post workday measurements during the study period. The intensity of their harvest activities allowed the project to evaluate the usability of the technology under harsh conditions. The wearable sensors were configured for bioimpedance spectroscopy measurements with 10 logarimthic spaced frequecnies in the range5k - 150k Hz. A single measurement takes less than a half-of-second (< 0.5 sec). For each pre and post measurement, we collected on average 732 samples/person. These measurements served as baseline and pos-work measurements. Collection Metrics: Day 1 - 64% (9/14) of the wearable devices deployed yielded valid measurement data. Day 2 - 78% (11/14) of the wearable devices deployed yielded valid measurement data. Successfully scanned and discovered all wearable devices during the field observations. Each wearable device contained a unqiue ID to prevent unwanted/unknown devices connection to our system. Field data was collected over two days that covered a combined 60+ arces across four green pepper fields. Prior to performing analysis, we pre-processed the datasets using data-driven signal processing techniques to remove noise artifacts and motion artifacts. Noise artifacts and motion artifacts observed from the data includes: larges spikes (1k - 100k ohms) resulting from armband adjustments (open circuits), movement during pre- and post-measurements, sensor movement against the subject's skin due to incorrect armband size and bad electrode connections with the electronics. Summary Statistics and Discussion of Results During the three-day period of field observation following orientation, study participants continued to work on preharvest operations. Post-workday survey responses indicated that the highest levels of work intensity, heat, thirst, and unwellness were reported on Day 2. It was the only day during the study period that any respondents characterized their workday as "very hot" or felt "unwell." While 64% ranked work tasks performed on Day 2 as "moderately strenuous," 29% felt they were "very strenuous." Results of urinalysis indicate study participants became increasingly hypohydrated as both the workday and workweek progressed. For example, on Day 1, the average USG value increased from 1.022 at the start of the workday to 1.031 by the end of it. Day 1 pre- and post-workday weight measurements similarly showed a 1.4% percentage loss, on average (Table 2). Some participants lost close to 3% of their body weight over the course of the Day 1 work shift. On Days 2 and 3, average pre-workday USG values were each higher than the previous day's value. Overall, results of urinalysis suggest study participants were hypohydrated for the duration of the study and severely hypohydrated from the end of Day 1. Hydration related parameters collected from each subject datasets include: Z (magnitude), Pha (phase angle), R (resistivity), Xc (reactance), Zh (normalized magnitude), Rh (normalized resistivity), and Zhigh/Zlow (impedance ratio). The normalized versions are derived by dividing the magnitude (resistivity) by the distance between the sensing electrodes (d = 2.4 cm). The normalized versions enable inter-subject comparision of bioimpedance measurements. All the bioimepdance parameters were taken for f = 20k (low-frequency) and f = 150k (high-frequency).

Publications

Progress 03/01/23 to 01/31/25

Outputs
Target Audience:Onda Vision Technologies focuses on farm owners who plant, harvest, and distribute specialty crops (e.g., fruits, vegetables, or nursery products). This niche agricultural sector holds particular significance in the Southern and Western regions of the United States, boasting a multi-billion-dollar industry. Specialty crops represents one of the most labor-intensive agricultural products because they are hand-picked, hand-cut, or hand-packed. Within the Specialty Crops Industry, farmworkers, the majority of whom are Latino, are at least 20 times more likely than other workers to suffer fatalities from heat-related illness. Our research project addresses the occupational safety needs of this vulnerable population who are at a higher risk for workplace injury and illness due to social, economic, and environmental disadvantages. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?Ongoing heat-related injuries and fatalities among farmworkers, the majority of whom are Hispanic, underscore the need to identify new technologies that anticipate and reliably prevent hazards and risks in the agricultural workplace through design (PtD). As a low-cost, marketable technology, hydration sensors have the potential to overcome social and economic safety disincentives in agriculture. Because the sensor data can be monitored by the user as well as a designated third party, such as an agricultural safety manager in a crop production setting, the technology could help validate water intake strategies and reduce morbidity and mortality, as well as economic costs, associated with heat-related illnesses (HRI). Yet understanding the impacts of agricultural workplace dynamics on prevention behaviors is key to its adoption. This study collected survey and biological data from a crew of 15 field crop workers to test the applicability of a wearable hydration sensor under agricultural conditions. How have the results been disseminated to communities of interest? Onda Vision Technologies participated in company panel discussion sponsored by the First Flight Venture Center (Research Triangle Park, NC). The company panel enabled Onda Vision to present our USDA-funded research to Ag Experient Station Directors from over fifty land grant institutions including representatives from the 1890 (HBCUs) and 1994 (Tribal Colleges) Land Grant Universities. We promoted our research efforts with hydration monitoring of specialty crop farm workers and announced opportunities for potential collaborations for our USDA SBIR Phase II proposal pursuit. Our technical manuscript is being prepared (work-in-progress)for submission to a scholarly journal covering environmental health and agricultural safety. The manuscript reports the results of biological and qualitative research conducted to test the applicability of the wearable hydration sensor under agricultural conditions. The target publications journals include: Journal of AgroMedicine and the International Journal of Environmental Research and Public Health. Onda Vision Technologies submitted a provisional patent application: U.S. Provisional Patent Application No. 63/693,440 "METHODS, DEVICES, AND SYSTEMS FOR MONITORING HYDRATION USING WEARABLE TECHNOLOGIES. Onda Vision also filed the provisional patent under the iEdison portal. What do you plan to do during the next reporting period to accomplish the goals?Onda Vision Technoloiges completed the Phase I feasibility study. The goal of this site study was to evaluate the usability of a wearable hydration sensor for on-farm safety surveillance. Investigating the impacts of agricultural workplace dynamics on HRI-prevention behaviors was a key objective. Conclusions that can be drawn from the study's findings: Research suggests ad libitum strategies did not adequately replace fluids lost during the agricultural workday. Results of urinalysis revealed study participants were hypohydrated, on average, for the duration of the fieldwork. Survey responses confirmed that few drank water hourly. Nonetheless, all were trained in heat safety and familiar with protective practices. Apparent discrepancy between USG values and perceived susceptibility to HRI implies that technological monitoring of hydration status in real time could improve detection of water deficits and recognition of heat hazards. Because the wearable hydration sensor posed no issues with agricultural work tasks, the accuracy of its readings should predict its potential to enhance agricultural workplace safety surveillance. Demonstrated the feasibility of collecting wearable bioimpedance data in a dynamic farming operation (e.g., preparing green peppers for harvest). For example, our wearable sensors did not impede or prevent the participant from performing the required arm movement and work-related tasks throughout the day. Demonstrated the ability to extract hydration-related bioimpedance parametersfrom noisy wearable sensor data. Phase II development look-ahead: Wearable Data Analysis Improve accuracy of hydration-related bioimpedance parameters by correcting the raw measurements against sweat and other physiological effects. For example, during work in hot/humid conditions, sweating leads to increased hydration of the skin and underlying tissues. The elevated skin moisture decreases the resistance across the outer layers, further lowering the total resistive component.Subtracting the impact of sweat on bioimpedance measurements involves isolating its influence on the readings and accounting for the impact on the bioimpedance measurements. Onda Vision developed deep learning neural network models to estimate the sweat-induced contribution and subtract it from the total bioimpedance measurement. To develop the initial models, we curated the data collected from Day 1 and Day 2 to construct training, validation, and tesing datasets. We designeda single deep learning model for each person, but trained the model using the person-specific data sourcedue to the unique physical characteristics that manifest the bioimpedance measusurements. Initial results indicate a 10-20% compensation in sweat-induced impedance contribution. Smart Safefy IoT Infrastructure Our initial assumptions related to installing our Smart Safety hardward (e.g., IoT Gateway) into existing farm infrastructure presents several challenges. Farm vehicles and trucks are not always located in a fixed location, can potentially leave the farm area to deliver crops during field operation, or are located several meters (> 75 m) from the targeted farmworkes. Our field experiments show this causes gaps and spuriousdata collections. In addition, these fixed locations prevents covering large acres (> 40)of crops. Thus, making continuous measurements difficult. Second, and more importantly, farmers do want foreign infrastructure placed throughouttheir crop fields. We plan to revise our Smart Safety IoT system using an aerial (drone) based approach. This enable large acre (> 40) coverage, AI-powered route planning, and continuous data collection. We received inital feedback on this approach during our field data collection experiments(Corbett Brothers Farms). We are in the inital planing stages to select an appropriate drone platform with the necessary payload capacity and flight endurance.

Impacts
What was accomplished under these goals? The goal of this site study was to evaluate the feasibility of a farm-level safety surveillance system that continuously monitors the hydration status of farmworkers. The goal was met by assessing the usability of a wearable hydration sensor in an agricultural setting and gauging farmworker reception to the technology. Major Activitesaccomplished under these objectives: In cooperation labor contractor (Trejo Management) and specialty crop grower (Corbett Brothers Farms), a small on-farm sensor network was set up near the state line between Florida and Valdosta, Georgia to evaluate the performance of the wearable hydration sensors.Trejo Management agreed to provide Onda Vision Technologies with access to their worker facilities and farming operations during the experiments. Their harvesting crew started the day at 7 am and ended the day at 7 pm. Provided a project introduction, obtained informed consent, and collected baseline survey and biological data from a crew of 15 agricultural workers during the study's orientation session in preparation for field research. Using audience response technology, the crew completed a 30-item questionnaire that assessed routine work experiences, fluid-intake practices, familiarity with heat safety, and demographics. The crew also provided a selfcollected urine sample and weight and height measurements. Baseline urine specific gravity, body mass index, and perceived workday thirst levels were established. Collected pre- and post-workday urine samples and weight measurements from the crew during three days of field research. Urine samples were analyzed via refractometry to assess hydration status at the start and end of each workday. Post-workday survey research data were also collected over the three-day field research period using audience response technology. The crew completed a 20-item questionnaire each evening that assessed self-reported beverage intake, exertion levels, cooling practices, perceived thirst, and response to the wearable technology. Conducted an exit interview on the last day of field research. Study participants were provided supplemental heat safety materials. Final feedback regarding the wearable hydration sensor technology was obtained.we completed the day by collecting end of day urine samples, bioimpedance measurements, and body mass measurements. Data Collection and Analysis Baseline data were collected during the orientation session held on the first day of the study. At the end of the session, study participants were fitted with an armband containing the hydration sensor and further instructed on its use during work hours. During the remaining three days of the study, biological data were collected at the housing complex before and at the end of the workday. Each morning, workers dropped off self-collected urine samples and were weighed as they departed for work. The same procedure was used upon their return. Care was taken to ensure the same clothing was worn during both weigh-ins. Once all biological data were collected each evening, the post-workday survey was collectively administered using audience response technology. All biological data collection and urinalysis procedures were performed by a single researcher for consistency. Wearable Data Collection Wearablesensors were outfitted for each person's upper bicep (left arm) to collect local bioimpedance measurements. The workers wore the devices and provided pre and post workday measurements during the study period. The intensity of their harvest activities allowed the project to evaluate the usability of the technology under harsh conditions. The wearable sensors were configured for bioimpedance spectroscopy measurements with 10 logarimthic spaced frequecnies in the range5k - 150k Hz. A single measurement takes less than a half-of-second (< 0.5 sec). For each pre and post measurement, we collected on average 732 samples/person. These measurements served as baseline and pos-work measurements. Collection Metrics: Day 1 - 64% (9/14) of the wearable devices deployed yielded valid measurement data. Day 2 - 78% (11/14) of the wearable devices deployed yielded valid measurement data. Successfully scanned and discovered all wearable devices during the field observations. Each wearable device contained a unqiue ID to prevent unwanted/unknown devices connection to our system. Field data was collected over two days that covered a combined 60+ arces across four green pepper fields. Prior to performing analysis, we pre-processed the datasets using data-driven signal processing techniques to remove noise artifacts and motion artifacts. Noise artifacts and motion artifacts observed from the data includes: larges spikes (1k - 100k ohms) resulting from armband adjustments (open circuits), movement during pre- and post-measurements, sensor movement against the subject's skin due to incorrect armband size and bad electrode connections with the electronics. Summary Statistics and Discussion of Results During the three-day period of field observation following orientation, study participants continued to work on preharvest operations. Post-workday survey responses indicated that the highest levels of work intensity, heat, thirst, and unwellness were reported on Day 2. It was the only day during the study period that any respondents characterized their workday as "very hot" or felt "unwell." While 64% ranked work tasks performed on Day 2 as "moderately strenuous," 29% felt they were "very strenuous." Results of urinalysis indicate study participants became increasingly hypohydrated as both the workday and workweek progressed. For example, on Day 1, the average USG value increased from 1.022 at the start of the workday to 1.031 by the end of it. Day 1 pre- and post-workday weight measurements similarly showed a 1.4% percentage loss, on average (Table 2). Some participants lost close to 3% of their body weight over the course of the Day 1 work shift. On Days 2 and 3, average pre-workday USG values were each higher than the previous day's value. Overall, results of urinalysis suggest study participants were hypohydrated for the duration of the study and severely hypohydrated from the end of Day 1. Hydration related parameters collected from each subject datasets include: Z (magnitude), Pha (phase angle), R (resistivity), Xc (reactance), Zh (normalized magnitude), Rh (normalized resistivity), and Zhigh/Zlow (impedance ratio). The normalized versions are derived by dividing the magnitude (resistivity) by the distance between the sensing electrodes (d = 2.4 cm). The normalized versions enable inter-subject comparision of bioimpedance measurements. All the bioimepdance parameters were taken for f = 20k (low-frequency) and f = 150k (high-frequency).

Publications

Progress 03/01/23 to 02/29/24

Outputs
Target Audience:Onda Vision Technologies focuses on farm owners who plant, harvest, and distribute specialty crops (e.g., fruits, vegetables, or nursery products). This niche agricultural sector holds particular significance in the Southern and Westernregions of the United States, boasting a multi-billion-dollar industry. Specialty crops represents one of the most labor-intensive agricultural products because they are hand-picked, hand-cut, or hand-packed.Within the Specialty Crops Industry, farmworkers, the majority of whom are Latino, are at least 20 times more likely than other workers to suffer fatalitiesfrom heat-related illness. Our research project addressesthe occupational safety needs of thisvulnerable population who are at a higher risk for workplace injury and illness due to social, economic, and environmental disadvantages. Changes/Problems:Technical Challenges Onda Vision Technologies has encountered several technical issues that caused delays to our wearable prototypes, completing the implementation of our cloud-based platform, and conducting our farm field experiments. Onda Vision Technologiestechnical challenges centered around the following: Wearable Prototypes: Onda Vision contracted a third-party contract manufacturer (Protronics) to produce twenty (20) new wearable device boards. Protronics has only produced 65% functional wearable devices (13). The technical issues with the remaining 35% of devices (7) include failure with the power buttons, Bluetooth connections, and unable to flash the embedded code. Impact: Insufficient number of wearable devices to perform the farm experiments. IoT Gateway: The manufacturer (Laird Connectivity) discontinued our initially selected IoT Gateway device (Laird IG60-BL654) and no longer supports the latest version of the AWS Greengrass software (v2.0). Onda Vision purchased the updated Laird IG60-BL654 (Laird Linux) gateway device. However, the newer IoT device requires writing additional software to perform the Bluetooth connection, data collection, and aggregation. These unexpected software tasks added extra time and testing to our development schedule. Impact: Prevents the aggregation and collection of bioimpedance measurements from the wearable sensors worn by each farmworker. Farm Experiments: Onda Vision Technologies delayed the farm experiments due to the technical issues encountered. Impact: Postpones the farm experiments to the 2024 Spring harvest season. Project Management Change Onda Vision Technologies experienced an unexpected circumstance with the University of Florida subaward. Dr. Maria Morera agreeded to conduct the research as an independent consultant to Onda Vision Technologies. Dr. Morera has maintained her program involvement and support with the project team.Dr. Maria Morera will perform the same research tasks as an independent contractor. The subaward change did not pose a risk to the performance or execution of the research task in any of the following: personnel, facilities, resources, research objectives, instrumentation, or level of expenditures.The change request didnot require any additional funds from the USDA. Since Onda Vision Technologies did not enter into a contractual agreement with the University of Florida, this approach requiredminimal changes to our technical plan and budgetary allocation. What opportunities for training and professional development has the project provided? Onda Vision Technologies participated in the AgLaunc365 accelerator program (13 - 17 November 2023)held in Memphis, TN. AgLaunch365 is a specialized program that supports farmer-selected agtech start-ups with business development training, potential equity investment and customer acquisition opportunities, and sophisticated on-farm trials. Onda Vision submitted an Executive Summary along with a Video Pitch to gain acceptance to the program. Onda Vision refined and presented our pitch during the challenge week to farmers and potential investors (AgLaunch Farmers LLC and Ag Ventures).Leveraging our current USDA SBIR Phase I feasibility study and farm experiments, we plan to utilize the connections and experiences from theAgLaunch365 programto boost and strengthen our SBIR Phase II proposal. Onda Vision Technologies is currently participating in the AgTech Propellor Cohort business program sponsored by First Flight Venture Center, Durham, North Carolina. The AgTech program helps entrepreneurs identify potential markets, develop a path to market strategy, and learn how to communicate the value of the idea to early stakeholders and adopters. Onda Vision received business pitch training during our participation in the AgLaunch Bootcamp. We received feedback from farmers, Ag consultants, and investors to help refine our value proposition to the farming community. Onda Vision attended the AgBio2023 conference sponsored by VirginiaBIo. The AgBio2023 conference was held in Danville, VA on July 19, 2023. The benefit from attending the conference allowedOnda Vision to present our research to farmers, potential industry partners, and public heath organizations. Onda Vision completed the 4-day AgLaunch Bootcamp sponsored by AgLaunch. AgLaunch is a non-profit organization utilizing a farm-centric innovation platform to efficiently bring new ideas into agriculture (https://aglaunch.com).Onda Vision presented our research and business pursuits to the farming community and investors within the agriculture industry. Onda Vision received feedback onour technical and business plans as well as access to potential beta testers and customers. The outcome of the program resulted in Onda Vision presenting a seven minute pitch to participatingfarmers and investors. Onda Vision sponsored a summer intern through the North Carolina Small Business TechnologyDevelopment Center (SBTDC) Summer Intern Program. Onda Vision's summer intern wasa full-time MBA student at the University of North Carolina Kenan-Flagler Business School. Theinternship focuses onthe market research and customer discovery to support Onda Vision's Commercialization Plan for the SBIR Phase II proposal. Onda Vision completed training from the Joint School of Nanoscience and Nanoengineering located on the campus of North Carolina A&T University. As an industry member of the JSNN, Onda Vision has access to start-of-the-art labortory resources and facilities to help fabricate and produce the core silver nanowire electrode sensors. Onda Vision also has access to JSNN students, faculty, and research scientists for future collaborations. Onda Vision received training the Scanning Electron Microscope (SEM) to help characterize the silver nanowire sensor electrodes used in their wearable hydration sensor. The training increases Onda Vision's technology capability and knowledge in fabricating and producing the core sensors used during the research project. How have the results been disseminated to communities of interest? Nothing Reported What do you plan to do during the next reporting period to accomplish the goals? Complete testing and evaluation of the fully assembled wearable devices. These devices will be employed during the data collection activities. Completethe farm experiments and data collections. The data will lead to determiningthe feasibility and usability of our wearable system within the context of a dynamic farming operation Analyzethe data collected wearable measurements todemonstrate the ability to extract useful information from noisy sensor data. Summarize our data and results in a scientific research article.

Impacts
What was accomplished under these goals? Farm Site Visit Onda Vision completed a site visit and early assessment of the farming operations targeted for the feasibility study. The site visit was with Trejo Management located in Valdosta, Georgia. Trejo Management is a farm labor contractor with operations located in Lake Park, Georgia. Onda Vision discussed the goals and objectives of the research and feasibility study. Trejo Management agreed to provide Onda Vision with access to volunteers for the feasibility study along with access to the farming operations (e.g., harvesting peppers) between early Spring / late Summer. Onda Vision observed cabbage and squash harversting operations. Oursite visit revealedpotential technical challenges along potential solutions for integrating Onda Vision's wearable technology into real-world environments.Crew member sizecan vary between 10 - 18 persons. Both harvesting crews operate within close proximity of each other and the tractor. Understanding and exploiting the current harvesting workflows provides critical insight into how Onda Vision can seamlessly integrate our Smart Safety platform into the daily farming workflow. Specialty crop harvesting methods and crew formations will drive the following system components: Infrastructure: determines the optimal approach for data collection. Distance: determines the required spatial coverage. Crew Motion: determines the noise level with the wearable data streams (e.g., walking, running, reaching). Time: determine the battery life to last during a full-days work schedule. Smart Safety Platform (IoT) Development Onda Vision's Smart Safety platform usesAmazon Web Services (AWS) IoT software components. The wearable sensors represent the sensors worn by each farmworker and the IoT Gateway devicedenotes the data aggregation and collection. Smart Safety sends the sensor data to the AWS Cloud via the Internet Gateway (e.g., Onda Vision's AT&T 5G Hotspot device). Smart Safety can scale to accommodate different crew sizes. The function and location of the IoT Gateway device serves a key role in the Smart Safety infrastructure as stated above. Our AWS Cloud infrastructure consists of AWS IoT Core, AWS IoT Rule, AWS Kinesis, and AWS S3 buckets for data storage. AWS IoT Core handles wearable device management and connectivity to the cloud. AWS IoT Rule Engine and AWS Kinesis Data Stream handle the input data from the wearable sensors. Each crew member will have a unique ID (e.g., worker_id) linked to the farm location. Additional metadata can be included with the input data streams such as time of day, date, and crop type. For our farm experiments, no personal or individual names of crew members will be used. AWS Kinesis Data Firehouse will send the input data to individual files located on AWS S3 buckets. The raw and processed (filtered) data sets will be available for offline line analysis. Contining to build our Smart Safety platform will include addressing the following: Number of wearable devices: depends on the crew size. Monitoring duration: depends on the harvesting workday schedule. Data rate: determines the input bytes per second. Data Storage: determines the overall requirements for cloud storage. Study Protocol Trejo Management has agreed to provide Onda Vision Technologies with access to their worker facilities and farming operations during the experiments. Their harvesting crew starts the day at 7 am and ends the day at 7 pm. Onda Vision will outfit each person with a wearable sensor to collect the bioimpedance measurements. Each day we will start at the worker facility by collecting baseline urine samples, bioimpedance measurements, and body mass measurements. Afterwards, we will transition to the farm to collect bioimpedance measurements during the workday. Finally, we will complete the day by collecting end of day urine samples, bioimpedance measurements, and body mass measurements. Onda Vision Technologies submitted an updated IRB to reflect the updated protocol. Analyze Sensor Data Onda Vision had twenty (20)prototype wearable devices producedfrom our contract manufacturer (Protronics). These wearable devices will support the farm experiments and data collection activities. Unfortunately, due to manufacturing delays, we had to postpone our initial scheduled data collection and experiments. Therefore, we haveno major progress to present during this reporting period. The farm data collection and analysis were also delayed due to thesetechnical challenges. We plan to complete this objective once the issues have been resolved.

Publications