Progress 08/01/24 to 07/31/25
Outputs
Target Audience:Target Audience The project successfully engaged a diverse and influential group of stakeholders across the agricultural production, technology, education, and academic sectors. Key target audiences included: 1. Vegetable Growers and Producers Primary stakeholders were vegetable growers, particularly those involved in large-scale or high-value crop production in Texas. These individuals were directly impacted by innovations in cultivation, harvesting, and post-harvest automation. Their feedback was vital for tailoring the robotic solutions to field-level challenges. 2. Agri-Industry Representatives This group consisted of professionals from agricultural corporations, industry associations, and agribusiness consultants. Their participation enhanced the strategic implementation of the bio-cell and ensured the relevance of this initiative to market dynamics, supply chain efficiency, and technology transfer. 3. Retailers, Processors, and Distributors Retailers (e.g., grocery chains, farmer's markets), along with companies involved in processing and distribution, represented the post-harvest segment of the value chain. Their input ensured that the technologies developed aligned with consumer preferences, food safety standards, and logistical considerations. 4. Professional and Research Communities The project also targeted researchers, engineers, and academic collaborators active in agricultural robotics, AI, and automation. Engagement was facilitated through participation in scientific forums such as: Ubiquitous Robot 2025 - Where hardware and vision-based calibration systems for robotic harvesters were presented. AI in Agriculture and Natural Resources Conference - Where work was shared on scene graph segmentation, hierarchical reinforcement learning, and crop yield forecasting. Smart Agriculture Workshop - Where the autonomous tomato harvesting cell was demonstrated. Beltwide Cotton Conferences and SA-CSSA-SSSA Annual Meeting - Where applications of UAV and machine learning in crop monitoring and prediction were highlighted. These conferences brought together professionals from land-grant universities, federal research agencies, precision ag startups, and tech-driven farming operations. Dissemination of project outcomes through these venues strengthened interdisciplinary collaboration, promoted open innovation, and attracted interest in future commercialization opportunities. 5. Cross-Sector Stakeholders in Texas The project engaged stakeholders across the Texas vegetable industry by inviting them to participate in milestone reviews, presentations, and field demonstrations. This included growers, ag retailers, robotic hardware integrators, and policymakers. Their collective expertise and feedback helped shape the direction of the project and ensured the resulting technologies were economically viable and broadly adoptable. 6. K-12 Outreach and STEM Education To inspire future generations of agricultural engineers and technologists, the project included two K-12 summer camp sessions during the summers of 2024 and 2025. These camps introduced middle and high school students to agricultural robotics, plant science, and AI through interactive, hands-on learning experiences. By engaging youth in the STEM pipeline early, the program contributed to broader impacts in education and workforce development. Changes/Problems:Major Changes or Problems in Approach and Reason(s) During the project's implementation, several significant adjustments were made to the technical approach based on field constraints, emerging opportunities, and the need for more biologically accurate and data-driven system integration. These changes were essential to preserve the scientific integrity of the work while aligning outcomes with the practical realities of building an autonomous bio-cell system. 1. Shift from Simple Robotic Control to SAM-T-Based Multi-Task Learning Original Plan: The initial design for robotic task execution emphasized rule-based and single-function models for operations like pruning, pollination, and harvesting. Revised Approach: Field trials and early feedback revealed the limitations of hard-coded behaviors in adapting to real-world plant variability. As a result, the project adopted a Structure-Aware Multi-task Transformer (SAM-T) learning framework, allowing the robot to generalize across multiple tasks and plant morphologies. This shift enabled unified training and better use of multimodal sensory inputs, enhancing robustness under variable lighting, occlusion, and plant architecture conditions. Reason: The complexity of real tomato plants--such as irregular growth, occluded fruit clusters, and overlapping branches--necessitated a more scalable and generalizable AI model than originally planned. 2. Transition from Visual Mapping to Point Cloud-Based Digital Twin with L-System Modeling Original Plan: The project initially intended to use 2D vision-based mapping and bounding box detection to guide robotic actions. Revised Approach: We transitioned to building a 3D point cloud-based digital twin, supplemented with L-system modeling for semantic representation of plant structure. This framework allowed precise spatial understanding and supported biologically meaningful task planning (e.g., identifying suckers near internodes or fruit clusters along secondary stems). Reason: 2D imaging lacked sufficient depth and structure information to support multi-layered decision-making in confined crop environments. The use of point clouds and L-systems enabled plant-specific task localization, improved collision-free motion planning, and enhanced the interpretability of structural plant data. 3. Expansion of Physiological Monitoring in Hydroponic Environments Original Plan: The original physiological monitoring was limited to leaf color and basic temperature/humidity sensors. Revised Approach: The project expanded its physiological monitoring program to include metrics such as photosynthetic efficiency, stomatal conductance, VOC emissions (e.g., hexanal and ethylene), and real-time nutrient uptake data via EC and pH measurements. Reason: To ensure the robot could make decisions that align with plant health and phenology, deeper insight into the physiological state of hydroponic crops became necessary. This also enabled better timing for pruning and harvesting operations, directly supporting the goals of precision, minimal-intervention agriculture. 4. Adjustment in K-12 Outreach Delivery Original Plan: Outreach events were originally planned as field trips to the lab or greenhouse during the school year. Revised Approach: Due to scheduling conflicts and school calendar constraints, the outreach was restructured into two intensive summer day camp programs (Summer 2024 and 2025) focused on STEM and precision agriculture. These camps provided age-appropriate activities around robotics, sensors, and smart farming concepts. Reason: This change increased accessibility and participation, allowing the project to reach a broader and more diverse group of middle and high school students in a more immersive environment. Summary of Impact These adjustments reflect a maturing and more nuanced approach to achieving the autonomous bio-cell vision. By integrating advanced AI, structured plant modeling, and physiological data streams, the project now operates with greater scientific rigor, technical flexibility, and alignment to practical agricultural needs. All major changes have enhanced the project's ability to deliver meaningful outcomes in robotic crop cultivation, with scalable implications for greenhouse farming and smart indoor agriculture systems. What opportunities for training and professional development has the project provided?The project has provided significant opportunities for both under-graduate, graudate students and K-12 learners, promoting self-sustained research capabilities, scholarly dissemination, and early STEM engagement. 1. Graduate, under-graduateStudent Research and Mentorship Graduate students played a central role in the project's technical development, including robotics hardware integration, plant phenotyping, SLAM navigation, and AI-based decision-making in collaboration with undergarduate studnets in engineering department. Through their involvement, they gained hands-on experience in experimental design, sensor calibration, algorithm development, and system testing. Importantly, they were mentored through the entire research pipeline, enabling them to initiate and lead their own independent research activities, laying the foundation for self-sustained scholarship beyond the duration of the project. 2. Manuscript Writing and Scholarly Dissemination Graduate researchers actively contributed to multiple conference presentations and journal manuscripts, enhancing their professional writing and communication skills. Key deliverables included: Peer-reviewed journal submissions (e.g., Precision Agriculture, Remote Sensing) Presentations at international and national conferences (e.g., Ubiquitous Robot 2025, AI in Agriculture 2024) Collaborative survey publications on plant feature extraction and segmentation techniques These experiences equipped students with publication literacy, co-authorship experience, and visibility in the academic research community. 3. K-12 Education and Outreach through Summer Day Camps In line with the project's broader impacts, two STEM-focused summer day camps were organized in 2024 and 2025 for K-12 students. These camps introduced youth to core principles of precision agriculture, robotics, and sensor-based farming systems through interactive modules, demonstrations, and hands-on activities. Topics included crop monitoring with drones, soil and nutrient sensing, autonomous navigation, and plant-health detection. The program inspired interest in agri-tech careers while fostering early understanding of engineering applications in food systems. 4. Cross-Disciplinary Exposure The project's integration of robotics, data science, plant biology, and systems engineering offered students a unique interdisciplinary training environment. Students regularly participated in lab meetings, design reviews, and field deployment exercises with faculty and collaborators from agricultural science, computer science, and engineering backgrounds. In summary, the project served as a robust platform for developing highly skilled graduate researchers and nurturing early STEM curiosity among K-12 students. These dual-track training opportunities are expected to have lasting impacts on workforce readiness and pipeline development in precision agriculture and agricultural automation. How have the results been disseminated to communities of interest?The project results have been actively disseminated to both academic and industry communities through peer-reviewed publications, conference presentations, and professional society engagement, ensuring broad exposure across the domains of precision agriculture, robotics, and agricultural systems engineering. 1. Conference Presentations at Key Technical Forums The research team presented at high-visibility national and international conferences to share real-time results and solicit feedback from experts in robotics, AI, and smart farming systems: At Ubiquitous Robot 2025, two key presentations were delivered: one detailing the hardware and system design of the autonomous robotic harvesting cell [1], and another focusing on robot manipulator calibration using Intel RealSense LiDAR for precise tomato pruning [2]. These forums engaged leading robotics researchers and engineers across academia and industry. At the AI in Agriculture and Natural Resources Conference 2024, three major contributions were presented: a scene graph-based approach for tomato plant segmentation and spatial modeling using point cloud data [4], a hierarchical reinforcement learning framework for autonomous harvesting in dynamic environments [5], and an integrated view of AI-driven crop monitoring and prediction to support automation-ready plant interaction systems. The Smart Agriculture Workshop 2024 served as a targeted venue to demonstrate the prototype autonomous tomato harvesting cell to a focused audience of agricultural researchers and industry collaborators [3]. These conferences not only enabled technical exchange with domain experts but also connected the research team with potential collaborators and end-users in precision farming and greenhouse automation. 2. Journal Publications in Leading Professional Outlets Key findings were also disseminated through scholarly publications in well-regarded peer-reviewed journals: A study under review in Precision Agriculture [6] introduced a phenomic and physiological maturity assessment framework for tomato harvesting that goes beyond traditional color-based ripeness detection, directly supporting decision-making in automated harvest scheduling. A comprehensive survey of plant feature extraction and segmentation techniques was published in Remote Sensing [7], summarizing and evaluating state-of-the-art approaches relevant for agricultural robotics and digital twin construction. These publications ensure long-term accessibility of the project's findings to the broader scientific community and contribute to the growing body of knowledge in AI-integrated farming systems. 3. Engagement with Professional Societies and Research Networks Participation in society-sponsored conferences--such as those affiliated with IEEE Robotics and Automation, AI in Ag & Natural Resources, and Ubiquitous Robots--allowed the project team to reach diverse audiences spanning academia, government research labs, and private-sector innovation hubs. This positioned the project within leading-edge conversations around sustainable food systems and agri-automation technologies. In summary, the project's dissemination strategy has successfully reached communities of interest through active publication, professional presentation, and participation in interdisciplinary technical forums, ensuring visibility, peer validation, and knowledge transfer across sectors. References Vemula, N., Lee, K., Um, D., Bhandari, M. "Hardware Prototype and System Apparatus of an Autonomous Robotic Harvesting Cell," Ubiquitous Robot, College Station, TX, June 2025. Nethala, P., Um, D. "Calibration of a Three-Axis Robot Manipulator with an Intel RealSense LiDAR Camera for Tomato Pruning," Ubiquitous Robot, College Station, TX, June 2025. Um, D. "Autonomous Tomato Harvesting Cell," Smart Agriculture Workshop, College Station, TX, Dec. 17, 2024. Nethala, P., Um, D., Fernandez, O., Bhandari, M., Landivar, J., Lee, K. "Scene Graph Generation from Point Cloud Data of Tomato Plants: Segmentation and Spatial Relationships," AI in Agriculture and Natural Resources Conference, April 2024. Nethala, P., Um, D., Huang, L., Starek, M. "Hierarchical Reinforcement Learning for Autonomous Harvesting Robots in Dynamic Environments," AI in Agriculture and Natural Resources Conference, April 2024. Chandana, K., Um, D., Tabassum, S., Inam, A.S., Sangmen, E.D. "Beyond Precision Agriculture, under review. Nethala, P., Um, D., Lee, K., Bhandari, M., Fernandez, O., Vemula, N. "Techniques for Plant Feature Extraction and Segmentation: A Survey," Remote Sensing, vol. 16, iss. 23, 2024. DOI: 10.3390/rs16234370 What do you plan to do during the next reporting period to accomplish the goals?To accomplish the goals of the autonomous bio-cell initiative, the next reporting period will focus on three interrelated technical thrusts: Advanced training of a multi-task harvesting robot using a Structure-Aware Multi-task Transformer (SAM-T), Development of a point cloud-based digital twin using L-system plant modeling, and Execution of an in-depth physiological study of hydroponic tomato growth to support phenomic task planning and adaptive harvesting strategies. 1. SAM-T-Based Robotic Training for Pruning, Pollination, and Harvesting The team will continue building and refining a SAM-T framework that enables the robot to perform complex crop management tasks across heterogeneous plant structures. The SAM-T model will learn from a combination of master-slave demonstrations and previously collected visual datasets. It will be trained to: Identify and remove unwanted suckers and dead branches (pruning), Conduct artificial pollination in precise floral zones, Detect and harvest ripe fruits using a ripeness-aware selection model. The transformer-based architecture will leverage multimodal inputs, including RGB images, depth maps, and temporal motion cues, to generalize across plant sizes, growth stages, and occlusions. Training will emphasize real-time adaptability in confined bio-cell conditions. 2. Digital Twin Plant Modeling Using Point Cloud and L-System Structure To support autonomous task planning, we will continue building a high-resolution digital twin of the tomato plant using point cloud data captured by stereo RGB-D and LiDAR sensors. This spatial data will be processed into a generative plant model using Lindenmayer systems (L-systems), enabling semantic representation of: Main stem and internode structures, Fruit-bearing branches and floral zones, Growth stages and predicted maturity timing. The combination of point cloud spatial data with L-system logic provides a hybrid framework for task localization, path planning, and motion prediction. This structured representation allows the robot to reason about where and how to interact with each plant based on its biological form and growth patterns. 3. In-Depth Physiological Study of Hydroponic Tomato Growth In tandem with the digital twin and robot control pipeline, we will conduct a detailed physiological analysis of hydroponically grown tomato plants in the NFT (Nutrient Film Technique) system. Key metrics to be measured and modeled include: Photosynthetic efficiency (via wearable optical sensors), Chlorophyll content and leaf color index (via ExG/NDVI), Stomatal conductance (via humidity and gas exchange probes), Fruit maturation signals (VOC emissions: hexanal, ethylene), Nutrient uptake and water status (EC/pH balance, root zone temperature). These data will inform both the AI's decision-making model (e.g., whether a fruit is ready for harvest) and the L-system's adaptive growth model. Ultimately, this physiological insight supports real-time crop health feedback, ensuring robotic actions are biologically appropriate and synchronized with plant development. Additional Goals for the Period Finalize SAM-T model integration on the Jetson hardware for on-device learning Validate digital twin with physical greenhouse observations Compare physiological profiles between NFT and soil-grown tomatoes Submit two journal papers (one on task-based learning, one on physiology-informed autonomy) Present findings at the 2025 IEEE/ASABE and AI in Agriculture conferences Through these activities, the project will deepen its integration of robotic autonomy, plant structure modeling, and biological intelligence to deliver a scalable, self-managing cultivation system--realizing the vision of the autonomous bio-cell.
Impacts
What was accomplished under these goals?
Project Achievements This research project aimed to advance the concept of an autonomous bio-cell--a closed-loop crop production system capable of performing cultivation tasks with minimal human intervention. The core objectives were: i) the development and implementation of engineering technologies for plant phenotyping and autonomous task execution, ii) the integration of computational and big data methods for decision-making, and iii) the design and automation of crop production tasks such as pollination, pruning, and harvesting. The following milestones were successfully achieved: 1. Development and Implementation of Autonomous Systems A fully functional hardware prototype of an autonomous robotic harvesting cell was developed and demonstrated [Vemula et al., Ubiquitous Robot 2025]. This system integrates mobility, manipulation, and real-time sensing for in-situ tomato harvesting tasks. Its deployment within a bio-cell setup demonstrated reliable navigation, plant interaction, and fruit detachment, validating the feasibility of autonomous cultivation within enclosed agricultural systems. 2. Precision Calibration and Task-Specific Manipulation Efforts were made to enhance precision control and spatial awareness of robotic manipulators. A calibration framework using Intel RealSense LiDAR was developed for 3-axis robots to improve the accuracy of pruning tasks [Nethala & Um, Ubiquitous Robot 2025]. This technical advancement supports the safe and selective removal of suckers and unnecessary branches in confined environments. 3. Advanced 3D Plant Modeling and Spatial Reasoning Using high-resolution point cloud data, a scene graph generation pipeline was proposed for capturing the spatial and structural layout of tomato plants [Nethala et al., AI in Agriculture 2024]. This enables real-time segmentation and localization of plant features, which is crucial for path planning, pruning, and selective harvesting within a digitally simulated crop environment. 4. Hierarchical Learning for Autonomous Decision Making To address complex task sequencing, a hierarchical reinforcement learning architecture was introduced for autonomous harvesting in dynamic crop environments [Nethala et al., AI in Agriculture 2024]. This architecture allows the robot to manage multi-stage tasks--such as identifying, grasping, and harvesting fruit--while adapting to unpredictable plant growth patterns. 5. Physiological and Phenomic Trait-Based Ripeness Detection Expanding beyond conventional color-based maturity metrics, the team developed a novel multi-modal tomato harvest maturity assessment system that integrates physiological signals and environmental data in a hydroponic NFT system [Chandana et al., under review, Precision Agriculture]. This work demonstrates a shift toward phenotype-driven harvesting and supports one of the fundamental goals of the autonomous bio-cell: real-time, biology-informed decision-making. 6. Knowledge Integration and Survey of Feature Extraction Techniques To unify the state-of-the-art approaches in plant sensing and recognition, a comprehensive survey of plant feature extraction and segmentation techniques was published [Nethala et al., Remote Sensing, 2024]. This review supports the selection of optimal sensing and modeling strategies for robotic implementation within data-driven farming environments. Collectively, these outcomes demonstrate significant progress toward the realization of an autonomous bio-cell. From core engineering designs to applied AI frameworks and phenomic integration, the project has built a multi-layered foundation for autonomous, scalable, and intelligent vegetable production systems.
Publications
- Type:
Peer Reviewed Journal Articles
Status:
Published
Year Published:
2024
Citation:
Prasad Nethala1 , Dugan Um1*, Kiju Lee2 , Mahendra Bhandari3 , Oscar Fernandez Montero4 , Neha Vemula5, Techniques for Plant Feature Extraction and Segmentation: A Survey, Remote Sensing (IF=4.2/SCI), vol. 16, iss. 23, 2024, https://doi.org/10.3390/rs16234370
|
Progress 08/01/23 to 07/31/24
Outputs
Target Audience:Target Audience Progress Report: (First-Year) During the inaugural year of the project, significant strides were made in identifying and engaging with various key audiences. The focus was primarily on stakeholders within South Texas, aiming to understand their needs and interests in the context of the research project titled "Revitalizing Tomato Farming in South Texas via Autonomous Growing and Harvesting Cell." 1. Farmers Market Attendees: Engaging with individuals at local farmers' markets was a crucial step in assessing the interest in locally-grown tomato products. The feedback gathered revealed a sustained enthusiasm for locally sourced produce. Notably, there was a specific interest expressed in locally grown, affordable strawberries with a continuous and reliable supply. This insight underscores the potential market demand for regionally produced fruits, and it serves as a valuable foundation for tailoring the project to meet consumer expectations. 2. Zordi CEO - Robotics in Agriculture: A pivotal meeting was held with the CEO of Zordi, a company recognized for pioneering the use of cutting-edge robotics in strawberry production. The key takeaway from this interaction was the importance of addressing cost-related challenges throughout the entire production cycle and market introduction. Understanding the successful strategies employed by a company with expertise in autonomous farming technologies will be instrumental in shaping the project's implementation and ensuring its viability in the broader agricultural landscape. 3. Corpus Christi Bay Productions LLC - Local Vegetable Producer: Engaging with a local vegetable producer, Corpus Christi Bay Productions LLC, provided valuable insights into the challenges and opportunities within the local farming community. Their perspective contributed to a more comprehensive understanding of the existing agricultural ecosystem in South Texas. Establishing collaborative relationships with local producers will be essential for fostering a supportive network and ensuring the success of the autonomous growing and harvesting initiative. 4.In July, asummer camp was offered to ten K-12 students. As part of the program, the students visited the Texas A&M University-Corpus Christi (TAMUCC) robotics lab on Friday morning. During the visit, a short lecture was presented, highlighting NIFA's commitment to supporting smart farm technology as a means to address environmental challenges.The students were introduced to an indoor hydroponic tomato growing system, where they participated in a fruit counting activity and interactive game. Following this, they engaged in hands-on experiences with automatic harvesting technology, providing them with practical insights into advanced agricultural practices. In summary, the first-year interactions with these diverse audiences have laid a solid foundation for the project. The enthusiasm observed at farmers' markets, the strategic insights gained from Zordi's CEO, and the local context provided by Corpus Christi Bay Productions LLC collectively contribute to shaping the project's direction. Moving forward, continued engagement with these and additional stakeholders will be critical for refining the project's goals and ensuring its alignment with the needs of the South Texas agricultural community. Changes/Problems:During the first year of the project, we evaluated various hydroponic plant growing systems, including Deep Water Cultivation (DWC), Nutrient Film Technique (NFT), and aeroponic systems. Our trials revealed that disease control in hydroponic systems, although initially considered less critical compared to soil-based growth, is essential for producing high-quality products. As a result, we plan to studyplant disease management in hydroponic systems using wearable plant sensors, such as Fiber Bragg Grating (FBG) sensors. Additionally, we observed that accurately sensing harvest readiness is crucial for the effective operation of autonomous harvesting systems. To address this, we intend to investigate microwave-based maturity scanning technology,Raman spectroscopy, or other creative methods to assess ripeness levels. What opportunities for training and professional development has the project provided?Training and Professional Development Opportunities Offered by the Project: The research project, "Revitalizing Tomato Farming in South Texas via Autonomous Growing and Harvesting Cell," has actively contributed to fostering training and professional development opportunities for individuals including faculty and graduate and undergraduate students involved in various capacities. The project's commitment to innovation and interdisciplinary collaboration has created a dynamic learning environment, offering the following opportunities: 1. Hydroponic Agriculture Training: - all of the project participants are engaged in hands-on training sessions focused on hydroponic agriculture practices. - Participants gained practical insights into germination, cultivation, and harvesting processes for a variety of crops, including tomatoes, cucumbers, and microgreens in hydroponic growthsystem. 2. Technological Skill Enhancement: - Graduate studentsreceived training in cutting-edge engineering technologies, including the development and implementation of a 3-axis robotic harvesting system and a delicate 6-DOF robotic hand for specialized harvesting. - Undergraduate students acquired skills in designing and optimizing a hydroponic growing chamber, showcasing a practical application of engineering principles in agriculture. 3. Data Analytics and Interpretation: - Training sessions were conducted to enhance proficiency in advanced computational methods and big data analytics for graduate students. - Participants learned to interpret and utilize agricultural systems data effectively, contributing to informed decision-making and improved project outcomes. 4. Interdisciplinary Collaboration: - Collaborative engagements with industry experts, such as the CEO of Zordi and CC pro, provided participants with unique insights into real-world challenges and solutions in autonomous farming technologies. - Enhanced understanding of the intersection between agriculture, technology, and business, fostering interdisciplinary collaboration. 5. Project Management Skills: - Team members gained experience in managing and coordinating research activities within the context of a multifaceted project. - Developed skills in project planning, execution, and evaluation, contributing to overall professional growth. 6. Networking and Industry Exposure: - Opportunities for networking were facilitated through interactions with attendees at local farmers' markets, CEOs of agricultural technology companies, and local vegetable producers. - Exposure to diverse perspectives and industry stakeholders broadened participants' professional networks. 7. Innovation in Agricultural Practices: - Engagement in research activities, such as the exploration of upside-down plant growth and light spectrum optimization, provided participants with opportunities to contribute to innovative agricultural practices. - Encouraged a mindset of continuous learning and adaptation to emerging technologies in the agricultural sector. Overall, the project has served as a platform for comprehensive training and professional development, equipping participants with a diverse skill set that spans agricultural practices, engineering technologies, data analytics, and interdisciplinary collaboration. These opportunities not only contribute to the success of the research project but also empower individuals with the skills and knowledge essential for advancing their careers in the evolving landscape of agricultural innovation. - Collaborative engagements with industry experts, such as the CEO of Zordi and CC product LLC, provided participants with unique insights into real-world challenges and solutions in autonomous farming technologies. - Enhanced understanding of the intersection between agriculture, technology, and business, fostering interdisciplinary collaboration. 5. Project Management Skills: - Team members gained experience in managing and coordinating research activities within the context of a multifaceted project. - Developed skills in project planning, execution, and evaluation, contributing to overall professional growth. 6. Networking and Industry Exposure: - Opportunities for networking were facilitated through interactions with attendees at local farmers' markets, CEOs of agricultural technology companies, and local vegetable producers. - Exposure to diverse perspectives and industry stakeholders broadened participants' professional networks. 7. Innovation in Agricultural Practices: - Engagement in research activities, such as the exploration of upside-down plant growth and light spectrum optimization, provided participants with opportunities to contribute to innovative agricultural practices. - Encouraged a mindset of continuous learning and adaptation to emerging technologies in the agricultural sector. Overall, the project has served as a platform for comprehensive training and professional development, equipping participants with a diverse skill set that spans agricultural practices, engineering technologies, data analytics, and interdisciplinary collaboration. These opportunities not only contribute to the success of the research project but also empower individuals with the skills and knowledge essential for advancing their careers in the evolving landscape of agricultural innovation. How have the results been disseminated to communities of interest?Dissemination of Results to Communities of Interest: Ensuring transparency and sharing the outcomes of the research project, "Revitalizing Tomato Farming in South Texas via Autonomous Growing and Harvesting Cell," with communities of interest has been a priority. The project has employed various channels and methods to disseminate results, fostering engagement and collaboration: 1. Community Workshops and Seminars: - Organized workshops and seminars tailored for local communities including faculty on campus, where project researchers presented key findings and insights. - Provided an interactive platform for community members to ask questions, offer feedback, and actively engage in discussions related to autonomous farming technologies and the revitalization of tomato farming. 2. Participation in Islander day: - Engaged with potential student sand their family to showcase project developments and outcomes in smart farm research. - The current research project has been introduced in local public to communicate the relevance and potential benefits of the autonomous bio-cell concept in tomato farming. 3. Collaboration with Agricultural Extension Services: - Collaborated with agricultural extension services to facilitate the dissemination of research results. - Shared practical information and best practices derived from the project to assist local farmers in adopting innovative technologies and enhancing their farming practices. The multifaceted approach to disseminating results ensures that information reaches diverse audiences, including farmers, industry professionals, educators, and the general public. By utilizing both traditional and digital communication channels, the project aims to create a well-informed and engaged community that actively participates in and benefits from the advancements in autonomous tomato farming in South Texas. What do you plan to do during the next reporting period to accomplish the goals?Next Reporting Period Goals and Action Plan: To further advance the goals of the project, "Revitalizing Tomato Farming in South Texas via Autonomous Growing and Harvesting Cell," the next reporting period will focus on in-depth research in autonomous pruning and harvesting. The strategic plan includes the following key elements: 1. Autonomous Pruning and Harvesting Research: - Objective: Develop and implement autonomous systems for efficient pruning and harvesting of tomato plants. - Actions: - Conduct comprehensive research on state-of-the-art techniques for autonomous pruning and harvesting in agricultural settings. - Explore innovative approaches to enhance the precision and efficiency of pruning processes, ensuring optimal plant growth and fruit production. - Investigate advanced technologies and methodologies for autonomous harvesting to streamline the process and maximize crop yield. 2. Focus Areas for In-Depth Research: - Objective 1: 3D Digital Twin Modeling: - Develop detailed 3D digital twin models of tomato plants, allowing for accurate representation and simulation of their growth patterns and structures. - Leverage digital twin technology to enhance understanding and decision-making in autonomous farming processes. - Objective 2: Symmetric Plant Descriptor Generation: - Investigate techniques for generating symmetric plant descriptors to facilitate precise identification and characterization of plant structures. - Enhance the efficiency of autonomous systems by creating comprehensive descriptors that capture the nuances of plant morphology. - Objective 3: Fruit Identification and Classification AI Study: - Implement artificial intelligence (AI) algorithms for the identification and classification of tomato fruits. - Train AI models to distinguish between ripe and unripe fruits, enabling targeted and efficient harvesting strategies. - Objective 4: Robotic Motion Control: - Develop and implement a 3-axis Cartesian version for robotic motion control, ensuring precise and coordinated movements for autonomous pruning and harvesting. - Explore the integration of a 6-DOF dextrous robotic version to enhance the adaptability and versatility of the robotic systems. 3. Collaboration and Integration: - Foster collaboration with experts in AI, robotics, and agricultural engineering to leverage interdisciplinary insights. - Integrate research findings from each focus area to create a cohesive and effective autonomous system for pruning and harvesting. 4. Validation and Testing: - Conduct rigorous validation tests in controlled environments to assess the accuracy and efficiency of the developed autonomous systems. - Collaborate with local farmers and agricultural communities to perform on-field testing, ensuring the practical applicability of the technologies in real-world scenarios. 5. Community Engagement: - Organize community workshops and information sessions to update local stakeholders on the progress of the research and gather valuable insights. - Seek feedback from farmers, industry professionals, and community members to enhance the project's relevance and impact. By concentrating efforts on these key areas, the next reporting period aims to make substantial strides in the development and implementation of autonomous systems for pruning and harvesting. The integration of cutting-edge technologies and interdisciplinary collaboration will contribute to achieving the overarching goals of the project, thereby advancing the revitalization of tomato farming in South Texas.
Impacts
What was accomplished under these goals?
Significant Achievement: Integration of Autonomous Bio-Cell Concepts in South Texas Agriculture The culmination of the first-year efforts in our research project, "Revitalizing Tomato Farming in South Texas via Autonomous Growing and Harvesting Cell," represents a significant achievement in advancing the autonomous bio-cell concept in agriculture. 1. Target Audience Contact: - Farmers' Market Engagement: Our interactions with attendees at local farmers' markets provided valuable insights into the strong interest in locally grown produce, particularly affordable strawberries. This information is crucial in aligning our autonomous bio-cell concept with the demands and preferences of the local community, setting the stage for a meaningful impact on South Texas agriculture. - Industry Collaboration: Meeting with the CEO of Zordi, a company excelling in robotic strawberry production, emphasized the importance of addressing cost-related challenges. This insight significantly influenced our approach to the project, reinforcing the need for a comprehensive solution from production to market introduction. - Local Producer Collaboration: Engaging with Corpus Christi Bay Productions LLC, a local vegetable producer, provided contextual understanding and fostered collaboration within the local agricultural community. Building such partnerships is pivotal for the success and sustainability of the autonomous growing and harvesting initiative. 2. Significant Products/Outputs: - Hydroponic Plant Growing Education: Developed foundational knowledge in hydroponic agriculture, successfully germinating and harvesting various crops within a 100-day period. This achievement sets the groundwork for implementing hydroponic practices within the autonomous bio-cell concept. - Key Research Activities: Initiated the development of a 3-axis robotic harvesting system, designed a delicate 6-DOF robotic hand, and explored the potential of upside-down plant growth. These activities are tangible steps toward realizing the autonomous capabilities outlined in the project proposal. - Light Spectrum Optimization: Through rigorous experimentation, we identified the optimal light spectrum--white and red light--for promoting tomato plant growth. This finding contributes directly to the efficiency and productivity goals of the autonomous bio-cell. 3. Significance of Achievements: - Our progress aligns seamlessly with the proposed research objectives, demonstrating a successful integration of engineering technologies, advanced computational methods, and big data analytics into the agricultural domain. - By developing and applying automation to tasks such as pollination, pruning, and autonomous harvesting, we are moving closer to the envisioned autonomous bio-cell that can revolutionize crop production in South Texas. - The achievements underscore the feasibility of implementing the autonomous bio-cell concept in diverse agricultural settings, with potential applications ranging from open fields to greenhouses or buildings. In summary, the significant achievements made during this reporting period reflect not only progress toward the outlined research goals but also a strategic alignment with the needs and preferences of the local community. The integration of autonomous bio-cell concepts, coupled with tangible developments in hydroponic practices and robotic technologies, positions the project as a transformative force in revitalizing tomato farming in South Texas.
Publications
- Type:
Journal Articles
Status:
Under Review
Year Published:
2024
Citation:
Prasad Nethala1 , Dugan Um1*, Kiju Lee2 , Mahendra Bhandari3 , Oscar Fernandez Montero4 , Neha Vemula5, Techniques for Plant Feature Extraction and Segmentation: A Survey, Remote Sensing, MDPI, 2024 (submitted)
|