Source: PRAG LLC submitted to NRP
ADVANCED AUTONOMOUS BLACK SOLDIER FLY LARVAL UPCYCLING FOR AGRICULTURAL PROTEIN PRODUCTION & COMPOSTING VIA CELLULAR AUTOMATA MACHINE LEARNING
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
COMPLETE
Funding Source
Reporting Frequency
Annual
Accession No.
1031710
Grant No.
2024-70031-41621
Cumulative Award Amt.
$125,000.00
Proposal No.
2024-00293
Multistate No.
(N/A)
Project Start Date
Jul 1, 2024
Project End Date
Feb 28, 2025
Grant Year
2024
Program Code
[8.12]- Small and Mid-Size Farms
Recipient Organization
PRAG LLC
171 FRANKLIN RD
LAKE MARY,FL 32746
Performing Department
(N/A)
Non Technical Summary
Our research aims to automate the BSFL compostingand preservation process into a single, off-grid system, so that it can efficiently produce preservedlarvae for farmers like Mr. Plasencia (a farmer we interviewed), helping him save his valuable time, keeping the larvae outof Mrs. Plasencia's kitchen, and substantially reducing the cost to feed his chickens. For Mr.Plasencia, the elegance of the design is that it would produce the feed without any input (asidefrom the initial compost and egg loading), allowing him during particularly busy weeks to ignorethe system entirely for prolonged periods, and still find the full harvest of larvae, preserved andready for his chickens when he has the time to check. And because the system is off-grid, he wouldbe free to place the system anywhere on his property, including far to the back of his land near thechickens, without having to stretch expensive and potentially hazardous power cables that far.
Animal Health Component
50%
Research Effort Categories
Basic
20%
Applied
50%
Developmental
30%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
40352302020100%
Knowledge Area
403 - Waste Disposal, Recycling, and Reuse;

Subject Of Investigation
5230 - Feed and feed additives;

Field Of Science
2020 - Engineering;
Goals / Objectives
This project seeks to bolster the efficiency and viability of small farms by improving andautomating black soldier fly larvae (BSFL) composters. These composters are currently used bymany farms for their ability to turn farm waste biomass from plants and animals into a highlyprized feed for fowl and fish. However, inconveniently for small farms, larval feed must be usedwithin 1-5 days, or it will pupate into flies and be lost. For small farmers, current means of larvalpreservation are sorely misaligned with their needs. Huge investments in industrial BSFLpreservation machines, or the improvised use of appliances like ovens and freezers shared for homefood use are the only options. To solve the disconnect, the research proposed makes novel use ofmachine learning and recycles the thermal energy produced in the composting process to fullyautomate the preservation of the larval harvest onboard the composter -- without any inputs fromthe farmer necessary. This allows for the sustainable production of valuable feed, fully preservedand ready for use by the farmer on-demand.
Project Methods
We propose a fully autonomous composter, harvesting its own heatenergy in the composting process, enabling dehydration and retention of larvae in an efficient,hands-free manner. We call our composter design an 'Advanced, Autonomous RetentionComposter,' or AARC. Automation of BSFL production and preservation leverages multipletechnologies: As in traditional composters, Hermetia Illucens eggs and compost are introduced ina closed space. By using machine learning, the composter's internal geometry will be redesignedto concentrate composting thermal energy into a more central point. There, it not only acceleratesthe composting process, but is used by a strategically placed thermoelectric generator (TEG) 1toproduce an electric current via the Seebeck effect. This current will charge a series of batteriesover the course of 15 days to generate sufficient stored wattage for the eventual dehydration oflarvae¹?. Near the end of the 15-day cycle, Hermetia Illucens larvae, having eaten their fill,naturally seek higher ground for pupation -- and after climbing up prepared ramps inside thecomposter, the larvae fall off the sheer edge at the end of the ramps and self-harvest into a heatingchamber¹?, whereupon a plate detects the increasing mass. This triggers a countdown in thechamber, which will utilize the stored power in the battery array to dehydrate the collected larvaebefore they are able to pupate into flies. Machine learning is again used to determine the optimalconditions to begin the dehydration sequence, calculating the ideal method for future use to ensurethe process goes as efficiently as possible each time. The dehydrated larvae, now fully sanitizedand preserved by the process, will be deposited into an optionally sealed container at the bottom,ready for use by the farmer whenever.

Progress 07/01/24 to 02/28/25

Outputs
Target Audience:The target audience for this project was small- to medium-sized farmers in rural or low-income areas who raise chickens, fish, or other livestock and are interested in lowering their feed and waste disposal costs. These farmers often lack time, labor, or financial resources to process black soldier fly larvae (BSFL) efficiently. We focused on designing a solution that automates BSFL composting and preservation, reducing the need for external energy sources or manual labor. Farmers like Mr. Plasencia, who was interviewed and provided input, represented our primary use case: individuals seeking affordable, off-grid tools to turn organic waste into protein-rich feed with minimal effort. Our efforts were driven by the goal of empowering these farmers to improve self-sufficiency, reduce costs, and reduce waste using accessible, low-maintenance technology. Changes/Problems:The primary issue encountered was insufficient power generation from the composting process. Despite various geometry optimizations and thermoelectric generator (TEG) configurations, the voltage output was too low to reliably charge the battery bank needed for dehydrating the larvae. As a result, we were unable to complete autonomous preservation of BSFL as originally intended. This led us to shift focus toward data collection, behavioral observation, and thermal profiling to better understand the limitations of passive energy harvesting in compost environments. These insights will inform future design iterations that may incorporate supplemental power sources or improved heat concentration methods. What opportunities for training and professional development has the project provided?This project provided hands-on training in thermodynamic modeling, compost system design, additive manufacturing, and biological experimentation. Team members gained practical experience in simulating and fabricating heat-optimized composters, managing BSFL growth cycles, and working with thermoelectric energy systems. The interdisciplinary nature of the work also helped build skills in mechanical assembly, environmental monitoring, and data logging. For several team members, it was their first opportunity to manage a full R&D cycle--from hypothesis and design to prototyping and field testing--enhancing both technical and project management capabilities. How have the results been disseminated to communities of interest?We shared our results through direct community outreach and public engagement. Presentations were given to 5local garden clubs, as well as to professional organizations such as Rotary Club chapters. We also spoke to students and faculty at the University of Central Florida, Valencia College, and Rollins College. Additionally, we discussed our findings and lessons learned on three public podcast appearances, helping raise awareness around BSFL composting and sustainable feed alternatives. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? We designed, built, and tested multiple prototype AARCs (Advanced Autonomous Retention Composters) to evaluate the feasibility of autonomous BSFL composting and preservation. Through these tests, we collected detailed thermal data, developed geometry simulations for internal composter layout, and conducted multiple cycles of larval rearing. We learned valuable lessons about BSFL behavior, humidity sensitivity, and the challenges of heat-based power harvesting. Ultimately, our efforts were unable to produce the voltage output necessary from the thermoelectric generators to sustain reliable battery charge for dehydration, but the process yielded significant insight into compost thermodynamics, larval distribution, and design improvements for future iterations. We have not given up on the process yet, and aim to iterate in the future ourselves.

Publications