Recipient Organization
Geo-Spider, Inc
4406 NW 77 Terrace
Gainesville,FL 32606
Performing Department
(N/A)
Non Technical Summary
Numerous challenges face the Florida and U.S. citrus industries, including global market pressure, urban and environmental pressures, labor issues, and emerging diseases and pests, thus underscoring the need for new production approaches. Over the last decade, over 200,000 acres of Florida citrus production has been lost due to the citrus canker eradication program and citrus greening known as HLB. HLB is now considered endemic in Florida and hundreds of thousands of acres are in step decline, with minimal hope of finding remediation solutions. Long term hope rests on finding HLB resistant varieties, and developing advanced management strategies that can control the psyllid population, repel them from groves, and manage trees in an optimal economic framework. One approach being considered is Advanced Production Systems (APS) that use high density semi-dwarfed trees, and open hydroponics with optimized nutrient and water availability, which accelerates plant growth. When combined with HLB tolerant rootstocks, the concept seeks to increase yield production per acre, while simultaneously shortening the time to return on investment, which means that grove life can be shortened by disease pressure and still remain viable economically. However, this new grove architecture will require new equipment systems to manage production and harvesting. The system concept proposed will offer a broad range of production and harvesting implements.In this Phase I project, we will advance a strategic aspect of fruit harvesting which will help improve harvesting performance, quality of yield, and provide growers with valuable geo-spatial data that can be used to improve production management. It is proposed in this Phase I to develop and in-line fruit quality and yield monitor that will be integrated into the vertical lift elevator on the APS Over the Top Citrus Harvester (APS-OTCH) that has been explored in previous Phase I and Phase II projects. This technology would be an add-on option that would monitor fruit harvest yield/quality and DGPS location for either the mass harvester (juice) or selective robotic harvester (fresh) that is being developed by GeoSpider, Inc. The technology will measure fruit mass flow directly by monitoring elevator drive torque/motor pressure. In addition, fruit count, size and volume will be estimated using optical based sensors that can also provide information about fruit coloration and can be used for fruit elimation via an in-line fruit seperator. The impact of citrus greening on fruit color, size and flavor quality all suggest that elimating poor quality fruit in the field will improve packout and reduce hauling costs. In future work, additional sensors could be added to this system to detect disease presence such as, citrus canker and blackspot. An experimental prototype of the OTCH elevator will be fabricated and sensors integrated with a controller to collect sensor and DGPS data. Appropriate experimental methods will be implemented to calibrate sensor data for yield and quality estimation, and then to evaluate the system performance in a laboratory setting. In the planned future Phase II project, we will integrate the various components from Phase I projects into a fresh citrus harvester comprised of multiple arms and material handling with fruit yield and quality monitoring. The Phase II prototype can then be tested at the UF APS grove in Citra, FL.
Animal Health Component
40%
Research Effort Categories
Basic
20%
Applied
40%
Developmental
40%
Goals / Objectives
The Florida citrus industry is facing growing global market pressures that threaten its long-term viability. The combination of low commodity prices both domestically and abroad, high U.S. labor prices and low labor productivity present significant challenges for the U.S. citrus industry. Over the last decade, over 200,000 acres of Florida citrus production has been lost due to the citrus canker eradication program and citrus greening known as HLB. While citrus canker is a significant threat to fresh citrus marketability, canker does not affect the quality of processed fruit. HLB, on the other hand, if left unchecked, kills an infected mature tree within 5 years and threatens the survival of the entire Florida Citrus Industry. Long term hope rests on finding HLB resistant varieties, and developing advanced management strategies that can control the psyllid population, repel them from groves, and manage trees in an optimal economic framework. One approach being considered is Advanced Production Systems (APS) that use high density semi-dwarfed trees, and open hydroponics with optimized nutrient and water availability, which accelerates plant growth. When combined with HLB tolerant rootstocks, the concept seeks to increase yield production per acre, while simultaneously shortening the time to return on investment, which means that grove life can be shortened by disease pressure and still remain viable economically. However, this new grove architecture will require new equipment systems to manage production and harvesting. Autonomous operations and equipment are well suited to APS, which, in tree crops, necessarily involve closely spaced smallish plants. Such trees have known efficiencies in their cultural and harvesting management and constitute ideal orchard systems for easy autonomous scouting, pruning, mowing, spraying, and harvesting. Robotic and mass harvesting solutions for fresh market fruits and vegetables have been studied by numerous researchers around the world during the past several decades. PrecisionAg Technology that could cull unwanted fruit, then geo-spatially map fruit quality and yield thus localizing fruit exhibiting dangerous disease and pest pressures, could provide growers with a valuable tool for their integrated disease and pest management program and improve packout and thus profitability.In this current proposal, we will develop an integrated fruit elevator which will serve to lift fruit from tree level to top deck, and while doing so monitor fruit quality and yield, so that unwanted fruit can be culled. This real-time data can then be synchronized with GPS data to create geo-spatial maps of fruit yield and quality that can be used in production management. This technology can be applied to both our fresh fruit harvester and our mass harvester, termed Citrus Yield and Quality Monitor (CYQM). However, the primary application will be the selective harvester, where on-board fruit quality and yield monitoring is more critical due to the higher crop value.This project will leverage design work that has been done in prior work by the PD and PI's during earlier Phase I and Phase II projects. The basic elevator concept is designed in CAD, but has yet to be rendered as fabrication part drawings for a final design. Likewise, many of our formerly developed machine vision algorithms for fruit detection and disease detections will be transferrable. As a result, the list of objectives that are defined below, are in various stages of pre-development.Refine/complete design of OTP fruit elevator with yield monitor implementationDesign vision-based fruit quality sensors for count/size/shape/color monitoring systemFabricate/assemble and integrate hardware components and sensor mountingIntegrate the data acquisition and control software with hardware.Conduct laboratory based performance experiments using HLB affected fruitData Analysis and Refinement, 7) Product evaluation, and 8) Revise commercial plan.
Project Methods
We will develop an integrated fruit elevator which will serve to lift fruit from tree level to top deck, and while doing so monitor fruit quality and yield, so that unwanted fruit can be culled. This real-time data can then be synchronized with GPS data to create geo-spatial maps of fruit yield and quality that can be used in production management. This technology can be applied to both our fresh fruit harvester and our mass harvester, termed Citrus Yield and Quality Monitor (CYQM). However, the primary application will be the selective harvester, where on-board fruit quality and yield monitoring is more critical due to the higher crop value. The list of objectives were defined previously, are in various stages of pre-development.Each technical objective has various levels of prior research progress from other work by team. It is proposed in this Phase I to develop and in-line fruit quality and yield monitor that will be integrated into the vertical lift elevator on the APS Over the Top Citrus Harvester (APS-OTCH) that has been explored in previous Phase I and Phase II projects. This technology would be an add-on option that would monitor fruit harvest yield/quality and DGPS location for either the mass harvester (juice) or selective robotic harvester (fresh) that is being developed by GeoSpider, Inc. The technology will measure fruit mass flow directly by monitoring elevator drive torque/motor pressure. In addition, fruit count, size and volume will be estimated using optical based sensors that can also provide information about fruit coloration and can be used for fruit elimation via an in-line fruit ejector. An experimental prototype of the OTCH elevator will be fabricated and sensors integrated with a controller to collect sensor and DGPS data. Appropriate experimental methods will be implemented to calibrate sensor data for yield and quality estimation, and then to evaluate the system performance in a laboratory setting. The first step in developing CYQM will be the development of the mechanical elevator. Whether harvesting fresh fruit or processed fruit, the fruit flow must be elevated from tree level to the platform deck so that the fruit can be cross-conveyed to a trailing transport vehicle that will accumulate fruit in bulk for roadside to the processors, or will containerize fruit for transport to the packinghouse. Due to issues associated with a high machine center of gravity, narrow between row spacing in high density groves, and the complexity of handling transport crates, it has been determined that it will be most effective for the containerization to take place on a trailing vehicle. In which case it will leave valuable space on deck for cross conveyors, fruit grading and separation. A concept for the mass harvester catch frame and internal cross conveyor will transport fruit to the infeed of the elevator. As fruit is shaken loose from the tree it falls vertically down onto either the catch frame or directly on to the cross conveyor. If it falls on the catch frame, fruit will gravity feed into the cross conveyor. A similar system will be used for fresh fruit, with the main difference being the way that robotically harvested fruit will be conveyed down to the cross conveyor.Now in this project we focus on the elevator itself, and will add the catch frame and cross conveyor in future development. The elevator is designed to look like an "S", where the fruit feeds in on the bottom left, and exits on the right. The infeed will have a powered belt to pull the fruit into the elevator, while the discharge will rely on the momentum of the fruit as it comes over the top of the elevator head drive to propel it out of the elevator housing. The elevator will use equally spaced paddles (with small upward lip on front edge) to capture the fruit and trap it between the front and back vertical faces of the inner elevator chamber. The outer chamber will run empty.During objectives 1-3 we will refine the concept designs that have already been created in SolidWorks, and fabricate the elevator, in cooperation with the Univ. of Florida Ag & Bio Eng. (ABE) Machinery shop. During this Phase I project we plan to implement a hybrid yield and quality monitor that uses sensor fusion from different sensor data types to generate a fused output. This means that we will try to implement a sensing and control scheme in the least intrusive manner possible, yet hopefully ending up with a much improved outcome. Therefore, we plan to use sensory input from both vision-based approaches and motion/pressure/torque based approaches. All of which will be mounted on the elevator housing. Once all of the mechanical fabrication and assembly is completed, we will begin to mount the various sensors and power up the hydraulic and electrical power and control systems. We will mount most all of the sensors in the vicinity of the head shaft of the elevator. The toque sensor will be mounted in-line with the head shaft and hydraulic motor, while the encoder will be driven by a timing belt mounted on the opposite end of the head shaft from the motor and torque sensor. The motor pressure sensor will be mounted on the pressure port based on motor rotational direction. The imaging system will either be mounted directly over the head shaft where the paddles open up, or slightly away from the elevator where the fruit's momentum carries them away from the conveyor. It will be important to locate the imaging system in a manner that will allow it to synchronize with the fruit ejectors. So this may require some preliminary testing to refine the concept. In this system several dissimilar sensor data will be fused together to improve yield estimation. This is because each approach has weakness, which can diminish yield estimation.Due to logistic constraints the performance and evaluation tests will be limited to laboratory-based experiments. The shortness of time to finish development, fabricate and assemble, conduct laboratory trials, integrate on full scale harvester, and coordinate with seasonal crop availability makes field trials impractical. However, all critical performance criteria can be evaluated in a laboratory setting other than GPS integration and mapping. These are mature technologies that can be applied later in Phase II with a strong confidence of success. Once fabrication and integration are completed, the assembled test rig will be situated in one of our team laboratories at the Univ. of Florida ABE. At this point, orange fruit samples will be collected that are composed of normal fruit and HLB infected fruit that exhibit greenness, mis-shapen and smallish characteristics. Three field boxes of orange fruit will be stored at the ABE environmental chambers to retain freshness and moisture content. All boxes will be counted and sized for major/minor axis diameter and weight, along with a qualitative assessment of coloration and shape. These fruit will provide the ground truth for CYQM trials. During experiments, care must be taken to maintain a reasonable fruit flow rate. Preliminary trials with additional fruit will be ran to establish a feeding method to maintain a reasonable flow rate that won't plug the elevator, but approaches the expected mass flow rate based on predicted yield per tree and harvester ground speed. A feed conveyor that is preloaded with fruit may be sufficient.Performance evaluation will seek to demonstrate whether this novel hybrid yield/quality monitor can successfully predict mass flow rates and fruit quality in a viable economic framework, with the primary success metrics listed. Once all trials are finished, the final report will be written and the commercialization plan revised, as needed.