The Agricultural Technology Research Program (ATRP) of the Georgia Tech Research Institute has approved a total of 10 research projects for the financial year 2011. The projects address research priorities in one of ATRP's five technical focus areas (advanced imaging and sensing, environmental and energy, information technology, robotics and automation, worker and product safety) and consist of continuation and new applied research as well as some special exploratory research topics. In addition to these research projects, the programme also supports an active technical assistance and technology transfer and outreach programme.

Continuation Research Projects

Worker Safety Research for the Poultry Industry

PI: Sim Harbert
Worker safety is an important and relevant concern for the poultry industry, as every year there are injuries that occur due to the harsh working environment. The industry’s list of top ergonomic concerns includes both highly repetitive movements such as deboning and lifting tasks. The research activities under ATRP in the area of worker safety align directly with these priorities under the following two tasks. The first task is a continuation of the EWAS (Ergonomic Work Assessment System) study to collect real worker data and study the differences between experienced and inexperienced workers as well as the levels of exposure for different deboning techniques at different line speeds and the impact on yield. The second study involves conducting an analysis of the efficacy of the Wii-Fit gaming system as a conditioning and/or prevention program for lifting tasks. Based on the results of the second study, a prototype will be created for a system to wirelessly capture the movement of workers while moving about the plant throughout their entire shift using compact hardware modules.

Novel Separation Technologies for Poultry Processing Liquid Streams

PI: John Pierson
Researchers are working to improve liquid stream reuse or rendered oil quality (e.g. reduce physical, chemical and microbiological contamination so as to prevent contamination or adulteration of product) via improved separations using dynamic filtration and adsorption chromatography systems. By targeting contaminant separations (e.g. fats, proteins, water-insoluble impurities), value-added by-product recovery and more rapid, cost-effective, and accurate detection of chemical and/or biological agents (either accidentally or intentionally placed in the food supply) should also be accomplished. Preliminary work using a VersaFlash Flash Chromatography System with different media and an initial analysis of key parameters affecting membrane separations and filtration indicate that adsorption and novel dynamic filtration separation techniques can be applied to selected poultry processing liquid streams (i.e. polished oil, both red chiller and chiller waters). FY 2011 efforts will expand adsorption work to functionalised adsorbents and simulated moving bed technologies, while dynamic filtration research will examine methods to improve separations, especially when pre-treating liquid streams in advance of adsorption systems.

Monitoring Bird Status in Broiler Housing Using Audio and Video

PI: Wayne Daley
It is well known that environmental conditions during broiler grow-out can affect the performance of the birds (usually measured as feed conversion ratio). The question to be addressed under this research is whether or not the bird itself can be used as a sensor. The goal of the research is to monitor various video and audio parameters of the birds to determine the correlation between environmental or medical conditions and their audio and video time series data. The concept is that these parameters could be useful as early predictors of flock health and performance. Work from the previous year indicates that it is possible to discriminate deviations from the normal temperature conditions by monitoring the audio. Results from the video are not as conclusive. In FY 2011, researchers will further investigate and confirm the use of bird vocalizations as early indicators of other production and environmental conditions. If successful, this audio monitoring could be used to take preemptive or corrective actions to maintain the health and viability of the flock, which in turn could improve production efficiency and bird welfare.

Cone Line Bone Detection System

PI: Wayne Daley
Researchers have demonstrated a new approach for the automatic screening of bone on the cone line by inspecting the frame (skeleton) after the meat has been removed. This not only assists in detecting bones that remain in the product, but also provides an opportunity to do real-time monitoring of production yield. The Missed Bone Screening System uses a special cone with internal illumination that backlights the frame giving the appearance of an x-ray image. The bones of interest are the clavicle and fan bones. The system was tested off-line and showed promise for implementation in a production environment. Initial results indicated there is a 0.5 per cent chance that the system would reject a product without bone fragments, and a 7.5 per cent chance that it would pass a product with bone fragments. Although the concept was demonstrated initially using a plastic cone, partner and stakeholder feedback has led to efforts focused on implementing the system using a steel cone. A laboratory prototype using the modified steel cone has been constructed, and FY 2011 plans are to conduct field trials/demonstrations along with algorithmic and software support to evaluate the concept under production conditions.

Intelligent Cutting and Deboning System

PI: Gary McMurray
Researchers have continued the development of the three key technology components of the Intelligent Cutting and Deboning System: tendon prediction system, bone detection algorithm, and force control to guide the blade around the bone. The team has been able to verify that the tendon prediction system is accurate to ±3mm. The image processing required for the tendon prediction is still being refined. The team has also been able to demonstrate a consistent bone detection algorithm for first contact of the blade with the bone as well as a simple force control algorithm to guide the blade around the bone. The force control will be optimized in the coming year. The system is now able to meet the testing goal of one bird every three minutes (the three minutes is taken up entirely by the setup of the bird in the system, which would not be required in a real system). For FY 2011, the team will begin to assemble the technology components into a prototype to allow for a yield test to be performed.

New Research Projects

Noise Abatement for Vibratory Conveyors

PI: Ai-Ping Hu
Noise abatement is a growing concern in industrial settings that are increasingly filled with machinery and automation. Vibratory conveyors are used widely in the food processing industry, including in poultry processing plants. Compared to more traditional gear-and-belt conveyors, they contain far fewer moving parts and exposed surfaces, and thus offer the advantage of being significantly easier to keep clean and maintain. Vibratory conveyors operate by essentially shaking and throwing product small distances toward a forward-progressing direction. The ensuing noise is generated by both the actuation of the vibratory tray surfaces and the collision between the surfaces and the (often frozen) product. The goal of this research project is to explore noise abatement for vibratory conveyors. Research activities will proceed along two tracks: (1) applying input shaping to modify the drive signal sent to the conveyor's actuator and (2) erecting acoustical shrouding structures to contain the noise.

Rapid Pathogen Detection Based on Surface Enhanced Raman Spectroscopy (SERS)

PI: Wayne Daley
The detection of pathogens in the food production chain is of significant concern to producers and consumers. A large number of different technologies have been developed for pathogen detection using optical, electrochemical, biochemical, and physical properties, as well as the conventional detection methods such as dry weight measurement, viable counting, and turbidity measurements. Conventional microbiological methods for determining the cell counts of bacteria employ selective culture, biochemical, and serological characterization. Although these achieve sensitive and selective bacterial detection, they typically require days to weeks to obtain a result. Other detection methods utilize equipment that is not only complex and expensive but also lacks portability. Researchers believe rapid pathogen detection can be accomplished using Surface Enhanced Raman Spectroscopy (SERS), specifically a Raman reporter labeled gold nanoparticle. Antibodies specific to the target of interest can be covalently linked on the shell surface for SERS measurement. SERS-coded gold nanoparticles are found to be highly stable and to be detectable down to attomolar particle concentrations because the gold nanostructure strongly enhances the Raman scattering signals. This project seeks to develop an ultra-sensitive, low-cost, field-usable pathogen detection prototype based on SERS.

Special Research Projects

Envisioning the Poultry Plant of the Future

PI: Wayne Daley
The goal of this project is to develop a vision for the poultry plant of the future. The plant of the future should be designed from the ground up with food safety and efficient use of natural resources (water, energy, and space) in mind. The vision is built on engineering models of the subsystems that validate the effectiveness of the concepts. The final result of the project is a list of technologies that are required to achieve the goals and a technology road-map to support the developed vision.

Chicken Egg Fertility Detection

PI: Robert Wallace
Detecting initial fertility in eggs quickly is of great interest to the poultry industry. Without early detection, valuable time and space is wasted on eggs that will never produce a chicken; depending on the age of the laying hens, 10 per cent to 30 per cent of eggs may be infertile. The focus of this project is to explore methods for rapidly testing and detecting the fertility of the eggs before they are placed in an incubator. Such a test would enable the poultry industry to save time and space in incubators by reducing the number of infertile eggs delivered by older laying hens. Fellow researchers at the University of Georgia estimate 100 per cent fertility could mean an additional $500,000 to an average size hatchery annually.

Product Sensing for Manipulation Using Ultrasonic Holography

PI: Wayne Daley
The ability to sense and locate whole parts for transfer and handling could significantly improve the automation of tasks such as cone loading, cut up, and deboning. Initial work has demonstrated that time-of-flight cameras can be used to generate 3D information for locating and handling product. In a separate initiative, ultrasonic holographic technology showed some promise for locating structures internal to the bird. The combination of these two technologies could help to address some of the sensing needs for current and future automation systems. This work will explore the potential of ultrasonic holographic techniques to provide information on the internal structure of poultry carcasses that could be useful for first and second processing.

Technical Assistance and Technology Transfer and Outreach Program Activities

Technical Assistance

Activity Leader: Doug Britton
Each year more than 30 technical assists, including workplace safety consultations, are provided to companies and individuals in Georgia’s poultry industry across the state. These assists range from simple inquiries regarding information or help needed to address a problem to extensive on-site consultations in which researchers collect data and provide a full report on their findings and recommendations. Over the past few years, the program has also offered in-plant energy assessments designed to help companies identify strategies of reducing energy usage and costs. ATRP uses input from all assists to gauge situations calling for new research initiatives.

Technology Transfer and Outreach Activities

Project Director: Angela Colar
On the Technology Transfer front, ATRP produces the programme newsletter PoultryTech and an in-depth Annual Report, both award-winning publications. Staff researchers produce numerous articles and technical presentations on research discoveries and emerging technologies while also generating patents and records of invention. The program also hosts a national conference on workplace safety in the poultry industry; exhibits at a range of industry events to provide information on research activities; provides support to the Information Systems Seminar for the US Poultry & Egg Association; and coordinates 'Poultry World', an exhibit designed to promote the poultry industry and its importance to Georgia, at the Georgia National Fair.