Integrated Systems Research and Development in Automation and Sensors for Sustainability of Specialty Crops

NON-TECHNICAL SUMMARY: Increased consumer demand for local foods will result in the development of local specialty crop markets. Expansion of local specialty crop production, however, is often constrained by low labor availability and high labor costs. Engineered sensors, electronics, and automation technologies offer the potential to revolutionize practices, address constraints to expanded local specialty crop production, and insure high food quality and safety. However, the basic engineering science required to apply these technologies to specialty crops is limited. <P>The overall purpose of this project is to thus extend engineering science and technology at the interface between engineered systems and specialty crop cultural systems. Weed management, for example, is critical to ensure high specialty crop yield, but represents a technical challenge as conventional weed control solutions can be difficult to implement in vegetable production. Herbicide-based weed control has a high embodied energy cost and is limited to non-organic systems. For small growers, manual weed control may be an option, but can be a major constraint to production expansion. Engineering solutions that integrate sensors and actuators for automated mechanical intra-row weed control are needed. <P>

Food safety concerns are rising in specialty crops production. Due to perceived health benefits, more fresh vegetables and fruits are consumed with a minimal processing. Contamination of such foods by pathogenic microorganisms is becoming an increasing threat to public health. Food-borne diseases are believed to cause an estimated annual 76 million illnesses, 5,000 deaths and 325,000 hospitalizations in the US. There is a need for early-warning pathogen detection systems to safe-guard public health. Sensor technologies will be investigated for field-based pathogen screening systems. Actuator development must accompany sensor development to realize the full benefit of sensor technology in automation systems. Vegetable crops, for example, must have precise plant removal actuators for weeding and thinning. Precise chemical application systems must be developed for non-organic systems. In other areas, mechanism design has not been optimized for the physical properties of the plant to which it is being applied. Mechanical tree crop harvesting technology, for example, requires a large energy input. Development in this area could substantially reduce producer costs and energy inputs.
Project investigations will thus include sensing and actuation for in-field automation systems; precision specialty crop application control; and pathogen detection for specialty crop food safety.

<P>The project approach will be to evaluate technologies for their applicability and to extend engineering science required to apply the technology. Expected outcomes include prototype systems; strategies for technology integration into specialty crop value chains; and reduced pesticide application rates and enhanced fertilizer use efficiency.<P> Anticipated benefits include: expanded high value local specialty crop production; safer fresh-produce; and enhanced technical capacity for solving specialty crop problems.

<P>

APPROACH: Investigations associated with this project will be focused in four areas associated with the specialty crop value chain including: sensing for in-field automation systems; reduced application control resolution for precision specialty crop production; mechanical systems for specialty crop automation; and pathogen detection for specialty crop food safety. Optical sensing technologies will be investigated for in-field detection of crop plants for automation of labor intensive tasks. Machine vision weed detection systems will be developed and tested. Multiple features of vegetative plants including color, morphology, and planting geometry will be fused for crop detection. Crop plant detection will be used to differentiate crop plants from weed plants for automated intra-row weeding systems and estimating plant centers for automated thinning systems. Precision fertilizer and pesticide applicator control systems will be investigated to enhance the placement of crop inputs in specialty crop production. Precision control systems will be developed to reduce the management width of applicators, allowing each individual plant row to be uniquely treated. Methods include techniques for up-sampling GPS positions, development of high speed liquid actuators, and application rate compensation for field and machinery dynamics. Reductions in input use will be quantified and the economic and agronomic benefits will be reported. Additional results will specify key performance indicators including control latency and sensitivity of controls to GPS correction source. Automation systems require the development of mechanisms that form the interface between machine and the specialty crop plant. Mechanism design can aid in the development of effective and energy efficient specialty crop automation systems. In parallel with sensing system work, mechanisms required for non-chemical vegetable crop weeding and thinning and tree crop harvesting systems will be investigated. Methods will include determining the spatial resolution of candidate actuator systems, energy requirements, and characterizing specialty crop plant physical properties on actuator performance. Current immunoassay-based pathogen detection requires fluorescence-labeled antibodies to be deployed as signal transducers, which limits their applications in in-field sensing schemes. Using an alternative nano-biosensor approach, the infrared fingerprints of the pathogenic microorganisms and viruses themselves will be used as characteristic signals that identify the targets, eliminating the need for fluorescence-labels. Sensors will be implemented as a microarray-type detection scheme. The mid-infrared signals revealing the identity of the pathogenic targets will be collected using an optical sensor. Signal analysis and post-processing software will be developed to achieve rapid identification of pathogenic targets at high accuracy and sensitivity. The project will be evaluated according to the outputs produced by the project team including journal articles, conference papers and presentations, intellectual property licensed and commercialized, students trained, and industrial partnerships.