Jason Ward

Assistant Professor

  • 919-515-8985
  • D S Weaver Labs NA

Selected Current Projects

 

Quantifying Crop Lodging Damage with UAS Imagery

In North Carolina, harvest season is hurricane season.  The possibility for weather-related crop damage is a real danger for producers.  After a severe weather event, crop health must be quickly assessed for the amount and severity of damage. Producers, insurance providers, and emergency managers need timely and accurate information to understand the scope of crop damage.  The objectives of this study are to compare specific measurement technologies to find the best suite of tools to quantify the severity of damage and area of impact.  Recently, multiple different deep learning object detectors were compared for their ability to automatically detect simulated lodged corn.  Traditional vegetative indices and machine learning techniques are being explored to identify the most efficient tool.  The measured impacts will be compared to the physiological impacts on the crop.  Finally, a best method of federal or commercial reporting and securing that data to current standards will be developed.

 

Cotton Quality Mapping

Modern cotton harvesting equipment has the capability to weigh, estimate moisture content of, and discretely identify round modules when completed.  These data along with bale location should allow tracking of gin fiber quality data all the way back to a field area via the permanent bale indicator (PBI) assigned at the gin and the radio-frequency Harvest ID (HID) tag assigned at harvest.  The first year of this project confirmed the proof-of-concept and the industry was receptive to the ability to translate fiber quality to a field basis. So far, this project has resulted the development of one of the first whole-field cotton fiber map in the US.

 

Automating the Cotton Replant Decision with UAS Imagery

Cotton replant decisions are often made based on a quick, ground-level visual inspection of fields. UAV plant counts are currently available but an expert decision tool is needed to help quickly process the resulting data into a replant recommendation.

 

Robot Systems to Support Pasture Animal Performance

Pasture animals are less comfortable and perform poorly when they are heat stressed. Heat stress can be mitigated with fixed or portable shade. Fixed shade causes animals to congregate which leads to negative environmental impacts and poor animal performance caused by muddy conditions and accumulated waste. Mobile shades are prone to damage, especially from wind, and can be difficult to move or operate. An autonomous robotic animal support platform is being developed which will autonomously move in a pasture which reduces localized environmental impacts, distributes waste, and protects forage. Environmental sensors will monitor the environment and respond to support the animal. Sensors will be used to detect how animals use the structure and if they are less stressed.

 

Improving Harvest Efficiency with Machine Data

Peanut harvesting is a complex series of field activities requiring multiple passes across the field with multiple implements.  There is limited time to get the job done before harvest will suffer.  Optimizing harvest efficiency, machine sizing, and machine settings are essential to getting in and out of the field as quickly as possible.  Modern farm equipment can generate substantial data during day-to-day operations.  The tools now exist to tap into this data and leverage that information to ensure that harvest operations are running as smooth as possible.  The objectives of this project are to collect that machine data and use it to measure harvest field operation efficiency and to understand how to best manage the equipment to save time and money.  The same data can be used to understand how to mitigate harvest risk within the allowable working days by properly sizing machines to the farm.

Education

B.S. 2003

Biosystems and Agricultural Engineering

University of Kentucky

M.S. 2004

Biosystems and Agricultural Engineering

University of Kentucky

Ph.D. 2012

Biological Engineering

Mississippi State University

Research Description

The Advanced Ag Lab works at the intersection of technology and farming. We leverage an understanding of agricultural methods and practices combined with technology, sensors, equipment, robotics, and data management to drive real-world decision making. We conduct applied research to identify existing or create new data in the modern precision agriculture environment and work closely with commodity experts to move from data collection to actionable insight. We develop and deliver innovative Extension programming that make the technology already on the farm more accessible and valuable. We defines precision agriculture as a methodology of data-driven decision making to improve output, manage impact, or reduce waste. This approach allows us to work across commodities and market sectors to create useful tools which allow best management at higher resolutions no matter the crop, animal, or equipment involved.

Grants

UAV Methods for Cotton Replant Decisions and Official Variety Trials
Cotton, Inc.(1/01/19 - 12/31/19)
Using UAV Imagery to Detect Crop Damage after Severe Weather Events
Corn Growers Association of NC, Inc.(4/15/19 - 4/14/20)
Cotton Quality Mapping and Harvest Logistics with RFID Tracking
Cotton, Inc.(1/01/18 - 12/31/19)
Aerial Imagery and Sensors for Peanut Management
NC Peanut Growers Association, Inc.(1/01/18 - 12/31/18)
Precision Agriculture Technologies for Cotton Production in North Carolina
Cotton, Inc.(1/01/12 - 12/31/19)