GTRI Agricultural Robotics
GOALS: Robotics has the potential to revolutionize inspection and harvesting tasks on commodity farms (corn, soybeans, and wheat) and high-value specialty farms (fruits and nuts). This is true from the standpoints of labor, economics, early disease detection, and crop assessment. Specialty farms stand to benefit the most because the tending and harvesting of their crops still require significant amounts of manual labor. By contrast, most aspects of commodity farms have already been automated. The Agricultural Robotics VIP Team will have as its goal leveraging recent advances in robotics to improve agricultural processes. Its emphasis will be on fielding functional devices/systems.
METHODS & TECHNOLOGIES: Mechanical design, physics, material properties, computer simulation, CAD, 3D printing, sensors, programming, rapid prototyping, experiment design, testing & evaluation.
RESEARCH/DESIGN ISSUES: Our planet’s human population is expected to exceed 9 billion by 2050. To keep pace with population growth, worldwide agricultural production will have to dramatically improve yields. A crucial enabler to improving yields is to better understand the genetic association between plant genotypes (gene profile) and plant phenotypes (the resulting physical expression of those genes). This involves identifying genes of importance that permit prediction of crop performance. Recent advances in DNA sequencing are reducing the cost and speed of plant genotyping. By contrast, efficient and precise methods for plant phenotyping have lagged further behind and have become a bottleneck for genetic association and yield improvement. Plant phenotyping methods typically involve intensive measurement of individual plants. Traditionally, this is accomplished by a grower manually scouting a field.
This project has developed a novel concept of using two-arm brachiating robots traversing a simple overhead parallel cable network installed above a soybean research field in Athens, GA to perform high throughput field-based phenotyping of traits that signify drought tolerance. “Brachiation” is a form of swinging used by apes to efficiently locomote within an unstructured, fairly low-infrastructure environment. Our developed robot (nicknamed “Tarzan”) is equipped with a sensor payload at the joint between its two arms that is used to take measurements of the soybean plants below. When eventually deployed as a multi-agent system within a crop field’s cable network, a team of brachiating robots will effectively comprise a re-organizing, re-deployable mobile sensor network for field-based phenotyping.
The Agricultural Robotics VIP Team will be tasked with advancing the hardware embodiment of Tarzan, beginning with designing, fabricating, and testing a more robust multi-purpose gripper and also adding a capability to visually detect the cable during brachiation.
MEETING TIME: Thurs, 3:00-3:50
ADVISOR: Ai-Ping Hu (GTRI)
PARTNERS & SPONSORS: GTRI, Harris Corporation
MAJORS, PREPARATION AND INTERESTS: ME, CS, ECE, MSE, Physics, Biology, BME, Industrial Design
CONTACT: Dr. Ai-Ping Hu, (404) 407-8815, firstname.lastname@example.org