WASHINGTON — A team of Virginia Tech mechanical, electrical, computer engineering and computer science students won a top prize in an annual competition for autonomous operating farm machinery.
The team competed against other university and industry teams at the agBOT Challenge, held each year at Gerrish Farms in Rockville, Indiana, and sponsored by the farm and airBridge LLC.
The team’s entry was an autonomous watermelon harvesting system that can locate, identify, sort and harvest ripe watermelons in a field.
How does it know if the melons are ripe? The same way humans do – by thumping them.
The Virginia Tech team won the top prize in the harvesting category and received $30,000.
Teams were scored on mechanics, software, innovativeness and execution.
The Virginia Tech team consisted of two groups of students: a mechanical engineering team that design and built the harvester, and a special studies team for the autonomous vehicle that towed the harvester, based on a modified ATV donated by Yamaha.
The entry used computer vision and machine learning that allowed the vehicle to locate watermelons, and the harvester determined whether or not they were ripe.
“We placed a microphone under the unit and angled it to the bottom of the melon where it captures the reverberations from the slapper,” said Hongxu “Howard” Guo, a double major in mechanical and electrical engineering.
“If the audio analysis indicates a sound ratio above a particular threshold, what we call the sub-band short-time energy ratio, then the watermelon is ripe and harvested. If not, it is left on the ground.”
Guo says the deep, hollow sound produced by a ripe melon has that distinct frequency range.
Virginia Tech students are already planning for the 2019 agBOT competition, where the categories will be weed and feed and soil testing and treatment.
See the autonomous harvesting system in action.