WSU researchers have been awarded a $1.2 million National Science Foundation grant to improve the efficiency of automated apple harvesting robotics.
Led by Ming Luo, Flaherty assistant professor in the School of Mechanical and Materials Engineering, the interdisciplinary team of researchers will develop a cyber-physical system that aims to integrate human intelligence and machine learning, enhance decision-making and actuation, and improve picking efficiency. The team includes Yan Yan from computer science, Manoj Karkee from biological systems engineering, Matthew Whiting from horticulture and landscape architecture, and graduate students Ryan Dorosh and Christopher Ninatanta.
Agricultural robotic technologies currently are inefficient for orchard operations because of the unpredictable environment there as compared to industrial settings, said Luo.
As part of the 3-year project, the researchers will create a network for human-robot interaction. The network will enable robots to function effectively in farm fields, allowing users to remotely monitor and guide the robots in addressing challenging tasks through teleoperation. Additionally, the robots will learn from the user’s input to enhance their harvesting efficiency.
The researchers are developing a simple, low-cost robot system with a soft body and a fabric arm. The design is intended to delicately harvest apples without causing damage to the trees, while also allowing the robot to respond quickly.
“Safety is a concern because the robot might accidentally collide with the tree,” said Luo. “Rigid robots require time to calculate the optimal path for apple picking with minimal damage, which is time-consuming. Therefore, we designed a soft robot that ensures safety in human-robot interactions and is also gentler on the trees. This reduces the computation time required for obstacle avoidance.”
The robot will eventually automatically help with other operations, like flower thinning, pollination, and pruning as well.
