Skip to main content
ARS Home » Midwest Area » East Lansing, Michigan » Sugarbeet and Bean Research » Research » Research Project #440962

Research Project: Deep Learning-based 3D Fruit-Tree Perception Through Efficient Multi-Sensor Fusion for Robotic Harvesting of Apples

Location: Sugarbeet and Bean Research

Project Number: 5050-43640-003-007-S
Project Type: Non-Assistance Cooperative Agreement

Start Date: Sep 15, 2021
End Date: Dec 14, 2022

Objective:
The main objective of this project is to develop a unified, efficient perception system to provide 3D fruit and tree branch information to support real-time planning and controls for robotic apple harvesting.

Approach:
The following approaches will be used, built on the existing ARS single-sensor perception system: 1) Develop a multi-sensor fusion system for accurate, robust perception of leaves, tree trunk and branches to substantially enhance the detection and localization of apple fruit; and 2) Develop algorithms for efficient dynamic moving window perception to support real-time robotic operation on a resource-limited platform. Different pre-fusion and post-fusion strategies will be considered to systematically combine these sensors.