Learning to Prune Branches in Modern Tree-Fruit Orchards
This addresses the labor-intensive task of pruning for orchard farmers, but it is incremental as it shows limited success compared to an ideal planner.
The researchers tackled the problem of automating dormant tree pruning in modern fruit orchards by developing a closed-loop visuomotor controller for robotic pruning, achieving a 30% success rate in real-world deployment, which is about half the performance of an oracle planner.
Dormant tree pruning is labor-intensive but essential to maintaining modern highly-productive fruit orchards. In this work we present a closed-loop visuomotor controller for robotic pruning. The controller guides the cutter through a cluttered tree environment to reach a specified cut point and ensures the cutters are perpendicular to the branch. We train the controller using a novel orchard simulation that captures the geometric distribution of branches in a target apple orchard configuration. Unlike traditional methods requiring full 3D reconstruction, our controller uses just optical flow images from a wrist-mounted camera. We deploy our learned policy in simulation and the real-world for an example V-Trellis envy tree with zero-shot transfer, achieving a 30% success rate -- approximately half the performance of an oracle planner.