Search papers, labs, and topics across Lattice.
This paper introduces AgriNav, an autonomous tractor system that integrates advanced weed detection and LiDAR navigation to enhance precision paddy farming. By employing a custom PyTorch implementation of a weed detection model alongside a lightweight CNN-FPN and a novel inverted-logic discrimination module, the system effectively distinguishes rice from diverse weeds while navigating under challenging canopy conditions. Simulation results reveal high confidence in crop row detection and significant reductions in detection inference regions, addressing critical challenges in site-specific weed management and reducing herbicide use.
AgriNav achieves over 90% confidence in crop row detection even during GNSS outages, revolutionizing precision agriculture with minimal herbicide reliance.
Site-specific weed management in paddy farming offers substantial reductions in herbicide use over conventional broadcast spraying, but field deployment has been limited by three persistent challenges: robust crop-row navigation under canopy where GNSS degrades, real-time visual discrimination between rice and morphologically diverse weeds, and the asymmetric cost of misclassifying rice as weed, which is irreversible. This paper presents AgriNav, an integrated autonomous tractor system built around four ROS-coupled modules: a custom PyTorch reimplementation of WeedDet for rice detection, a parallel lightweight 1.68M-parameter CNN-FPN variant with asymmetric class weighting, an inverted-logic discrimination module that protects the rice class through a hardcoded confidence-gate veto, and a 6-state constant-velocity-turn-rate Extended Kalman Filter fusing GNSS, IMU, and wheel odometry with three-level outage bridging. Our primary system-level contribution is a four-mechanism LiDAR-camera fusion bridge that uses the navigation LiDAR for region-of-interest constraint, world-coordinate projection, ground-plane filtering, and bidirectional confidence fusion at zero additional hardware cost. Simulation experiments demonstrate continuous position tracking through a 20-second GNSS outage, crop row detection confidence above 0.9 throughout operation, and rice-detection confidences from 0.32 to 0.95 across paddy, aerial, and post-flood imagery. The LiDAR ROI constraint reduces detection inference region by an estimated 30 to 50 percent.
