Design and Experiment of an Orchard Variable-Rate Spraying System Based on Canopy Volume Sensing

To enhance the adaptability of orchard spraying to canopy spatial variability, a LiDAR-based variable-rate spraying system was developed for trellis orchards. The system used a Livox MID-360 LiDAR to acquire canopy point clouds and generated spray control values based on canopy volume in different nozzle action regions. A canopy partitioning method combining left–right division, travel-direction grids, and sector-based spraying regions was proposed. And the voxel occupancy method was used to estimate canopy volume in each partitioned unit. Based on the calculated canopy volume, target flow rates and PWM control values were determined. Pressure stabilization and FIFO queue-based delay compensation were adopted to improve flow regulation stability and sensing–execution matching. System performance was evaluated through nozzle flow calibration, dynamic targeting tests, and spray deposition tests. Results indicated a strong correlation between the PWM duty cycle and nozzle flow rate under different supply pressures (R2 > 0.98). At forward speeds of 0.3, 0.5, and 0.8 m/s, the mean targeting offsets following delay compensation were 112.5, 118.6, and 126.4 mm, respectively. Spray deposition tests showed that variable-rate spraying produced deposition responses corresponding to canopy volume or density differences. The developed system can realize real-time canopy sensing, sectional spray control, and dynamic spraying execution, providing a technical reference for precision spraying in trellis orchards.