A Method for Canopy Thickness Detection Using Frequency-Modulated Continuous Wave Radar

Accurate canopy thickness measurement is vital for advancing smart orchard sprayers and reducing pesticide use. We present three FMCW radar-based algorithms to estimate canopy thickness and assess their accuracy through experiments. Laboratory tests on simulated canopies showed a strong correlation between estimated and actual thickness. However, stability and relative error were significantly affected by leaf area density (LAD) and detection distance. When distance was 40.0–110.0 cm, LAD 1.5–4.2 m2/m3, and actual thickness 30.0–120.0 cm, the coefficient of variation and relative error remained within ±14%. Preliminary outdoor trials on three isolated tree canopies served strictly as an early proof-of-concept, not a full orchard evaluation. The Db algorithm performed poorly, with relative error up to 74.9%, whereas the Da and Dc methods proved robust, with maximum relative errors of 11.93% and −15.19%, respectively. These findings highlight that, with further refinement, Da and Dc algorithms are highly promising for precision variable-rate spraying systems in orchards.