Drones Over the Paddy: How UAVs Are Rewriting DSR

Drones, or unmanned aerial vehicles (UAVs), equipped with advanced sensors, cameras and GPS technology, are transforming DSR in several ways — from seed sowing to fertilizer spreading and pesticide spraying — by addressing the key challenges of uneven seeding, labour-intensive operations and inefficient weed management that have held mechanized DSR back. Traditional broadcasting methods often result in uneven seed distribution and damage to land topography, leading to suboptimal crop establishment and lower yields. Drones distribute seeds evenly across a field in a fraction of the time, ensuring uniform germination and minimizing seed wastage — outperforming manual methods in both speed and accuracy. In field trials, drone seeding at a rate of 40 kg/ha significantly outperformed manual broadcasting, delivering superior growth and 13% higher overall productivity compared to the traditional manual approach at 60 kg/ha.

Current drone seeding, however, still lacks the precision of true row placement — developing mechanisms for row-specific seed placement via drones could further reduce seed rates and optimize crop geometry, an active area of ongoing research. That gap matters because row placement, not just even distribution, is what the mechanized seeders covered elsewhere in this series use to enable better weed management and nutrient placement — closing it would let drone seeding capture those same benefits on top of the speed advantage it already has.

Weed management is a critical DSR challenge, and traditional manual knapsack spraying is labour-intensive, slow, and exposes workers to toxic chemicals. Drones offer a transformative alternative, outperforming traditional mist-blower methods by ensuring timely herbicide application while significantly reducing labour intensity and physical hardship for farmers. Research has established that bispyribac-sodium 10% SC applied at 35 g a.i./ha during the 2-to-3-leaf weed stage, using a UAV spray system with a spray volume of just 30 L/ha, delivers effective weed control — roughly ten times less carrier volume than manual methods.

The economics back this up. Drone application of pretilachlor followed by bispyribac-sodium recorded cost ratios of 2.27 and 2.09, energy-use efficiency of 10.86 and 9.55 MJ, and energy productivity of 0.81 and 0.71 kg/MJ across trials — all significantly reducing weed density and dry weight compared with unweeded plots, and resulting in higher grain yields, net returns and benefit-cost ratios. A separate trial of the same drone-applied sequence recorded a grain yield of 5,286 kg/ha, a net return of ₹51,631/ha, a benefit-cost ratio of 2.17, and an energy productivity of 0.30 kg/MJ under wet direct-seeded rice.

Artificial intelligence is sharpening this further. An optimized YOLOv8n-DT model enables drones to use precise 'prescription maps' that target actual weed density, increasing identification accuracy while saving 15.28% of herbicide compared to uniform spraying. A separate detection method (RPAAS), with up to 95% accuracy, achieved a 45% reduction in herbicide use while minimizing crop injury by treating only where weeds actually are, rather than spraying the whole field uniformly.

One of the more counterintuitive findings is that dose reduction, done correctly, can outperform the full recommended dose. Research applying a herbicide combination (triafamone 20% + ethoxysulfuron 10%) via drone at 70% of the recommended dose (50.6 g a.i./ha) not only cut costs by 30%, but also improved crop-weed dynamics, yield, and weed control efficiency (reaching 95%). At 100% of the recommended dose (67.5 g a.i./ha), by contrast, the herbicide caused phytotoxicity — hindering initial crop growth and resulting in higher weed competition and lower weed control efficiency (71–73%). The lesson is that precision application at a calibrated dose can beat brute-force full-strength spraying, provided the standard operating protocol for that specific drone-herbicide combination has actually been established.

The direct-seeding trials tell a similar story about calibration mattering more than raw quantity: drone seeding at 40 kg/ha significantly outperformed manual broadcasting at 60 kg/ha, meaning the drone trial used a third less seed while still achieving better crop growth and 13% higher productivity — precision in placement, not a higher seeding rate, was what delivered the gain.

The economic case for drones goes beyond any single trial: tasks such as seeding and spraying, which would traditionally take days or even weeks with manual labour, can now be completed in a matter of hours, which is especially valuable in regions facing rural labour shortages. Although the initial investment in drone technology can be high, the long-term savings on labour and inputs, combined with the yield gains covered above, make a strong economic case over the equipment's working life.

The barriers to wider adoption are nonetheless real: high initial investment and a lack of technical expertise, particularly in developing regions, alongside the absence of standard operating protocols (SOPs) for drone operations, which limits their efficiency and effectiveness. Recent studies have begun establishing SOPs for specific herbicide-drone combinations, spray volumes and application heights, but this work remains patchy across the many herbicide and crop-stage combinations Indian farmers actually use. Where SOPs do exist, they're specific and tested: bispyribac-sodium at 35 g a.i./ha in 30 L/ha of carrier at the 2–3 leaf weed stage is one; triafamone plus ethoxysulfuron at 70% of the recommended dose is another. Outside these documented combinations, a farmer or drone operator is largely extrapolating from the nearest tested protocol rather than following one written for their exact herbicide and crop stage.

As drone technology increasingly incorporates AI and machine learning for real-time monitoring and decision-making, and as government and private investment in training and infrastructure grows, the expectation in the literature is that these SOP gaps will close — but until they do, drone DSR's precision remains a double-edged tool that rewards correct protocols and punishes guesswork.