Precision weed management in maize

Abstract

Weeds are the largest biological constraint on maize productivity, and management based on uniform herbicide spraying and repeated hand weeding is becoming costly, labour-intensive and environmentally questionable. Precision weed management (PWM) adjusts the place, timing, dose and method of control to the spatial and temporal variability of weeds within a field. This review summarises the weed problem in maize, with emphasis on Indian conditions, and describes the sensing platforms, deep-learning classifiers, spot and patch sprayers, cameraguided hoes and robots that make up current PWM systems. Field evidence shows herbicide savings of at least 50% with patch spraying, and an average of 59% reported for commercial camera-based sprayers in corn, soybean and cotton, although savings depend strongly on weed pressure. Adoption is held back by cost, poor interoperability between sensors and machines, data demands and a gap between research and on-farm practice. Combining PWM with integrated weed management, timing control to the critical period and using low-cost, service-based delivery appears the most realistic route for smallholder maize systems.