Advances in local perception for orchard robotics

The development of digital technologies, cost pressures and the increasing need for sustainability have heightened interest in the application of robotics and automation to improve the efficiency of agricultural operations. Sensors for autonomous navigation require precise positioning and perception to keep robots on track, avoid obstacles and correctly identify target objects such as fruit. Sensors capable of providing three-dimensional information, such as stereo cameras, time-of-flight cameras and laser scanners, are emerging as effective solutions. Colour, multi- or hyperspectral and thermal cameras are also widely used for real-time crop sensing. This chapter reviews the advantages and limitations of these sensors for practical farming operations.