LIDAR-based Navigation Rover for Fields with Smart Pest Sprayer using Machine Vision

Abstract

Agricultural robots that have been unmanned ground vehicles are a replacement for the tractors that are now used in fields. Future fleets of autonomous agricultural robots may carry out cultivation operations like sowing, fertilizing, pest spraying, and harvesting. The intention is to demonstrate that a genuine agricultural robot is competitive with existing technology and might even take a clear edge. Agriculture production’s most important step is pest bug management. Pest insects cause almost all agricultural plants to become harmed, weak, or even die. This leads to lower quality, lower yields, and unsellable plants or plant products due to damage. Insects continue to harm products that have been prepared or kept even after they have been harvested. Regardless of its exact design, a professional agricultural robot will be a complex and expensive machine. A Retractable sprayer arm will then be used in this study’s implementation to spray the insecticide underneath the crop’s leaves. Using LiDAR (Light Detection and Ranging) data, the research presented in this paper aims to establish a broad and reliable approach enabling autonomous robot navigation within a crop field. The autonomous mobile navigation should be as reliable as feasible. The method that was presented is based on the extraction of lines from 2D point clouds. In a greenhouse or an outdoor setting, such as a farm with open fields, the retractable sprayer boom can also be operated in a retractable manner. By deploying the autonomous sprayer, a successful pesticide management system is anticipated.