Validation of High-Definition Maps Using Agricultural and Autonomous Vehicles

农业数字技术的进步使高分辨率作物性能数据得以捕获，为改进决策提供了深刻的见解。然而，现有的农业技术解决方案在很大程度上是孤立开发的，没有与其他解决方案集成，导致最终用户感到沮丧，减缓了采用速度。该项目将利用高精度的RTK设备绘制葡萄园内的所有特征，包括：入口、道路、排水沟、灌溉歧管、传感器、柱子和葡萄藤；并将这些数据转换成开源地图。这些地图的主要用户是机器人和无人机供应商，他们依靠详细的地理空间信息来安全可靠地操作他们的平台。直接使用单一的统一地图将消除昂贵的单独调查。这将是新兴自动驾驶技术的关键推动因素，对它们的普及至关重要。虽然该项目侧重于葡萄园，但这同样适用于其他行作物（灌木水果、草莓、啤酒花、果园、高价值蔬菜）。耕地作物本质上主要是二维的，因此通过shapefile共享相对简单的技术进行映射，与之不同的是，行内作物目前还没有描述性的映射方法。这一限制意味着种植者无法轻松部署尖端农业技术来监测作物的变化和问题。该项目将研究核心制图技术和技术合作，但也有扩展的空间，以研究其他方法，在更多的技术平台上验证导航地图，以获取更有洞察力的害虫、疾病或杂草、产量和树冠结构数据。

Digital technology advances for agriculture have enabled high-resolution crop performance data capture, offering great insight for improved decision-making. However, existing Agri-tech solutions have largely been developed in isolation, do not integrate with other solutions, causing frustration to end-users, slowing adoption.The project will utilise highly accurate RTK equipment to map all features within the vineyard including: entrances, roadways, drains, irrigation manifolds, sensors, posts, and vines; and convert this data into open-source maps. Key users of these maps are robotics and UAV suppliers, who rely upon detailed geospatial information to safely and reliably operate their platforms. Direct access to a single unified map would eliminate costly separate surveys. This will be a key enabler to emerging autonomous technologies and will be critical to their uptake. Although the project focuses on vineyards, this will be equally applicable in other row crops (bush fruit, strawberries, hops, orchards, high value vegetables).Unlike arable cropping which is largely 2D in nature and therefore mapped using relatively simple techniques shared via shapefiles, there are no current descriptive mapping methods for row crops. This limitation means growers are unable to easily deploy cutting-edge agri-tech to monitor crop variation and issues. The project will investigate core mapping techniques and technology collaboration, but there is scope to extend it to investigate additional methods, validating the navigational maps across more technology platforms to capture more insightful data of pests, disease or weeds, yield, and canopy structure.