Automatic Path Planning for Unmanned Ground Vehicle Using UAV Imagery

Abstract

Field machines play an important role in the management of agricultural environments. Increasing use of automated machines in precision agriculture has gained significant attention of farmers and industries to minimize human work load to perform tasks such as land preparation, seeding, fertilizing, plant health monitoring and harvesting. Path planning is considered as a fundamental step for agricultural machines equipped with autonomous navigation system. For mountain vineyards, path planning is a big challenge due to terrain morphology and unstructured vineyards. This paper proposes a workflow to generate an automatic coverage path plan for unmanned ground vehicles (UGVs) using georeferenced imagery taken by an unmanned aerial vehicle (UAV). First, image acquisition is performed over a vineyard to generate an orthomosaic and a digital surface model, which are then used to identify the vine rows and inter-row terrain. This information is then used by the algorithm to generate a path plan for UGV.