An optimized field coverage planning approach for navigation of agricultural robots in fields involving obstacle areas

Technological advances combined with the demand of cost efficiency and environmental considerations lead farmers to review their practices towards the adoption of new managerial approaches including enhanced automation. The application of field robots is one of the most promising advances among automation technologies. Since the primary goal of an agricultural vehicle is the complete coverage of the cropped area within the field, an essential prerequisite is the capability of the mobile unit to cover the whole field area autonomously. In this paper, the main objective is to develop an approach for coverage planning for agricultural operations involving the presence of obstacle areas within the field area. The developed approach involves a series of stages including the generation of field-work tracks in the field polygon, the clustering of the tracks into blocks taking into account the in-field obstacle areas, the headland paths generation for the field and each obstacle area, the implementation of a genetic algorithm to optimize the sequence that the field robot vehicle will follow to visit the blocks, and an algorithmically generation of the task sequences derived from the farmer practices. This approach has proven that it is possible to capture the practices of farmers and embed these practices in an algorithmic description providing a complete field area coverage plan in a form prepared for execution by the navigation system of a field robot.