Hybrid topological location and mapping for autonomous agricultural robots

Overview & Synopsis

This paper considers the problem of autonomous navigation in an agricultural field. It proposes a localization and mapping framework based on semantic place classification and key location estimation, which together build a hybrid topological map. The approach has been assessed through off-line data, recorded in real conditions, on diverse fields, at different seasons. The results have shown the interest of the approach, which allows: (i) simple and easy-to-update map to be obtained; (ii) the use of artificial landmarks to be avoided, thus (iii) autonomy of agricultural robots to be improved.