Vision-GPS Fusion for Guidance of an Autonomous Vehicle in Row Crops

Thomas Bak

Abstract:	This paper presents a real-time localization system for an autonomous vehicle passing through 0.25 m wide crop rows at 6 km/h. Localization is achieved by fusion of measurements from a row guidance sensor and a GPS receiver. Conventional agricultural practice applies inputs such as herbicide at a constant rate ignoring the spatial variability in weed, soil, and crop. Sensing with a guided vehicle allow cost effective mapping of field variability and inputs may be adjusted accordingly. Essential to such a vehicle is real-time localization. GPS allow precise absolute sensing but it is not practical to guide the vehicle relative to the crop rows on an absolute coordinate. A row guidance sensor is therefore included to sense the position relative to the rows. The vehicle path in the field is re-planned online in order to allow for crop row irregularities sensed by the row sensor. The path generation is thus controlled by location relative to the local field while the actual path execution is carried out in absolute GPS coordinates. The solution is a system that fuse data from a relative and an absolute measurement system while ensuring accurate row operation at high work rates.
Published in:	Proceedings of the 14th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GPS 2001) September 11 - 14, 2001 Salt Palace Convention Center Salt Lake City, UT
Pages:	423 - 451
Cite this article:	Bak, Thomas, "Vision-GPS Fusion for Guidance of an Autonomous Vehicle in Row Crops," Proceedings of the 14th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GPS 2001), Salt Lake City, UT, September 2001, pp. 423-451.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In