Abstract

The aim of the project was to find a method for automatic positioning a row-weeder in cereal crops. This would increase the precision of the weeding and lower the demands on the driver, compared to a manually guided row-weeder. The report deals only with the problems connected to the automatic control-system and to the question of how to detect the rows to be followed. With the aid of the Swedish and foreign agricultural data-bases of the Ultuna library, a search for literature, dealing with the problem of automatic positioning of a row-weeder was made. The literature found was used in a study which is summarized in the report. The result of the study shows that no adequate method exist with regard to precision, demands on working environment and cost. While searching for a suitable method a new idea ocurred; To mount the row-weeder firmly to a vehicle and to guide the whole vehicle automatically along the rows. Instead of only positioning the implement. To test this idea a model was derived. With the model different positions of a transducer and of the implement could be tested together with different regulators. Thus the system with the highest possible precision could be found. The result of the simulation tests showed that it ought to be possible to guide a row-weeder, mounted on a vehicle, up to a speed of 8 km/h. An experimental machine was equipped in order to test the results of the simulation in practice. The experimental machine was equipped with an electrical motor, for the controlling of the steering, and a microprocessor-based regulator. Two prototype transducers were designed and tested. Both were ment to run between two rows and to mechanically detect the position of the vehicle in relation to the rows. Two types of experiments were carried out. Firstly, the machine was tested on a course where planks (5*10 cm) served as substitutes for cereal rows. The machine was exposed to different kinds of disturbances so that the responses could be studied, while using different regulators and different positions of the transducer. Secondly the machine was tested in spring barley so that the practically achievable levels of precision could be found. When the tests were done (at the end of June) the crop was high enough to permit good possibilities for mechanical detection. There were though, many perennial weeds that influenced the transducer negatively. This did not effect the good performance of the control system, however. The result of the tests in spring barley showed that the machine, at a speed of 2 km/h, had a maximum deviation of 2.7 cm. The standard deviation was 1 cm. No tests could be done at higher speeds as the actuator set a limit to the performance of the system. The conclusions drawn in this paper are: A system consisting of an implement firmly attached to a vehicle, can be automatically guided accurately enough, and that more work must be put into improving the transducers. Some proposals for improvements of the transducers are made. In the discussion part of the paper the author gives an account of how the project can be followed up in the future.

Keywords

radhacka

Published in

Rapport - Sveriges lantbruksuniversitet, Institutionen för lantbruksteknik
1990,
ISBN: 91-576-4375-X
Publisher: Institutionen för lantbrksteknik, Sveriges lantbruksuniversitet

Permanent link to this page (URI)

https://res.slu.se/id/publ/125615