Although the use of pesticides in crop production has been brought down considerably over the last years there is still a need for further reductions. However, the methods used until now - mainly improved management practices and more efficient equipment - are not considered to have sufficient potential for reaching this goal. It is therefore time to look for new solutions.
One promising possibility is to utilise the new information technologies that already are being applied in many mass produced devices to deliver higher performances at still lower costs. These technologies can also be used to develop new field machinery systems that are operating autonomously and able to perform highly accurate down to the spatial resolution of individual plants. This could for example be a weeding operation by using a tillage tool, an energy dose of laser light or a single small droplet of herbicide onto individual weed plants. Such systems could not only dramatically reduce or substitute the amount of chemicals applied; they will also remove the risk of herbicide leaching known from present systems. An additional perspective is that individual plant identification may be used to control only the most harmful weed species while leaving the remaining population for the benefit of biodiversity and soil protection.
A major problem of implementing such new systems has up to now been the costs. But if these new technologies are combined with autonomous vehicles or robots the labour expenses can be reduced considerable. Furthermore, the potential daily working hours can be doubled or tripled. This means that labour costs and basic machinery costs will be reduced, which somewhat compensates for the higher costs of automation. A consequence of this is also, that the size and weight of machinery will drop significantly, which leads to another advantage of reduced soil compaction damages. The economic break even for this equipment is therefore not unrealistic in the near future.
To order to contribute to this development the Department of Agricultural Sciences at KVL is working on the following projects:
1. Autonomous Mechanisation System for Seeding and Weeding
(Short description)
The objective of the project was to establish a fully operational autonomous tractor for crop production. A predefined route plan navigates the tractor automatically across the field and also a predefined task plan controls the implement in order to execute an application task. The system in general allows a very accurate crop plant establishment even in predefined spatial patterns and a weeding operation including within row control. The operation is carried out completely unmanned and the working quality is better or at least comparable with real-time operator control.
Partners: Hans W. Griepentrog, Michael Nørremark (KVL, Environment, Resources and Technology), Nils A. Andersen, Ole Ravn (Oersted-DTU, Automation), Olaf Heinemann (Hako, Marketing), Till Meinel (Kverneland, Accord), Svend Christensen (DIAS, Department of Agricultural Engineering)
Timeframe: 2001-2006
Financing: Danish Research Agency/The Danish Ministry of Food, Agriculture and Fisheries
Responsible: Hans W. Griepentrog
2. Autonomous Weeder for Christmas Trees and other Tree Crops
In the project a driverless machine is being developed for weed control in Christmas trees. The machine drives between the tree rows and cuts the weeds below the branches with a rotor cutter. The navigation is based on GPS and information about the tree positions. Tree positions are used to define a route plan for the machine and for moving the cutter laterally. The developed system appeared to be well suited for the purpose and is considered technically and economically realistic. Furthermore the environmental perspectives in general are very promising.
Partners: Henrik Have, Jon Nielsen (KVL, Environment, Resources and Technology), Frans Theilby (KVL, Centre of Forestry, Landscape and Planning), Lillie Andersen (DIAS, Horticulture), Erik Anderson (Christmas tree grower)
Timeframe: 2003-2005
Financing: Danish Ministry of Environment
Responsible: Henrik Have