Projects
(S3-CAV), ERA-NET ICT-AGRI2 action
Precision farming relies on the ability to accurately locate the crops or leaves with problems and to accurately apply a local remedy without wasting resources or contaminating the environment. This project develops a unifying framework allowing incorporation of many different types of sensor data, methods for creating 3D maps and maximising map accuracy to facilitate operations on a narrow scale with a smaller environment footprint, methods for combining this data to make relevant information easily visible to the farmer, and methods for incorporating real-time sensor data into historical data both to increase precision during applications and to provide fast automated safety responses.
(QUAD-AV) , ERA-NET ICT-AGRI action
Autonomous vehicles are being increasingly adopted in agriculture to improve productivity and efficiency. For an autonomous agricultural vehicle to operate safely, environment perception and interpretation capabilities are fundamental requirements. The present project will focus on the development of sensors and sensor processing methods to provide an autonomous agricultural vehicle with such ambient awareness. The “obstacle detection” problem will be specifically addressed.
The obstacles that might be encountered in the field can be separated into four overall categories that should be detected and handled in different ways: positive obstacles, negative obstacles, moving people/animals/obstacles, and difficult terrain. Further, obstacles may vary greatly from situation to situation, depending on type of crop, fruit, vegetable or plant grown, curvature of landscape as well as other factors. Owing to the variety of situations and problems that may be encountered, no sensor exists that can guarantee reliable results in every case. Any candidate sensor has its strengths and drawbacks. Therefore, a complementary sensor suite should be used to gain the best performance.
The idea of this project is that of using different sensor modalities and multi-algorithm approaches to detect the various kinds of obstacles and to build an obstacle database that can be used for vehicle control. For instance, bearing and distance to the nearest collision can be estimated and used by the path planner to change route or to lower the speed if an obstacle is in close proximity to the vehicle’s planned path.