Autonomous Navigation Support from Real-Time Visual Mapping
|Funding from Vinnova||SEK 5 718 000|
|Project duration||September 2019 - December 2022|
|Venture||National Aeronautical Research Program 7|
|Call||Research project in aviation technology - spring 2019|
Purpose and goal
Autonomous UAV navigation is disturbance sensitive, for instance due to intentional electronic attacks. An alternative to GNSS is to compare images acquired by the UAV with ground reference data. The project tackles challenges like: choice of support data (aerial / satellite images, terrain models, 3D point clouds) UAV camera requirements how to calculate position/orientation accurately and quickly enough. The project will provide a basis for development of more reliable autonomous navigation, both in terms of improved accuracy and as “backup” when GNSS is unavailable.
Expected results and effects
The project will provide a basis for the development of visual mapping systems, which will be useful both as "backup" in cases where GNSS fails and to improve the accuracy of position and orientation estimates for aircraft. In a longer perspective, the project is expected to: Facilitate the UAV use of i.a. Swedish blue light authorities; Strengthen Sweden´s technical capacity, Contribute to strengthened Swedish research, development and innovation capacity; and Contribute to reduced risk of accidents for UAVs.
Planned approach and implementation
Daniel Sabel is employed by Spacemetric AB, which produces the image management software Keystone. Daniel is granted temporary leave from Spacemetric to run the research project. The research will be conducted at KTH´s School of Electrical Engineering and Computer Science, EECS. Daniel will have a supervisor from EECS, Dr Atsuto Maki, and a supervisor from Spacemetric, Dr Torbjörn Westin. Daniel will work closely with Spacemetric´s development team to perform experiments and gradually operationalise his research results into Spacemetric´s image management software Keystone.