Reliable and Robust Multi Sensor State Estimation and Prediction (Navigation without GPS)



GPS and an INS are the common tools to solve the navigation problem for outdoor environments. When flying close to objects, relying on GPS only for position estimates is not reliable. The goal of the research is the development of a system that is robust against errors or loss of the GPS signal. The state estimation of the UAV will be aided with different on board sensors. For aerial manipulation and landing on mobile platforms it is also important to estimate the state of external objects. Special attention is paid to autonomous landing assisted by multi-sensor data processing including GPS, INS, vision, radar and laser scanner. The flying robots research group will use different sensors like mono/stereo vision and laser scanners to solve these problems.


URL for this article
http://www.dlr.de/rm/en/desktopdefault.aspx/tabid-7660/13033_read-32820/