View Abstract Sign in for premium content
The demand and development of UAV operations is increasing rapidly and with it the need for safe navigation for UAVs in the urban environment. The primary means of UAV positioning is GNSS, which suffers from locally introduced error sources such as multipath and non-line-of-sight signals. A promising method to decrease the effects of signal reflections is by applying imaged based masking. By comparing the signals received and the local surroundings with image detection methods, signals that are determined to be non-line-of-sight can be filtered out. Within this work, the masking algorithm is put to the test, demonstrating the positioning integrity for a UAV flight measurement campaign in the urban environment. The performance of the proposed methodology is evaluated by comparing the position error and integrity bounding with the same metrics obtained without masking applied. In this work it has been shown that the use of image based masking can aid in filtering satellite signals which suffer from non-line-of-sight. Removing these satellites from the position equation results in a higher accuracy of the positioning, and as a consequence better error over bounding of the protection levels. It is observed that other methods such as increasing the C/N0 or signal strength threshold do not necessarily provide the same results as the proposed image based masking can. This shows that a simple and low cost implementation of camera based masking can help to improve the accuracy and integrity of positioning solution.