Interest in the application of unmanned systems to Antarctic exploration has increased because those can replace perilous and challenging tasks for human. Precise heading determination is one of the required navigation capability of unmanned systems, but it is hardly achievable in Polar Regions. The reason is that a magnetic compass, which uses horizontal geomagnetic force to create orientation measurement, is inaccurate and deviates near the magnetic pole where vertical geomagnetic force is dominant. Unexpectedly large heading error may cause an unmanned vehicle to lose its control and finally results in an unsuccessful mission. In this regard, we present a method that uses Global Navigation Satellite System (GNSS)-based heading measurement instead of magnetic heading measurement to improve heading determination performance. From the previous work (Kim et al. 2019), the use of local-area differential GNSS (DGNSS) has been proposed to cope with the limited GNSS performance due to low satellite visibility in high latitudes. By exploiting this architecture, GNSS-based heading measurement is calculated as a course angle of DGNSS position. In this study, the feasibility of the proposed method is analyzed through simulations, and several flight tests are conducted to assess the performance of the GNSS-based heading estimation while the magnetic compass is turned off.

Keywords: Antarctic exploration, GNSS-based heading measurement, heading improvement