In addition to position sensing, GPS receivers can be leveraged for orientation sensing too. We place multiple GPS receivers on drones and translate their relative positions into orientation. Such an orthogonal mode of orientation sensing provides failsafe under Inertial sensor failures – a primary cause of drone crashes today. This paper integrates GLONASS satellite measurements with GPS for enhancing the orientation accuracy. Accurate estimate of orientation depends upon high precision relative positioning of the GPS receivers. While GPS carrier phases provide high precision ranging data, the phases are noisy and wrap after every wavelength which introduces ambiguity. Moreover, GPS signals experience poor SNR and loss of satellite locks under aggressive flights. This can severely limit both the accuracy and the amount of carrier phase data available. Fortunately, integrating the ubiquitously available Russian GLONASS satellites with GPS can double the amount of observations and substantially improve the robustness of orientation estimates. However, the fusion is non-trivial because of the operational difference between FDMA based GLONASS and CDMA based GPS. This paper proposes a temporal differencing scheme for fusion of GLONASS and GPS measurements, through a system called SafetyNet. Results from 11 sessions of 5–7 min flights report median orientation accuracies of 2◦ even under overcast weather conditions.