Method for Correcting Unreliable Readings of a GNSS Receiver When Using a Cost-Efficient Autopilot
摘要
This paper addresses a critical vulnerability in low-cost UAV autopilot systems in case of GNSS data unavailability, which can result from external interference or intentional spoofing. In the paper, we use a method that compensates for erroneous GNSS readings by substituting them with corrected signals derived from an external Inertial Navigation System (INS) combined with high-precision barometric altimeter measurements. The proposed approach integrates the inertial unit’s angular velocity and acceleration data using an Extended Kalman Filter (EKF3) to generate accurate flight parameters that mimic the standard GNSS output, thereby ensuring reliable navigation. To validate this method, a series of field experiments were conducted using a VTOL UAV platform equipped with a HEX Cube Orange+ flight controller. During the test, an interference generator simulated false GNSS signals, causing significant deviations in altitude and position readings from the built-in GNSS receiver. In contrast, the INS-based system maintained a stable and accurate flight profile, as confirmed by telemetry data processed in MATLAB. Comparative analyses of flight altitude, groundspeed, and positional data clearly demonstrate the proposed method’s effectiveness in mitigating the impact of GNSS jamming. This work contributes to enhancing the safety and reliability of UAV operations in environments prone to GNSS interference.