New UAV Velocity Estimation Using Array of Hall Effect Sensors for Indoor Navigation

Abstract

Small Unmanned Aerial Vehicles (UAVs) cost, size, weight, and versatility make it adequate for many autonomous applications in our daily life (e.g. mapping, surveillance, inspection). The key point for a successful autonomous navigation system is the integration between Inertial Navigation Unit (INS) and Global Navigation Satellite Systems (GNSS). Without GNSS signals the INS accumulated errors cause massive drift in the navigation solution, especially if the onboard INS is a low-cost Micro Electro Mechanical Systems (MEMS) based. In order to substitute the GNSS during unavailability periods, other sensors such as cameras, Light Detection and Ranging (LIDAR) and Radio Detection and Ranging (RADAR) must be used. Regardless of the utilized sensor to substitute the GNSS system, its characteristics must be adequate for such small UAVs (regarding size, cost, weight, and power consumption). From this perspective this paper proposes an unconventional approach to estimate quadcopter drone forward velocity. The proposed approach is based on unusual utilization of an array of hall effect sensors “Air-Odo”. Two Hall effect sensors were utilized to accommodate the velocities of such small drones in indoor environments. To stand on the ability of such approach to enhance the navigation solution during complete GNSS signals outage, an experimental flight was carried out in an indoor field. The proposed approach was able to estimate the velocity of the drone with Root Mean Square Error (RMSE) of 0.4 m/s. The navigation performance was enhanced by more than 98% compared to INS dead reckoning with 2D RMSE of 4m in 180 secs of complete GNSS signal outage.