Suas Wind Sensing with Computational Fluid Dynamics and a Distributed Flush Airdata System

Small unmanned aircraft systems (sUAS) have proven their effectiveness for measuring both the inertial and aircraft-relative wind. Multiple methods for wind measurement from sUAS exist, but one of the more common instruments is the multi-hole probe (MHP). While the MHP is accurate and simple to use, there are two main drawbacks: 1) the MHP airdata system can cost several times that of the sUAS, and 2) the probe itself is often exposed to damage during routine operations. Flush airdata systems (FADS) are an alternative method of wind sensing, and work with pressure ports mounted flush with the aircraft surface. This removes any external components, thereby mitigating the risk of damage to the airdata system. The work presented details the implementation of a FADS for sUAS. Computational fluid dynamics simulations were used to determine the port locations of the FADS. Airframe locations were sorted based on the total sensitivity over a range of angles of attack and sideslip. Upon completion of hardware installation, the FADS was calibrated in flight using an onboard MHP. A portion of the flight testing was reserved for validation of the FADS. Multi-layer feedforward neural networks are employed to produce estimates of the angle of attack and sideslip, while static and stagnation ports on the fuselage measure airspeed. In the validation portion of flight tests, the FADS exhibited an overall mean error of 0.12 m/s in airspeed, but errors in angle of attack and sideslip were unbiased. Root-mean-square errors were 0.42 m/s, 0.65 ◦ , and 0.87 ◦ , respectively. Additionally, 97.7% of the errors in airspeed were within 1 m/s of the MHP, while 93.8% and 87.3% of the angle of attack and sideslip errors were within 1 ◦ . Flight tests show that a FADS can be calibrated in flight, and is an effective method for measuring the aircraft-relative wind from a small UAS