Taylor, Brian2015-06-192015-06-192015-06-10http://hdl.handle.net/11299/172658The goal for this flight was to repeat the auto land. We modified the flare guidance to track 0.1 m/s descent rate instead of 0.5 and increased the airspeed in flare to 16 m/s. We decided to engage flare higher (around 20 ft) to let everything settle more before touching down. The launch, circuit around the airspace, and approach were again flawless. The flare guidance started very well, but stalled the aircraft late in the flare and resulted in a cartwheel. From a quick look at the data, this seems to be from a high pitch up that outran the auto throttle. We're considering increasing the theta and phi tracking bandwidth and putting in some reduced theta command limits in the flare to keep it from happening again (notionally, something like 0 to 8 or 10 degrees).enUMore Test RangeFenrir Flight 28Attitude and Airspeed HoldRigid Body Control Law SynthesisPilot Commanded Phi and ThetaAutolandZdot Vertical Velocity TrackerDataset