June 21, 2022 | Archer Flight Test Program

By: Geoff Bower, Chief Engineer

As we continue to advance the Maker Flight Test campaign, we recently began flying with our Tilt Propeller System activated. During our flight last week, we saw the first use of the Tilt Propeller System (TPS) for active control in hover, and the system performed as expected. Following the successful first hover flight of Maker in December 2021, the engineering team has been focused on installing and testing the TPS in Maker to prepare for more advanced transition envelope expansion flights. The TPS is a system of actuators, sensors and software that can move the angle of the forward six propellers relative to the wing in order to control the aircraft from hover through the transition to wing-borne flight in cruise.

We have now returned Maker to the air with an updated configuration that supports transition flight. The assumptions we made around the TPS have been validated in our early test flights and, now that we’re expanding our testing program, I thought I would explain a bit of why we conduct flight testing, how we minimize risk during flight test, and some of what we expect to learn from it. 

Maker flight testing is the culmination of years of hard work at Archer in close collaboration with our suppliers -- designing, assembling, system testing, and debugging. Flight testing is ultimately how we prove that our aircraft design meets its requirements. It also allows us to acquire valuable data to continuously improve the design tools and reduce risk on our production program. 

Before we fly, we exercise every system to the greatest extent practical on the ground and in simulation. Some examples include proof loading of structural components beyond the loads we ever expect them to see in flight, running our propulsion system on a test stand and in a wind tunnel, and running hardware-in-the-loop (HIL) simulations that exercise our flight control hardware and software.

Towards the end of the ground testing phase we begin to close some of the control loops that will be active when the vehicle takes flight. This starts with tests of single motors and actuators to make sure they perform nominally and don’t have negative impacts on other systems. We then move to testing all actuators at once as they’ll be used in flight. One of our last tests before flight is to activate the inner loop controllers that maintain the vehicle attitude (roll, pitch, and yaw) to check for ground resonance or other aeroservoelastic interactions. We also simulate complete flights by playing back actuator commands from our Hardware-In-The-Loop (HIL) sim to the vehicle while it is strapped to the ground. The only parts of the control system that we can’t test on the ground are the outer loop controllers that allow the vehicle to track a pre-planned trajectory. We still have high confidence in these controllers due to extensive simulation and flight testing of our sub-scale aircraft.

During flight test, we minimize risk to the aircraft by methodical flight envelope expansion. A flight envelope is defined by a set of limits that the aircraft stays within; things like airspeed, altitude or climb rate. Early flights begin within a conservative envelope where we have high confidence in the vehicle stability by simulation and ground testing. Additional tests extend that envelope in small steps one dimension at a time. Key test parameters are monitored by engineers on the ground, and if something unexpected is seen, the aircraft is returned to the previously cleared envelope either automatically (for time sensitive issues), or via command by the ground station operators. 

During our flight tests we record gigabytes of data from multiple systems. This data is used to update our simulation models to further reduce risk in future flights. I like to think of it as a growing bubble of knowledge, reducing the fog (or uncertainty) at the boundaries of the current envelope.

Some of the key data the team will be looking for during our upcoming flight test campaign include:

  • Flight mechanics model validation, such as trim motor RPMs and power draw as a function of airspeed
  • Control system stability margins estimated using system identification methods
  • Processing data to improve simulation models of vehicle aerodynamics, battery performance, and motor efficiency to match flight test data. These will be used to predict performance for future flights
  • Characterizing the vibration and thermal environments for future airborne equipment qualification
  • Gathering acoustic data to validate our prediction toolchain

This is an extremely exciting time here at Archer as we get to see all our hard work literally take flight. We remain on track to achieve transition flight this year and, along with the Archer team, I look forward to sharing further progress as we continue to expand the Maker flight envelope.