We recently tested arducopter on a turboshaft powered helicopter. First flights proved it works extremely well (as expected :-) ), still solving some mechanical issues though.
Flight controller is pixhawk, helicopter is a modified Vario XLV running a Wren MW54 twin stage engine, main rotor diameter is 2,5m.
Main code was modified to fit an H4X swashplate arrangement and turbine management. I added three phases connected to a switch, for turbine off/ turbine startup-ground idle (basically this turbine requires the pilot to move the throttle from fully closed to beyond half travel, then back to g. idle, to initiate start-up sequence--I included this routine in arducopter)/ flight idle (which consists of a power curve).
After some adjustments , I noticed that flight idle curve is almost "flat" once you get to the perfect M/R RPM point, for this helicopter being 790 RPM (while turbine is working between 75-85 % over the whole collective travel).
Here a short footage of the early test flights: