Rotorcraft autorotation is a particularly challenging and hazardous maneuver. Reasons for this include: 1) Highly coupled aircraft dynamics, 2) Highly non-linear aircraft dynamics, 3) Critical aircraft states must be extracted from various sources that are not co-located, making pilot synthesis difficult, 4) Instrument scales and depiction methods make integration of state information difficult, 5) Pilot unfamiliarity with entry conditions/terminal constraints, and 6) A pilot's scanning strategy must usually undergo rapid and unnatural transitions during the maneuver. This paper puts forward a novel training display and methodology that takes advantage of automation's potential as a high-speed decision aid and the strengths of human pattern recognition and conditioning. The methodology applies optimal control theory to solve for a helicopter's trajectory and the required control inputs. A preview of the commanded input suite is displayed to the pilot, which will dynamically update as the vehicle state changes in time. Using this the pilot should be able to execute numerous autorotation maneuvers previously considered outside the operational envelope, in addition to performing 'standard' autorotations with a high degree of control consistency and accuracy. The preview display's function can also be extended to serve as an on-board pilot cueing aid. An autorotation cue set developed from functional requirements is presented, along with results of a preliminary experiment incorporating the display.