A new closed-loop control system is developed and evaluated for use in automatic carrier landings. The system is based on a new tracking filter that uses angle of attack and airspeed measurements from the airplane, in addition to the usual ship-based radar measurements. The filter dynamic model is based on actual flight dynamics, using the additional measurements, and thus achieves significant radar noise sensitivity reduction by eliminating the existing dependence on numerical differentiation of the radar output. An additional feedback loop that blends aircraft model estimates with radar measurements is also added to the system. Nonlinear optimization techniques are used to determine a set of optimal filter and control gains for the entire closed-loop system. A detailed digital computer simulation, verified with available flight data, indicates that the use of the flight-dynamics-based tracking filter and the addition of the feedback loop dramatically improves the noise rejection sensitivity in the automatic carrier landing system.