This paper explores the feasibility of using information from a vehicle equations of motion as an aiding source for an Attitude Heading Reference System (AHRS) mechanized using low cost inertial sensors. This kind of an aiding scheme is particularly useful for operations in environments where information from GPS is not available such as indoors. An identified dynamic model of a miniature helicopter is fused with the outputs of a low cost rate gyro triad to mechanize an AHRS with bounded output errors. The information from the dynamic model and the rate gyros is fused using an Extended Kalman Filter (EKF). The rate gyro outputs are used to drive the EKF time update equations. Measured control inputs are used to drive the helicopter equations of motion. The output of the equations of motion form the measurements for the EKF. By post processing data from a flight test of a miniature helicopter (Blade CX2) in an indoor laboratory, it is shown that attitude errors can be bounded even in the presence of large rate gyro null-shifts.