Optimal Control Formulation and Non-Linear Trajectory Generation for Envelope Protection

In a manned rotorcraft system with an active control system, using an active sidestick has significant advantages. Critical control position estimated by envelope protection systems are used to position softstop locations on the active sidestick to provide tactile cues to the pilot for envelope protection. In this paper, a new approach is proposed for estimating critical control positions from the adaptive estimate of limit parameter dynamics. The adaptive estimate of limit parameter dynamics is developed by augmenting a linear model of limit parameter dynamics with a single-hidden layer neural network. The weights of the neural network are tuned on-line to capture modeling uncertainty arising from the linear approximation of limit parameter dynamics and or changes in system configuration. The control limits corresponding to the limit boundary are computed from the optimal control solution that takes the limit parameter response to the limit boundary while minimizing a linear quadratic objective function. The optimal control problem is solved in real-time using the Nonlinear Trajectory Generator(NTG) package. The system predicts the hub moment limit without a priori training of its neural network and solved for the critical control position where a tactile softstop is placed as a limit avoidance cue. The method is evaluated in piloted simulations, at the Georgia Tech Carefree Maneuver Lab, within the Real-Time Interactive Prototype Technology Integration Development Environment (RIPTIDE) developed at the Army/NASA Rotorcraft Center.