Entry Guidance by Onboard Trajectory Planning and Tracking

While numerical predictor-corrector entry guidance algorithms have in recent years demonstrated great potential, a standing concern still is that the computational requirement such an algorithm demands may exceed the capability of the flight computer. An alternate entry guidance method combining a fullynumerical predictor-corrector algorithm with Linear Quadratic Regulator (LQR) based trajectory tracking control is explored in this paper. The predictor-corrector algorithm is used to plan a reference trajectory onboard. This reference trajectory is then tracked by an LQR tracking law, which provides the guidance commands. When necessary, the reference trajectory is updated periodically by the predictor-corrector algorithm with respect to the current vehicle state to eliminate any accumulated tracking error. The effectiveness of the LQR gains has been known and is demonstrated again in this paper to be independent of the particular reference trajectory used to generate them. Therefore a set of LQR gains can be generated offline, and stored online for tracking any reference trajectory planned by the predictor-corrector algorithm. Such an entry guidance strategy significantly reduces onboard computational requirements and still retains the adaptability of a numerical predictor-corrector entry guidance algorithm. The implementation details are presented. Numerical results are provided to demonstrate the effectiveness of the proposed method.