Ascent trajectory optimization for air‐breathing vehicles in consideration of launch window

Abstract

Summary This paper studies the optimal ascent trajectory for a one‐stage‐to‐orbit, air‐breathing vehicle. This kind of vehicle allows endoatmospheric lateral maneuver and thus can extend the launch window. To obtain the optimal orbit‐insertion trajectory in consideration of launch time, a novel optimization model is proposed by designing longitudinal and lateral profiles separately. The longitudinal flight aims to meet the orbital shape that contains the eccentricity and the angular momentum, which are transformed to boundary constrains in altitude, path angle, and velocity. For lateral motion, the orbit inclination and right ascension of ascending node are designed to vary simultaneously as the range. By designing an altitude profile, the terminal altitude and path angle are naturally met and removed from the constraints and accurate orbit insertion is realized by searching the total range and the initial azimuth. Then, the original problem is converted to a parameter optimization problem and solved by a hybrid optimization algorithm, in which the solution is first searched by a particle swarm optimization method and then refined by a modified golden section method. Simulation tests the proposed method with various scenarios. Compared to traditional launch vehicles, a launch window over 1.5 hours is derived without consuming the propellant in the upper stage.

Multiconstrained Ascent Trajectory Optimization Using an Improved Particle Swarm Optimization Method

“…Mathematical Models of the Launch Vehicle. The equations of motion of the launch vehicle are as follows [5]:…”

“…Trajectory optimization of the launch vehicle is in general considered a typical nonlinear optimal control problem [1]. Initially, analytical methods are developed to solve ascent trajectory optimization problems based on the optimal control theory [2][3][4], which usually appears in the indirect methods [5][6][7]. In this kind of method, the optimization problem is first converted by formulating a Hamilton function and then solved by solving a two-point boundary value problem using the maximum principle [8].…”

Adaptive dynamic surface control using neural networks for hypersonic flight vehicle with input nonlinearities

“…Some of the earlier works of others were explored in Refs. 6-8 to solve problems such as nonlinearity, strong coupling, status and control constraints, and parameter uncertainties of the HFV systems. However, a single control method was insufficient.…”