Energy-Informed Strategies for Low-Thrust Trajectory Design in Cislunar Space

Bonnie Jessenia Prado Pino, Purdue University

Abstract

As cislunar and outer space exploration regains worldwide popularity, the low-thrust spacecraft technology, whether in the form of solar sails, electric propulsion or nuclear propulsion, has seen a major increase in the last two decades, as new technologies arise that not only seek for a reduction of the size of the spacecraft —and/or the payloads— but also to minimize the cost of spaceflights, while trying to approach further destinations in our solar system. Mission designers are being challenged with the need to develop new strategies to generate rapid and informed initial guesses for low-thrust spacecraft trajectory design, that are easily converged into fully continuous solutions in position, velocity and mass states, in a high-fidelity dynamical model that incorporates the true ephemerides and perturbations of the gravitational attracting bodies acting on the spacecraft as it navigates through space. In an effort to explore further mission options for spacecraft traveling in the lunar vicinity, new interest arises into the problem of constructing a general framework for the initial guess generation of low-thrust trajectories in cislunar space, that is independent of the force models in which the orbits of interest are de ned. Given the efficiency of the low-thrust engines, most vehicles are equipped to perform further exploration of the cislunar space after completion of their primary science and technology demonstrations in orbits around the Moon. In this investigation, a generalized strategy for constructing initial guesses for low-thrust spacecraft traveling between lunar orbits that exist within the context of multiple dynamical models is presented. These trajectories are converged as mass-optimal solutions in lower fidelity model, that are easily transitioned and validated in the higher-fidelity ephemeris model, and, achieve large orbital plane changes while evolving entirely within the cislunar region. The robustness of the initial guess generation of the spacecraft’s path, depends highly on the fidelity of the dynamical model utilized to construct such trajectories, as well as on the numerical techniques employed to converge and propagate them into continuous solutions. Other researchers have extensively investigated novel techniques for the generation of initial guesses for the low-thrust spacecraft trajectory design problem including, but not limited to, patched conics strategies, methodologies for the transformation of impulsive burns into nite burns, the orbit chaining framework and, more recently, artificial intelligence schemes. This investigation develops an adaptive orbit chaining type approach that relies on the energy parametrization of periodic orbits that exist within the context of the circular restricted three-body problem, to construct informed initial guess for the low-thrust spacecraft trajectory. A variety of multiple transfer applications for vehicles traveling between orbits in the cislunar region is explored for a wide range of low-thrust spacecraft with varying thrust acceleration magnitude. The examples presented in this investigation are consistent with the low-thrust parameters of previously own missions that utilized the same propulsion capabilities, such as, the DAWN mission and the Japanese Hayabusa missions 1 and 2. The trajectories presented in this work are optimized for either propellant consumption or time- of-flight in the lower-fidelity model, and later transitioned into a higher-fidelity ephemeris model that includes the gravitational attraction of the Sun, the Earth and the Moon. Two strategies are explored for the transition of trajectories from a lower-fidelity model to the higher-fidelity ephemeris model, both of which are successful in retaining the transfer geometry. The framework presented in this investigation is further applied to the upcoming NASA Lunar IceCube (LIC) mission to explore possible extended mission options once its primary science and technology demonstration objectives are achieved. It is demonstrated in this investigation that the strategies developed and presented in this work are not only applicable to the specific low-thrust vehicles explored, but it is applicable to any spacecraft with any type of propulsion technology. Furthermore, the energy-informed adaptive algorithm is easily transition to generate trajectories in a range of varying dynamical models.

Degree

Ph.D.

Advisors

Howell, Purdue University.

Subject Area

Energy|Astronomy|Systems science

COinS