Low-thrust spacecraft trajectory design and optimization

Student thesis: Doctoral Thesis


Low-thrust (LT) propulsion has attracted the attention of researchers and the space industry in the past few decades due to its superior propulsive efficiency compared to chemical rockets. Because of the low cost associated with LT propulsion, space missions with increasing degrees of sophistication, complexity, and scientific yield are envisioned for future projects, which would otherwise be impossible due to financial and technological constraints. The preliminary design of such missions is a tedious and time-consuming task; it demands advanced tools to quickly explore different solutions and evaluate them against multiple conflicting criteria to determine the best candidate. For that purpose, numerous methods and tools for optimizing LT trajectories have been proposed. However, most of these tools have limitations that seek performance improvements.

This thesis aims to contribute to the field of LT spacecraft trajectory design and optimization in two ways. The first contribution is to use the recently developed MOLTO-IT tool to determine an optimal LT interplanetary trajectory to Saturn as a part of a novel concept for an exploration mission of the inner large moons of Saturn. The software uses an approximate dynamical model to perform a quick search of time and fuel-optimal LT multi-gravity-assist trajectories. However, MOLTO-IT does not take the thrust bounds of a real LT engine into account, which may result in unfeasible solutions, especially if the attainable thrust levels are very low. To overcome this issue, propulsion constraints are incorporated into the original code to obtain realistic solutions. The tool is employed to determine the optimal number and sequence of fly-bys to reach Saturn with a hyperbolic excess speed of 1km/s in a reasonable flight time while minimizing the characteristic launch energy C3 and maximizing the arrival mass.

Results show that it is possible to reach the planet in 12.34 years (between January 21, 2028 and May 24, 2040) and 1014 kg arrival mass with the launch C3 of 27.04 km2/s2 and the initial mass of 1500 kg via an Earth-Venus-Venus-Earth-Saturn trajectory. This is a significant improvement with respect to similar solutions available in the literature.

The second main contribution of this thesis is in extending the applicability of the MOLTO-IT tool to the three-dimensional case for better estimation of trajectories with substantial out-of-plane motion. A mission to Ceres is used to benchmark the improvements relative to the original code. Results reveal that trajectories found by 3D MOLTO-IT are much closer to the optimal solutions obtained with a higher-fidelity, computationally intensive, direct-method-based optimizer in terms of launch and arrival dates, in-plane and out-of-plane average LT accelerations, and propellant consumption. Furthermore, when used as initial guesses for the advanced optimizer, the 3D MOLTO-IT trajectories require fewer iterations and less time to converge to the optimal solution than their counterparts obtained with the original version of MOLTO-IT.
Date of AwardApr 2023
Original languageAmerican English
SupervisorElena Fantino (Supervisor)


  • Low-Thrust propulsion
  • Multi-gravity assist maneuvers
  • Trajectory design
  • Multi-Objective optimization
  • Hybrid optimal control problem

Cite this