Physics-informed neural networks for spacecraft trajectory optimization with exact boundary conditions
Organizations: TU Graz, Graz, 8010, Austria Graduate Student Researcher, Institute of Theoretical and Computational Physics, Petersgasse 16/II · Julius Maximilian University Würzburg, Würzburg, 97074, Germany Professor, Institute of Computer Science, Am Hubland. · German Aerospace Center (DLR), Hardthausen am Kocher, 74239, Germany Group Leader, Institute of Space Propulsion, Am Langen Grund. · Brown University, Providence, Rhode Island, 02912 Fellow, Division of Applied Mathematics, 182 George Street. ·
Abstract
We use neural networks to optimize trajectories by modeling the position as a function of time and minimizing the thrust with respect to the network parameters. This corresponds to an unsupervised training procedure in which physical laws are incorporated directly into the optimization objective. In standard physics-informed neural networks, initial and terminal states are enforced through additional weighted loss terms, resulting in a multi-objective optimization problem that is highly sensitive to the choice of loss weights and can lead to poor convergence and reduced accuracy. To address this challenge, we construct a simple transformation, exploiting the mathematical structure of trajectory optimization problems, to enforce boundary conditions exactly a-priori. This transformation can be interpreted as a reparameterization of the solution space that embeds the boundary conditions directly into the function. This converts the original constrained optimization problem into an unconstrained one over admissible trajectories. I.e. the transformation eliminates the boundary condition penalty terms and reduces the problem to a single-objective optimization. Crucially, trajectory optimization comes with a low-dimensional boundary, which allows particularly simple and stable constructions of distance functions compared to general domains in partial differential equations. Numerical experiments demonstrate that the proposed approach meaningfully improves convergence behavior and solution accuracy.
Figures & tables
| = | neural network |
| = | neural-network parameters |
| , = | spacecraft position vector and velocity vector |
| = | neural-network trajectory |
| = | transformed neural network trajectory satisfying exact boundary conditions |
| = | time of flight (TOF), terminal time |
| = | physical time |
| # | ||
|---|---|---|
| 1 | 0.5 | |
| 2 | 1.0 | |
| 3 | 0.5 |
| Example | Method | mean of | variance of | CPU-time / s |
|---|---|---|---|---|
| Swing-by 2D | OpenGoddard (baseline) | - | ||
| (Sec. 5.1 ) | PINN with soft BCs ∗ | |||
| PINN with exact BCs | ||||
| Swing-by 3D | OpenGoddard (baseline) | - | ||
| Sec. 5.2 | PINN with soft BCs | |||
| PINN with exact BCs |
Appendix figures & tables3 assets
Supplementary material from the paper’s appendix.
Appendix
| Dim. | Best | |||
| 2 | Optimizable | 2000 | 100000 | 0.324 |
| 3 | Optimizable | 2000 | 100000 | 0.421 |
Trajectory Optimization
Trajectory optimization computes motion or control sequences that minimize costs while satisfying system dynamics, safety limits, and task constraints.
See all 223 papers