Compositional Diffusion Models for Powered Descent Trajectory Generation with Flexible Constraints
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913534292525056 |
|---|---|
| author | Briden, Julia Du, Yilun Zucchelli, Enrico M. Linares, Richard |
| author_facet | Briden, Julia Du, Yilun Zucchelli, Enrico M. Linares, Richard |
| contents | This work introduces TrajDiffuser, a compositional diffusion-based flexible and concurrent trajectory generator for 6 degrees of freedom powered descent guidance. TrajDiffuser is a statistical model that learns the multi-modal distributions of a dataset of simulated optimal trajectories, each subject to only one or few constraints that may vary for different trajectories. During inference, the trajectory is generated simultaneously over time, providing stable long-horizon planning, and constraints can be composed together, increasing the model's generalizability and decreasing the training data required. The generated trajectory is then used to initialize an optimizer, increasing its robustness and speed. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_04261 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Compositional Diffusion Models for Powered Descent Trajectory Generation with Flexible Constraints Briden, Julia Du, Yilun Zucchelli, Enrico M. Linares, Richard Robotics Machine Learning Systems and Control Optimization and Control This work introduces TrajDiffuser, a compositional diffusion-based flexible and concurrent trajectory generator for 6 degrees of freedom powered descent guidance. TrajDiffuser is a statistical model that learns the multi-modal distributions of a dataset of simulated optimal trajectories, each subject to only one or few constraints that may vary for different trajectories. During inference, the trajectory is generated simultaneously over time, providing stable long-horizon planning, and constraints can be composed together, increasing the model's generalizability and decreasing the training data required. The generated trajectory is then used to initialize an optimizer, increasing its robustness and speed. |
| title | Compositional Diffusion Models for Powered Descent Trajectory Generation with Flexible Constraints |
| topic | Robotics Machine Learning Systems and Control Optimization and Control |
| url | https://arxiv.org/abs/2410.04261 |