← Back to the feed
Space & aviationResearch findings

Neural Operator for Collision-Aware Trajectory Planning in Spacecraft Swarms

Original published 2026.08.01 07:10 KSTAdded to Blackdeck 2026.11.19 15:03 KST

What has changed?

A neural operator that maps initial state distributions of spacecraft and obstacles to collision-avoidant, fuel-efficient trajectories is proposed.

Key points

  1. A method for achieving both fuel efficiency and collision avoidance in spacecraft swarms is introduced.
  2. The neural operator generates trajectories that avoid collisions and conserve fuel based on state distributions.
  3. The model generalizes zero-shot to 1,000 spacecraft and 11,000 obstacles without retraining.

Connect it to your work

The model generalizes zero-shot to 1,000 spacecraft and 11,000 obstacles without retraining.

What remains unproven

This research note is based on an abstract; check the original for test conditions and application scope.

Originals used for this note

Links open the original publisher in a new tab.

  1. arxivNeural Operator Learning for Collision-Aware Trajectory Planning of Spacecraft Swarms