ORBEX-A

Adaptive control for orbital capture

Adaptive model predictive control for approaching tumbling objects in orbit, with explicit safety constraints that account for uncertain dynamics.

Context
M.S. research · UCLA
Collaboration
With Brett T. Lopez
Areas
Robust control · MPC · Orbital robotics
AI concept illustration of two spacecraft approaching over Earth, connected by a translucent teal approach corridor
Orbital rendezvous and capture, illustrated with a conceptual approach corridor. AI-generated concept illustration.

Problem

Capturing a tumbling, noncooperative object requires a chaser to rendezvous while the target motion and environmental dynamics remain uncertain. A controller must adapt without discarding the safety margins that make the approach robust.

Contribution

My master's research at UCLA, carried out with Brett T. Lopez, focused on adaptive tube control for orbital rendezvous and capture.

The work develops adaptive dynamic tube model predictive control around nonlinear rendezvous optimization, robust tubes, set-membership identification, and target-aware docking constraints.

The public implementation supports single- and multi-chaser missions and reports physical margins, tightened constraints, parameter estimates, target attitude, and inter-chaser spacing.

Outcome

The research was published at the 2025 IEEE Aerospace Conference and is accompanied by a reproducible public simulation codebase.

Simulation is the basis of the results, with constraint margins and parameter estimates available for examining controller behavior.

Simulation results

Adaptive tube behavior

Simulated positional norm and tube boundary versus orbital true anomaly, with an inset showing their separation
The simulated positional norm (blue) follows the adaptive tube boundary (red dashed). The inset resolves the small separation between the curves. This is a computational result from the repository's paper assets. Source

How to cite this work

Peer-reviewed conference paper · IEEE Aerospace Conference, 2025