Action Chunking with Transformers for Image-Based Spacecraft Guidance and Control
This addresses the problem of sample-efficient and precise control for spacecraft guidance, navigation, and control, though it is incremental as it adapts existing ACT methods to a new domain.
The paper tackled spacecraft guidance and control by developing an imitation learning method using Action Chunking with Transformers (ACT), which achieved high performance with only 100 expert demonstrations (6,300 interactions), outperforming a meta-RL baseline trained with 40 million interactions in accuracy and smoothness for an ISS docking task.
We present an imitation learning approach for spacecraft guidance, navigation, and control(GNC) that achieves high performance from limited data. Using only 100 expert demonstrations, equivalent to 6,300 environment interactions, our method, which implements Action Chunking with Transformers (ACT), learns a control policy that maps visual and state observations to thrust and torque commands. ACT generates smoother, more consistent trajectories than a meta-reinforcement learning (meta-RL) baseline trained with 40 million interactions. We evaluate ACT on a rendezvous task: in-orbit docking with the International Space Station (ISS). We show that our approach achieves greater accuracy, smoother control, and greater sample efficiency.