Learning Adaptive Multi-Task Guidance, Navigation, and Control via Hypernetworks
2026-07-27 • Robotics
Robotics
AI summaryⓘ
The authors created a new system called HYPER-GNC that helps space robots perform different tasks like moving at the right speed, docking, inspecting, and avoiding obstacles using one single controller. Instead of making a separate controller for each task, their approach uses a smart network that adjusts its settings based on the task it needs to do. This method works well even with unexpected disturbances and can handle new tasks without retraining. They tested it both in simulations and on a real satellite model, showing it works reliably in practical situations.
reinforcement learninghypernetworkactor-critic policyguidance navigation and control (GNC)multi-task learningroboticssimulation-to-reality gaporbital roboticstask embeddings
Authors
Ricard Marsal I Castan, Aman Arora, Antoine Richard, Andrej Orsula, Cédric Pradalier, Miguel A. Olivares-Méndez
Abstract
Autonomous free-flying robots in orbital environments require controllers that are both versatile and resource-efficient, yet maintaining a separate, task-specific policy for each mission profile is architecturally brittle and limits operational flexibility as requirements evolve. We introduce HYPER-GNC, a multi-task reinforcement learning framework in which a hypernetwork maps physics-informed task embeddings to the weights of a shared actor-critic policy, enabling a single compact controller to master four distinct GNC tasks: velocity tracking, docking, inspection, and navigation with obstacle avoidance. The continuous embedding space allows the controller to generalize to novel mission configurations at deployment time without any retraining. Extensive experiments demonstrate that HYPER-GNC achieves sample efficiency comparable to single-task specialists while maintaining stability under significant inertial perturbations and external body wrenches. We further validate the framework on a physical satellite emulator, successfully bridging the simulation-to-reality gap across all mission profiles. Code, trained models, and deployment scripts are made publicly available to support reproducibility.