# Reward-Free Continual Adaptation for Resilient Space Robots

arXiv: https://arxiv.org/abs/2608.23452 · HTML: https://mlsanity.com/paper/2608.23452/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Latent dynamics models and model-based RL, one of 11 directions in World models.
- Builds on 5 of its 15 references that are in the map.
- First posted on arXiv 24 Aug 2026.

## Placed in

- World models / [Latent dynamics models and model-based RL](https://mlsanity.com/wm/latent-dynamics-models-and-model-based-rl/)

## Builds on (most cited first, 5 of 5)

- [Domain Randomization for Transferring Deep Neural Networks from Simulation to the Real World](https://mlsanity.com/paper/1703.06907/) (2017)
- [A Definition of Continual Reinforcement Learning](https://mlsanity.com/paper/2307.11046/) (2023)
- [Finetuning Offline World Models in the Real World](https://mlsanity.com/paper/2310.16029/) (2023)
- [Space Robotics Bench: Robot Learning Beyond Earth](https://mlsanity.com/paper/2509.23328/) (2025)
- [Learning Tool-Aware Adaptive Compliant Control for Autonomous Regolith Excavation](https://mlsanity.com/paper/2509.05475/) (2025)

## Abstract (arXiv)

Space robots operate in extreme environments where hardware degradation can critically compromise traditional control strategies. While continual reinforcement learning offers a promising mechanism for online adaptation, it inherently requires access to a reward signal during deployment. However, precise reward computation in space is often infeasible due to the lack of external tracking systems and the overall complexity of the environment. To address the challenge of unobservable rewards, we introduce a reward-free continual learning framework that leverages latent-state world models. By pre-training a model-based agent across diverse simulations, the world model learns a robust predictor of the reward structure within its latent space. Upon deployment to an environment with severe hardware degradation, we freeze the observation encoder and reward predictor to update only the transition dynamics of the world model through unsupervised rollouts. By training the policy entirely on imagined trajectories generated by this updated world model, the agent adapts to altered dynamics without receiving new rewards. We demonstrate our approach across simulated planetary traversal, orbital navigation, and precision assembly tasks subjected to severe morphological failures.