AI-Enabled Capabilities to Facilitate Next-Generation Rover Surface Operations
Cristina Luna, Robert Field, Steven Kay
TL;DR
The paper tackles the bottleneck of slow rover mobility by introducing an integrated AI-enabled stack that couples long-range obstacle detection (FASTNAV/FOD), multi-robot coordination (CISRU), and semantic terrain classification (ViBEKO/AIAXR) validated in Mars/Lunar analogue environments. The approach yields sustained autonomous speeds around 0.7–1.0 m/s, a 20 m obstacle-sensing horizon, and high terrain-classification accuracy, supported by space-grade hardware implementations. Key contributions include a validated multi-robot collaboration framework, robust perception pipelines, and synthetic-data–driven terrain models, culminating in TRL4 readiness with a roadmap to higher TRLs. The work has practical significance for Artemis lunar operations, Mars sample return, and ISRU demonstrations, enabling safer, faster, and more capable planetary surface missions through autonomous, human-robot collaboration.
Abstract
Current planetary rovers operate at traverse speeds of approximately 10 cm/s, fundamentally limiting exploration efficiency. This work presents integrated AI systems which significantly improve autonomy through three components: (i) the FASTNAV Far Obstacle Detector (FOD), capable of facilitating sustained 1.0 m/s speeds via computer vision-based obstacle detection; (ii) CISRU, a multi-robot coordination framework enabling human-robot collaboration for in-situ resource utilisation; and (iii) the ViBEKO and AIAXR deep learning-based terrain classification studies. Field validation in Mars analogue environments demonstrated these systems at Technology Readiness Level 4, providing measurable improvements in traverse speed, classification accuracy, and operational safety for next-generation planetary missions.