AI & ML New Capability

Automates the entire robotic data generation loop, including a self-resetting mechanism that restores unstructured workspaces without human intervention.

arXiv · March 13, 2026 · 2603.11811

Yongzhong Wang, Keyu Zhu, Yong Zhong, Liqiong Wang, Jinyu Yang, Feng Zheng

Why it matters

The requirement for manual environment resets is the single biggest bottleneck in scaling physical robot learning. By using a VLM-orchestrated forward-reverse planning system to reset the scene, RADAR enables 24/7 autonomous data collection, moving toward 'perpetual learning' systems.

From the abstract

The acquisition of large-scale physical interaction data, a critical prerequisite for modern robot learning, is severely bottlenecked by the prohibitive cost and scalability limits of human-in-the-loop collection paradigms. To break this barrier, we introduce Robust Autonomous Data Acquisition for Robotics (RADAR), a fully autonomous, closed-loop data generation engine that completely removes human intervention from the collection cycle. RADAR elegantly divides the cognitive load into a four-mod