RobOralScan: Learning Active Intraoral Scanning for Robotic Dental Reconstruction

2026-06-25

AI Summary

This study addresses the inefficiency and variability inherent in manual full-arch and long-span intraoral scanning, which are highly susceptible to occlusion and operator experience. To overcome these limitations, the authors propose the first reinforcement learning–based robotic framework for autonomous intraoral scanning. The approach introduces a novel geometric memory–driven three-state observation space and a tooth-level coverage reward mechanism, integrated with a progressive training strategy and proprioceptive fusion to enable closed-loop, uniform scanning control. In simulation, the method achieves an average coverage of 92.58% (with a minimum of 88.45% per tooth) and a Chamfer distance of 0.00838, meeting clinical acceptability criteria in 8 out of 10 evaluation runs. Notably, the system demonstrates successful zero-shot transfer from simulation to real-world deployment.

Abstract

Intraoral scanning is widely used for digital optical impressions in prosthodontic, implant, and orthodontic treatment, but full-arch and long-span scanning remain labor-intensive tasks with limited automation. In the confined oral cavity, operators must continuously adjust scanner motion while accumulating narrow field-of-view observations, making reconstruction quality sensitive to missing tooth surfaces and operator workload. We propose RobOralScan, which, to the best of our knowledge, is the first reinforcement learning (RL)-based pipeline for robotic automatic intraoral scanning. RobOralScan introduces a geometric memory-based observation space that accumulates partial scan observations into a tri-state geometric representation, allowing the policy to reason over scan history and insufficiently observed regions. It further introduces tooth-wise coverage learning, combining coverage-aware reward signals and a progressive training scheme to improve global reconstruction coverage while reducing uneven coverage across individual teeth. The learned policy selects relative scanner motions from accumulated geometric memory and robot proprioception for closed-loop scan control within the oral workspace. RobOralScan achieves a Chamfer Distance of 0.00838, an average coverage of 92.58%, a lower-tail per-tooth coverage of 88.45%, and a normalized AUC of 0.6674, completing the scan criterion in 8 of 10 evaluation episodes. Furthermore, zero-shot sim-to-real experiments demonstrate its practical feasibility on a physical robot-scanner setup.

Problem

• intraoral scanning
• robotic dental reconstruction
• full-arch scanning
• automation
• 3D reconstruction

Innovation

• reinforcement learning
• intraoral scanning
• geometric memory
• coverage-aware reward
• robotic dental reconstruction