RTPolicy: Real-Time (32.9 Hz) Multi-View Visuomotor Policy for Robotic Manipulation on Edge Devices
RTPolicy: Real-Time (32.9 Hz) Multi-View Visuomotor Policy for Robotic Manipulation on Edge Devices
This research introduces RTPolicy, a multi-view visuomotor policy optimized for low-latency robotic manipulation on edge devices. By combining view-aware feature fusion with visual compression and a flow-matching action generator, the system achieves real-time inference at 32.9 Hz (30.35 ms latency) on an RDK S600 device while maintaining high performance. On the RoboTwin~2.0 benchmark across 50 tasks, RTPolicy achieves a mean success rate of 52.26%, outperforming the strongest baseline ($\pi_0$ at 46.4%). Under asynchronous streaming inference on a real bimanual platform, it sustains a 72.0% mean success rate. The method reduces computational costs for visual representation and action generation while preserving necessary visual information.
The paper presents RTPolicy, a multi-view visuomotor policy designed for low-latency robotic manipulation on edge devices, achieving 32.9 Hz inference rates and a 52.26% mean success rate on the RoboTwin~2.0 benchmark.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
Reported numbers
inference latency
30.35
View original evidence
reaming inference on a real bimanual platform, where RTPolicy attains a mean success rate of 72.0\% and sustains real-time inference at 30.35 ms (32.9 Hz) on the RDK S600 and 149.72 ms (6.7 Hz) on the RDK S100, both in INT16. Project Page: https://d-robotics-ai-lab.github.io/RTPolicy.project.
Open source S5inference latency
149.72
View original evidence
reaming inference on a real bimanual platform, where RTPolicy attains a mean success rate of 72.0\% and sustains real-time inference at 30.35 ms (32.9 Hz) on the RDK S600 and 149.72 ms (6.7 Hz) on the RDK S100, both in INT16. Project Page: https://d-robotics-ai-lab.github.io/RTPolicy.project.
Open source S5
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
reaming inference on a real bimanual platform, where RTPolicy attains a mean success rate of 72.0\% and sustains real-time inference at 30.35 ms (32.9 Hz) on the RDK S600 and 149.72 ms (6.7 Hz) on the RDK S100, both in INT16. Project Page: https://d-robotics-ai-lab.github.io/RTPolicy.project.
Open source S5
nsive experiments demonstrate that this design reduces the computational cost of visual representation and action generation while preserving visual information required for manipulation. On the RoboTwin~2.0 single-task clean evaluation across 50 tasks, RTPolicy attains a mean success rate of 52.26\%, outperforming the strongest leaderboard baseline $\pi_0$ (46.4\%). We further evaluate RTPolicy under asynchronous st
Open source S4
Source：arXiv Robotics — research abstracts · arxiv.org