RAPID: Reconfigurable, Adaptive Platform for Iterative Design

Zi Yin*, Fanhong Li*, Shurui Zheng, Jia Liu†
Tsinghua University
*Equal Contribution †Corresponding Author
arXiv Code

RAPID enables tool-free, plug-and-play reconfiguration of robotic manipulation systems, reducing setup time from minutes to seconds.

Abstract

Developing robotic manipulation policies is iterative and hypothesis-driven: researchers test tactile sensing, gripper geometries, and sensor placements through real-world data collection and training. Yet even minor end-effector changes often require mechanical refitting and system re-integration, slowing iteration. We present RAPID, a full-stack reconfigurable platform designed to reduce this friction. RAPID is built around a tool-free, modular hardware architecture that unifies handheld data collection and robot deployment, and a matching software stack that maintains real-time awareness of the underlying hardware configuration through a driver-level Physical Mask derived from USB events. This modular hardware architecture reduces reconfiguration to seconds and makes systematic multi-modal ablation studies practical, allowing researchers to sweep diverse gripper and sensing configurations without repeated system bring-up. The Physical Mask exposes modality presence as an explicit runtime signal, enabling auto-configuration and graceful degradation under sensor hot-plug events, so policies can continue executing when sensors are physically added or removed. System-centric experiments show that RAPID reduces the setup time for multi-modal configurations by two orders of magnitude compared to traditional workflows and preserves policy execution under runtime sensor hot-unplug events.

Interactive 3D Preview

Explore the RAPID hardware design in 3D. Use your mouse to rotate, zoom, and pan the model.

Hot-plug

The driver-level Physical Mask abstraction derives modality presence from USB hot-plug events. This enables auto-configuration and graceful degradation: policies can continue executing even when sensors are physically added or removed at runtime.

Hot-plug performance plot

The video below demonstrates that USB hot-plug events correctly trigger topic and mask changes in real-time as sensors are physically connected and disconnected.

Multi-Modal Configuration

RAPID supports diverse multi-modal configurations including wrist cameras, tactile sensors, and motor state feedback. Use the slider to explore different configuration states.

Multi-modal configuration.

Direct-Hand Micro-Motion

RAPID's compact, lightweight, and wearable design allows the human hand to directly sense force and tactile feedback during data collection. This enables capturing perception-driven, fine-grained manipulation tasks—such as striking a match, handwriting, wrist flicking, or tossing small objects—that require subtle force control beyond traditional teleoperation.

BibTeX

@article{yin2026rapid,
  title     = {RAPID: Reconfigurable, Adaptive Platform for Iterative Design},
  author    = {Yin, Zi and Li, Fanhong and Zheng, Shurui and Liu, Jia},
  journal   = {arXiv preprint arXiv:2602.06653},
  year      = {2026},
}