Matt Strong

Matt Strong

CS PhD at Stanford University

← Paper Digest

ARTS: An Open-Source, Reproducible Three-Arm Platform for Autonomy Research in Minimally Invasive Surgery

Jiawei Ge, Chunwoo Kim, Antony Goldenberg, Douglas McDonough, Justin Liao, Bo Ning, Sylvia Ladstaetter, Katya Van Anderlecht, Hao Ding, Laura Connolly, Ethan Kilmer, Francis Fan, Shouyue Hu, Tyler Salvador, Brendan Burkhart, Pit Henrich, Richard Cha, Tito Porras, Mathias Unberath, Kevin Cleary, Axel Krieger

November 2026

TL;DR An open-source surgical robot built from purchasable parts for about USD 137,000, with two UR5e arms driving da Vinci instruments, a UR5 stereo endoscope and a haptic surgeon console. On the real arms it reaches 0.14-0.25 mm ISO 9283 repeatability and a 58 ms hand-to-arm latency.

What they do

Learning-based surgical autonomy needs data, but most of it comes from the da Vinci Research Kit (dVRK), which runs on retired clinical hardware that labs cannot buy. ARTS is an open three-arm platform built only from purchasable parts. Two UR5e arms drive wristed da Vinci X/Xi instruments (energy instruments included), a UR5 holds a stereo endoscope, and a haptic console lets a surgeon teleoperate. Designs, firmware, software and a priced bill of materials (USD 137,320) are released.

How it works

  • Hardware. Two FLIR cameras image the endoscope’s optical channels, triggered by a custom board that measures each frame’s left-right skew. A safety relay holds all three arms in place (a powered, Category 2 stop).
  • Control. Teleoperation, learned policies or scripted motion all publish a tool-tip target. A 500 Hz quadratic program over nine joints tracks it, with the RCM, insertion depth and joint limits as hard constraints and an extra row that stops the UR5e from rolling the shaft.
  • Teleoperation. Haply Inverse3 handles with a Hall-effect jaw grip, motion scaled down 4x, and da Vinci-style clutch and camera pedals.
  • Software. ROS 2 Jazzy on a real-time kernel with pinned dependencies, installable on bare metal or in containers.

Results

  • Repeatability of 0.14-0.25 mm per pose; the dVRK Classic on the same protocol reached 0.09-1.24 mm. Physical RCM error 0.34-0.37 mm, measured with an optical tracker.
  • Latency of 58 ms from hand to arm and 68 ms from camera to monitor. Stereo skew 2.7 ± 3.2 µs, with no dropped frames over 30.5 min.
  • All three experts completed the FLS peg transfer (93-145 s) and two of three novices did.

Limitations

Tests were in free air without a cannula, so trocar and tissue forces were not measured. Instrument zeroing varies by a few degrees between instruments, and the jaw grip updates only about every 90 ms. No independent group has rebuilt the platform yet.

Figures

From Figure 1: The ARTS platform. The patient cart carries two UR5e instrument arms with da Vinci X/Xi instruments an...
From Figure 1: The ARTS platform. The patient cart carries two UR5e instrument arms with da Vinci X/Xi instruments and a UR5 stereo endoscope arm; next to it are the support tower, the surgeon console and the supervisor station with the stop button.
From Figure 2: Joints and coordinate frames of an instrument arm, with the remote center of motion (RCM) point on the...
From Figure 2: Joints and coordinate frames of an instrument arm, with the remote center of motion (RCM) point on the instrument shaft.
From Figure 3: Suturing (left) and peg-transfer (right) sequences performed with ARTS, seen from an external camera.
From Figure 3: Suturing (left) and peg-transfer (right) sequences performed with ARTS, seen from an external camera.
Abstract

Research on surgical autonomy needs open platforms that any laboratory can build, yet the most widely shared platform depends on donated clinical hardware, and open alternatives on industrial arms each lack part of the surgical chain. We present the Autonomous Robotic Translational System (ARTS), an open-source three-arm platform for autonomy research in minimally invasive surgery, built entirely from purchasable parts for about USD 137,000. Two UR5e arms hold wristed da Vinci X/Xi instruments, a UR5 arm holds a stereo endoscope with two FLIR cameras, the surgeon console provides haptic handles, footswitches and a stereo viewer, and a hardware stop independent of the software lets a supervisor freeze all three arms in place. The ROS 2 software pins every dependency, installs from the same scripts on bare metal or in containers, and records stereo images with hardware-measured timing. On the real arms, ARTS commands each arm at its controller's full rate (e.g., 500 Hz on the UR5e), the two cameras expose within 12 microseconds of each other, hand motion reaches the arm in 58 ms, and the endoscope image reaches the monitor in 68 ms. The ISO 9283 pose repeatability is 0.14-0.25 mm, comparable to a da Vinci Research Kit (dVRK) Classic arm on the same protocol, and an optical tracker measured a mean remote-center-of-motion error of 0.34-0.37 mm. Three expert and three novice operators performed the Fundamentals of Laparoscopic Surgery (FLS) peg transfer through a laparoscopic trainer box, and one of the experts sutured on a suture pad. The hardware designs, firmware and software are available at https://github.com/MRE-Lab-UMD/arts.

Summaries are written automatically from each paper's text. Figures and videos come from the paper and its project page; all credit goes to the authors.