[Paper Review] An Open-Source Robotics Research Platform for Autonomous Laparoscopic Surgery

Ariel Rodriguez, Lorenzo Mazza+5 more|arXiv (Cornell University)|Mar 9, 2026

Surgical Simulation and Training0 citations

TL;DR

The paper presents an open-source, robot-agnostic RCM controller with a closed-form analytical velocity solver and a full ROS-based platform using UR5e and Panda arms to enable teleoperation, data collection, and autonomous policy deployment for laparoscopic tasks, validated across phantom, ex vivo, and in vivo porcine procedures.

ABSTRACT

Autonomous robot-assisted surgery demands reliable, high-precision platforms that strictly adhere to the safety and kinematic constraints of minimally invasive procedures. Existing research platforms, primarily based on the da Vinci Research Kit, suffer from cable-driven mechanical limitations that degrade state-space consistency and hinder the downstream training of reliable autonomous policies. We present an open-source, robot-agnostic Remote Center of Motion (RCM) controller based on a closed-form analytical velocity solver that enforces the trocar constraint deterministically without iterative optimization. The controller operates in Cartesian space, enabling any industrial manipulator to function as a surgical robot. We provide implementations for the UR5e and Franka Emika Panda manipulators, and integrate stereoscopic 3D perception. We integrate the robot control into a full-stack ROS-based surgical robotics platform supporting teleoperation, demonstration recording, and deployment of learned policies via a decoupled server-client architecture. We validate the system on a bowel grasping and retraction task across phantom, ex vivo, and in vivo porcine laparoscopic procedures. RCM deviations remain sub-millimeter across all conditions, and trajectory smoothness metrics (SPARC, LDLJ) are comparable to expert demonstrations from the JIGSAWS benchmark recorded on the da Vinci system. These results demonstrate that the platform provides the precision and robustness required for teleoperation, data collection and autonomous policy deployment in realistic surgical scenarios.

Motivation & Objective

• Provide an open-source, robot-agnostic platform enforcing the trocar RCM constraint for laparoscopic robotics.
• Deliver a full-stack ROS-based system enabling teleoperation, data collection, and autonomous policy deployment.
• Validate RCM accuracy and trajectory smoothness across phantom, ex vivo, and in vivo porcine procedures.
• Demonstrate data collection and policy learning readiness through imitative learning on collected datasets.

Proposed method

• Develop a closed-form analytical velocity controller that deterministically enforces RCM in Cartesian space for any suitable manipulator.
• Integrate two hardware configurations (UR5e and Franka Panda) within a ROS-based software stack for teleoperation and policy deployment.
• Implement a ZMQ-based server–client architecture to decouple policy inference from ROS and include a safety controller for action validation.
• Record raw sensor data and endoscope videos during teleoperation to create structured datasets for learning.
• Apply fifth-order polynomial velocity profilers to generate smooth motion toward targets within jerk/acceleration limits.

Figure 2: The platform is built with modularity and scalability at its core, with inter-component communication abstracted through ROS topics to remain hardware-agnostic. This architecture allows input devices and robot actuators to be readily exchanged, and new modalities can be incorporated by sim

Experimental results

Research questions

• RQ1 Can an open-source, robot-agnostic RCM controller achieve sub-millimeter RCM deviations across phantom, ex vivo, and in vivo settings?
• RQ2 Is the platform capable of producing expert-level trajectory smoothness comparable to established benchmarks while enabling autonomous policy deployment?
• RQ3 Does the ROS-based full-stack architecture support robust teleoperation, data collection, and policy rollout across hardware and environments?
• RQ4 Can imitation learning policies trained on collected data achieve high success rates in laparoscopic tasks like bowel grasping and retraction?

Key findings

• RCM deviations remain sub-millimeter across phantom, ex vivo, and in vivo trials.
• Trajectory smoothness metrics (SPARC, LDLJ) are comparable to expert demonstrations from the JIGSAWS benchmark on the da Vinci system.
• An imitation-learning policy trained on the platform’s dataset achieved 85% success (17/20 roll-outs).
• The platform maintains robust performance under teleoperation, demonstration recording, and autonomous policy deployment in realistic surgical scenarios.
• The system is modular and hardware-agnostic, supporting UR5e and Panda arms with stereoscopic endoscopy and immersive visualization.
• The RCM controller operates without iterative optimization, using a closed-form analytic velocity solution for deterministic constraint enforcement.

Figure 4: Immersive stereoscopic visualization through the Meta Quest headset running Endomersion [ 11 ] showing the surgical workspace captured by the endoscope. Only the left eye view is shown; the right eye receives the corresponding stereo pair to enable depth perception.