Skip to content

Anatomical Mesh-Based Virtual Fixtures for Surgical Robots (IROS 2020). Repo contains code for interactive demo and dvrk demo.

Notifications You must be signed in to change notification settings

mli0603/PolygonMeshVirtualFixture

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Anatomical Mesh-Based Virtual Fixtures for Surgical Robots

This is the offical repo for our work Anatomical Mesh-Based Virtual Fixtures for Surgical Robots accepted in IROS 2020. We present a framework to automatically generate arbitrary complex virtual fixtures for any geometries given a polygon mesh (STL) file. The framework is integrated with core CISST-SAW libraries, thus supports all CISST-based robots with a few lines of change, including but not limited to dVRK and UR. To apply virtual fixture to your robot, there are very few lines needed to change in your control code. This repo contains an interactive demo as well as example code of the virtual fixture working with dVRK.

If you find our work relevant, please cite

@article{li2020anatomical,
  title={Anatomical Mesh-Based Virtual Fixtures for Surgical Robots},
  author={Li, Zhaoshuo and Gordon, Alex and Looi, Thomas and Drake, James and Forrest, Christopher and Taylor, Russell H},
  journal={arXiv preprint arXiv:2006.02415},
  year={2020}
}

Table of Content

System Requirement

We have tested the code functionality on Ubuntu 16.04, 18.04 and 20.04.

Dependencies

  • 3D Slicer: We use 3D Slicer as a front end for visualization. It is required to use 3D Slicer 4.8.1 for OpenIGTLink compatibility. To download, please use this link here and extract the files. Double click on the executable named Slicer to run 3D Slicer.
  • OpenIGTLink: We use OpenIGTLink to communicate between 3D Slicer and CISST robot. To install, follow instruction here.
    • Note: you will need to make install after compiling the code.
    • Note: you will need to install the SlicerIGT extension in 3D Slicer. Please see here for how to do so.
  • ROS: To install ROS, pleaes follow instruction here
  • Third-Party Packages:
    • For Ubuntu 16.04 and 18.04
      sudo apt install libxml2-dev libraw1394-dev libncurses5-dev qtcreator swig sox espeak cmake-curses-gui cmake-qt-gui git subversion gfortran libcppunit-dev libqt5xmlpatterns5-dev # most system dependencies we need
      sudo apt install python-wstool python-catkin-tools # catkin and wstool for ROS build
    • For Ubuntu 20.04
      sudo apt install libxml2-dev libraw1394-dev libncurses5-dev qtcreator swig sox espeak cmake-curses-gui cmake-qt-gui git subversion gfortran libcppunit-dev libqt5xmlpatterns5-dev # most system dependencies we need
      sudo apt install python3-wstool python3-catkin-tools python3-pip # catkin and wstool for ROS build, pip for python package install
      pip3 install osrf-pycommon # catkin tool needs osrf

Compilation

Please follow the following system command

source /opt/ros/melodic/setup.bash # or use whatever version of ROS is installed!
mkdir ~/catkin_ws # create the catkin workspace
cd ~/catkin_ws # go in the workspace
wstool init src # we're going to use wstool to pull all the code from github
catkin init
cd src # go in source directory to pull code
git clone https://github.com/mli0603/PolygonMeshVirtualFixture.git dvrk_mesh_vf # clone the code in a folder called dvrk_mesh_vf
wstool merge dvrk_mesh_vf/scripts/dvrk_mesh_vf.rosintall # add required repos
wstool up # now wstool knows which repos to pull, let's get the code
catkin build # ... and finally compile everything

To run the demo

Visualization

  • Launch 3D slicer and load the demo scene.

    • Load the demo scene. Click on File, then Add Data.

    • Select Choose File(s) to Add and select the scene file located at scene/Demo-Scene.mrml.

    • You should now see a scene with pediatric skull and two spheres (red and blue, but the spheres may overlap in the beginning so you only see one).
    • The red sphere represents the servo-ed position (i.e. commanded position from the MTM or mouse) and the blue sphere represents the measured position (i.e. actual position with virtual fixture imposed). If the red sphere enters the skull, blue sphere should be stopped at the exterior shown in the GIF.

  • Connect to IGTL-Bridge.

    • Navigate to Modules->IGT->OpenIGTLink IF. Select the IGTL connector.

    • Check Active box.

Interactive Demo - Simple teleop

This interactive demo runs a simple teleoperation "robot" where the robot position (blue sphere) is controlled by the red sphere. Drag the red sphere to move the robot around. The robot is subject to the mesh virtual fixture, thus it cannot penatrate the skull.

  • Follow steps for visualization above.

  • Source catkin workspace environment values by source ~/catkin_ws/devel/setup.bash

  • Start roscore by roscore.

  • Start the demo by rosrun dvrk_mesh_vf simpleTeleop.

  • Transform the skull into robot coordinate frame by navigating to Modules->IGT->OpenIGTLink IF in 3D Slicer. In I/O Configuration, select Skull to PSM transformation and click on Send.

  • To control the robot position (blue sphere), drag the red sphere.

dVRK

  • Follow steps for visualization above.

  • Source catkin workspace environment values by source ~/catkin_ws/devel/setup.bash

  • Start roscore by roscore.

  • Base on the file share/console-MTMR-PSM2Derived-TeleopDerived.json, create a configuraiton file that matches your MTMR and PSM2 (or MTML and PSM1) serial number.

  • Launch dVRK by rosrun dvrk_robot dvrk_console_json -j share/console-MTMR-PSM2Derived-TeleopDerived.json.

  • Transform the skull into robot coordinate frame by navigating to Modules->IGT->OpenIGTLink IF in 3D Slicer. In I/O Configuration, select Skull to PSM transformation and click on Send.

Simulated dVRK

  • Follow steps for visualization above.
  • Source catkin workspace environment values by source ~/catkin_ws/devel/setup.bash
  • Start roscore by roscore.
  • Start the simulated console by rosrun dvrk_robot dvrk_console_json -j share/console-MTMR_KIN_SIMULATED-PSM2Derived_KIN_SIMULATED-TeleopDerived.json. Optionally, you can visualize the PSM by roslaunch dvrk_robot dvrk_arm_rviz_only.launch arm:=PSM2.
  • Inform teleoperation logic that PSM is simulated by rostopic pub -1 /PSM2/set_simulation std_msgs/Bool "data: true"

  • In the console, first setup teleoperation

    1. Check Direct Control
    2. Enable Operator
    3. Home the robot
    4. Start teleoperation.



  • Move the second joint of MTM such that its value is 20. This step moves the tip above the skull.

    • Enable Direct Control in MTMR tab.
    • Change the joint value, click Move.
      • NOTE: you have to set it to 10 then 20 to avoid large joint motion.
    • Now you should see the tool tip is above the skull in 3D Slicer.

  • Transform the skull into robot coordinate frame by navigating to Modules->IGT->OpenIGTLink IF in 3D Slicer. In I/O Configuration, select Skull to PSM transformation and click on Send.

  • In the console,

    • Adjust the second joint such that its value is 10
    • Now you should see the tool tip (blue sphere) is stopped at the exterior of the skull in 3D Slicer, while the MTM (red sphere) passes through the skull, demonstrating VF is working.

Log

  • 2020.10.06: We have finished upgrading our code to be crtk compatible.
  • 2020.09.21: We have moved core funcitonalities of handling mesh into CISST library as cisst-mesh component in branch devel. We are working on upgrading the exisitng code base to be crtk compatible.
  • 2020.07.15: We are working on integrating some of the core components into JHU-CISST Library for minimal dependencies.

Acknowledgement

Special thanks to Anton Deguet for helping open-sourcing the code :)

About

Anatomical Mesh-Based Virtual Fixtures for Surgical Robots (IROS 2020). Repo contains code for interactive demo and dvrk demo.

Topics

Resources

Stars

Watchers

Forks