Isaac Sim implementation of the AMBF Surgical Robotics Challenge developed by Johns Hopkins LCSR Lab. This implementation has been developed and tested on Ubuntu 20.04
See: https://github.com/surgical-robotics-ai/surgical_robotics_challenge To find the work that this project is based on.
Ensure that the correct version of ROS is installed and sourced on your system. ROS1 Noetic can be found here: http://wiki.ros.org/noetic/Installation/Ubuntu
AMBF and directions for installation can be found here: https://github.com/WPI-AIM/ambf
Isaac Sim can be downloaded here from NVIDIA's wesbsite: https://developer.nvidia.com/isaac/sim
For additional resources, see Documents for summarized installation steps and notes for installations.
Refer to README in Assets and Scripts to access Isaac Sim Surgical Challenge for use
- After installation of all prerequisite programs, open Isaac Sim version 4.5.0
- Navigate to file and open either version of the Surgical_Challenge.usd (Surgical_Robotics_Challenge_New or Surgical_Robotics_Challenge_Old) You should see the following scene in the view monitor
- Start roscore in a seperate terminal
- Navigate to preferred ROS operated controller
cd Isaac-Sim-Surgical-Robotics-Challenge/scripts/surgical_robotics_challenge/teleoperation
- Play Isaac Sim Simulation
- Run desired teleoperation control with the following command line
python3 [controller].py -c isaac_sim --three false
From Isaac Sim 4.0.0 and on, you are able to activate and deactivate different assets within the scene. To activate another thread in the scene, see below:
- Rigid Body Thread: Set Physic Scene step to 300, Activate "Needle_Thread" and deactivate others. Turn OFF GPU in physics scene.
- Deformable Body Thread: Set Physic Scene step to 60 (default), Activate "Needle_Deformable" and deactivate others. Turn ON GPU in physics scene.
- Particle Body Thread: Set Physic Scene step to 60 (default), Activate "Needle_Particle" and deactivate others. Turn ON GPU in physics scene.
Repositories for other control devices can be found in the links below for MTM and Phantom Omni
Master Tool Manipulator (MTM) a. https://github.com/jhu-dvrk/dvrk-ros
Phantom Omni b. https://github.com/WPI-AIM/ros_geomagic
