University of Iowa, Mechanical Engineering
Langley Research Center – Simulation and Path Planning for Continuum Robots
Final goals of your project?
The primary goal of my project was to develop a simulation framework for continuum/soft robotics in Gazebo (which only supports rigid dynamics).
Describe what you did during the internship.
During my internship, I developed a Gazebo/ROS2 framework for the simulation of continuum systems in Gazebo, and I developed a numerical method for solving the set of nonlinear PDEs that make up the Cosserat rod model. This model captures all relevant modes of deformation for continuum/soft systems that are suitably slender. I implemented the numerical method in a Matlab ROS2 node and used it to control the behavior of a discretized Gazebo model. I also validated this numerical method against existing solvers in the literature. This established a framework for the simulation of continuum/soft systems in Gazebo with an external ROS2 node acting as a dynamics solver for the system.
Did you achieve your goals? What were the results?
I achieved most of what I set out to achieve however, there is significant room for future work. Specifically, I would like to have a real system that needs to be digitally cloned so more concrete tests can be run. The current results of the work are a generalized framework for the simulation of soft/continuum systems in Gazebo. This is novel and applicable to several ongoing projects.
Describe positive lessons learned.
Before this project, my experience with Gazebo and ROS was limited. I now have a baseline understanding of the capabilities of these tools and a level of comfort to approach tasks leveraging them in the future.
Describe negative lessons learned.
Bureaucratic tasks within government agencies (while important) delay progress and are often frustrating.