When developing autonomous vehicle controllers that will act directly on the physical environment, software-in-the-loop (SWIL) testing is important to ensure reliability of the codebase and system. Existing programs to do this job are often difficult to install, clunky, and may lack the granularity or complexity needed for certain experiments. Here, an approach is implemented that uses Docker’s containerization to facilitate running and maintaining the simulation. The simulation is comprised of a Python training and evaluation environment, modified so that it can interface with the same ROS components that will be running on a hardware-in-the-loop (HWIL) testbed. This containerized simulation can be easily run by new researchers and with new versions of the controller, making it convenient and useful for controller designers to run SWIL regression testing before deployment.