Simulation-as-a-Service

• A key enabler for Software in Loop processes is the ability to use Simulation as a Service (SaaS) rather than just isolated instances of software.

• Metamoto is a simulation software which can be used to generate synthetic data for training purposes as well as to integrate our sensor and control models with preexisting models provided in the software.

• We utilize System Under Test (SUT) architecture developed by Metamoto that utilizes containers and remote procedural calls for communication to deploy our prototype algorithms to interface with the simulation engine thereby providing SaaS.

Methodology

• Utilizing a co-simulation approach, to leverage the unique advantages of Metamoto, ROS and MATLAB/Simulink.

• Metamoto is a light-weight yet robust simulation engine with rich, granular control over environmental parameters.

• ROS is the basis of software like Autoware.AI, which is one of the best tools to develop self-driving cars.

• Docker containers run directly on the machine kernel and don’t require any hypervisor.

• The multitude of toolboxes provided by MathWorks and the simple GUI based development environment is a boon.

A dbc file was created that will be linked with Python code for transmission & reception of message

The messages sent on CAN from Nuvo were received by Simulink running on MABX.

By extending the Steer-by-Wire system for accommodating the pedals in the Logitech G29, a similar Throttle-by-Wire and Brake-by-Wire system was implemented.

MicroAutoBox interface indicating reception of actuator data