Case Study: General Motors achieves reliable low-velocity maneuvering in self-driving vehicles with MathWorks' MATLAB and Simulink

GM Develops Low-Velocity Maneuvering Capability in Self-Driving Vehicles Using MATLAB and Simulink

General Motors faced the challenge of enabling reliable low-velocity maneuvering in the final 100 meters of autonomous trips, where GPS can be unavailable and road markings are often absent. To tackle this, GM used MathWorks’ MATLAB and Simulink to model, simulate, and develop the perception and control algorithms needed for these complex close-range scenarios.

MathWorks helped GM build a 3D virtual world in Simulink and generate photorealistic computer-vision algorithms in MATLAB, integrating CarSim for vehicle dynamics, Unreal Engine for rendering, ROS2 for visualization, and dSPACE for validation to replicate fisheye camera images and accurately simulate vehicle behavior without GPS. The MathWorks-based workflow connected systems from proof of concept to production, accelerated development, improved architecture management, and delivered measurable outcomes such as accurate last-100-meter simulation and faster readiness for commercialization.

General Motors

Jonathan Naor

General Motors