Development of a Virtual Framework for Vehicle Control Logic Test and Validation using a Driver-In-The-Loop (DIL) Simulation

Current simulation tools can assist in expediting the development and validation of powertrain control systems compared to traditional methods that exclusively use physical testing. In this study, a co-simulation platform has been developed by connecting Matlab/Simulink, IPG CarMaker, GT-Suite, and PTV Vissim together to create a virtual calibration and validation environment. The purpose of developing this platform is to save time and reduce costs during vehicle testing by largely replacing the calibration and robustness evaluation carried out on physical vehicles with a simulation-based approach. The platform constructs a virtual environment that replicates the local road network as well as populates the roads with a randomly generated traffic pattern that the simulated vehicle can interact with. By combining the strengths of each software listed above, the developed virtual environment can provide realistic drive scenarios for testing new vehicle control algorithms. One of the primary challenges during development of the virtual test platform involved the driver model producing inputs while interacting with traffic that did not replicate the human behavior observed in on-road data. To avoid the complexity of developing a new driver model, the virtual platform was adapted to become a Driver-In-the-Loop (DIL) system. Outputs from a physical steering wheel, accelerator pedal, and brake pedal on the simulation rig are connected to the vehicle dynamic model thus providing the required driver inputs for this real-time simulation. With this platform the engineer can quickly generate new test conditions as well as repeat previous conditions to support Failure Mode and Effects Analysis (FMEA) of the prototype control logic. This paper explains the structure and configuration of the DIL simulator and demonstrates its application for control logic validation.