This presentation was made at the NAFEMS Americas "Creating the Next Generation Vehicle" held on the 14th of November in Troy.

The automotive engineering community is now confronting the largest technology transformation since its inception. This includes the electrification of powertrains for more efficient consumption and cleaner emissions, the reinvention of the battery with fast wireless charging capabilities and finally the advent of a fully autonomous vehicle. Compounding to these technology changes, the automotive companies design verification process is moving away from a major reliance on physical testing to almost a full virtual simulation product verification process.

The automotive engineering community is now confronting the largest technology transformation since its inception. This includes the electrification of powertrains for more efficient consumption and cleaner emissions, the reinvention of the battery with fast wireless charging capabilities and finally the advent of a fully autonomous vehicle. Compounding to these technology changes, the automotive companies design verification process is moving away from a major reliance on physical testing to almost a full virtual simulation product verification process.

Resource Abstract

The advancements in autonomous and ADAS equipped vehicle will start a new era of mobility. One of the challenges for acceptance of ADAS and autonomous vehicles is their safety. According to Rand corporation report, in order to prove with more than 95% statistical significance that an autonomous vehicle is safer than human driving, hundreds of millions of miles of autonomous driving data will be needed. Also, the number of different scenarios (traffic, weather, communications, faults, etc.) a vehicle can encounter is very large. It is not realistic to generate the entirety of that data using fleets of vehicles, hence virtual simulation is the only viable option. The key benefits of virtual simulation are – low cost, no risk to humans, and efficiencies from of parallelization. At the Simulation Innovation and Modeling Center (SIMCenter) of the Ohio State University, we are working with the automotive industry community and academic researchers to develop an open-source simulation building block-set to co-simulate traffic, vehicle dynamics, sensors and scenes. The open source nature of the simulation block-set allows participation of the entire automotive industry to solve pre-competitive research problems. The block-set will be a part of an evolving simulation eco-system to support the AV/ADAS testing. The block-set is modular in nature to allow users to either use it as a stand-along simulation tool or import only relevant pieces into projects. The block-set is based on Unreal Engine, but for traffic simulation, SUMO, an open-source simulation tool, is used. The vehicle dynamics and sensor models are also developed within Unreal Engine or other generic C++ code. It has plug-in points for ADAS/AV control logic, detection algorithms, and other proprietary models (e.g. sensors). This block-set will allow validation and verification of ADAS controls and perception algorithms in large numbers of scenarios and at edge cases. The simulations can be run in parallel, and specific test cases can also be visualized and individually analyzed. In this talk, I will present the architecture of the simulator and demonstrate its capabilities using an example.