Validation Platform of Autonomous Capability for Commercial Vehicles 2019-01-0686
Global deployment of autonomous capability for commercial vehicles is a big challenge. In order to improve the robustness of autonomous approach under different traffic scenarios, environments, road conditions, and driver behaviors, a combined approach of virtual simulation, vehicle-in-the-loop (VIL) testing and proving ground testing have been established for algorithms validation. During the validation platform setup, different platforms for different functionalities have been studied, including open source virtual testing environment (CARLA, AirSim), and commercial one (TruckMaker from IPG). We also cooperate with Mcity to do proving ground validation. In virtual testing, the functionality of sensors (camera, radar, Lidar, GPS, IMU) can be applied in the virtual environment. In VIL testing, real world and virtual test will be connected for different validation purposes. The proving ground testing is performed for specific real-life scenarios and high safety. Several challenges have been overcome during implementation, including sensor system consistency, data communication protocol establishment, FLOPs of computing device and etc. In this paper, a self-implemented autonomous driving framework, including perception, planning and control algorithm, will be validated in different virtual and physical validation approaches. The perception algorithm is deep learning based lane keeping and objective detection system, several popular neural network has been bench-marked. Planning is based on hybrid A* algorithm. The control algorithm is utilizing MPC. Several test case studies for algorithm testing will be discussed. And conclusions will be made on the established validation platforms for adopting autonomous driving technologies. And the next steps for the ADAS system performance improvement will be discussed.