Continuous Evaluation of Methods for Sensor Data Fusion of Video- Based Driver Assistance Systems, Using Virtual Road Tests

摘要

The vehicle of the future will support drivers in critical situations and alert them to potential hazards byrnmeans of Advanced Driver Assistance Systems (ADAS). Sensor-based perception of the relevant traffic conditions andrnsituation is essential in this context. The current evaluation approach of real-world road tests will no longer be sufficient duernto the high complexity of the system interactions. New simulation methods are needed to test and evaluate ADASrn(Advanced Driver Assistance Systems), using virtual road tests with realistic vehicle behavior and a complex trafficrnenvironment. This paper describes a new simulation technology which has the capability of integrating virtual cameras inrnaddition to the commonly known environment sensors (Radar, Lidar, ultrasonic) in the IPG CarMaker vehicle dynamicsrnsimulation for functional testing of sensor data fusion algorithms.