Development of a Virtual Reality Based Vehicle Simulator System for Test and Development of ASV, Telematics and ITS 2007-01-0946
A driving simulator system named KAAS(KATECH Advanced Automotive Simulator) is developed in KATECH (Korea Automotive Technology Institute) as a test and development environment of advanced automotive electronics such as ASV devices, Telematics, ITS, and etc. This simulator system is the largest driving simulator in Korea. There were several technical challenges during the development process. First, Korea had no experience of building a full spherical dome system with 360 deg FOV(field of view) before KAAS. Second, because of the large weight of the dome system (5.5 ton), a large size 6DOF motion platform was placed under the dome system which is the largest electric motion platform (7 ton capacity) in Korea. Third, the vehicle dynamics model was developed for three years including field test. Fourth, the KAAS system has eight channel visual system (largest in Korea) to cover 360 deg FOV. Fifth, the KAAS system provides real-time CAN communication which allows several different HILS(hardware-in-the-loop) system to be synchronized with the main driving simulator scenarios. In order to satisfy the original purpose of the simulator, the KAAS system is integrated with several real-time HILS system, an in-vehicle network simulation system, wireless communication simulation systems, high speed signal analysis devices, a driver perception analysis system, and a GPS signal simulation system. This paper mainly discussed on the driving simulator part of KAAS system. Because multi-year in-depth research was required for each component, the development of each component can be regarded as separate research topics. As a first step of simulator validation, the output signals from the vehicle dynamics (translational acceleration and angular rate in XBody, YBody and ZBody) are compared with the actual measurement of dome motion using a 6 axis gyroscope/accelerometer package. The results show the fair match of acceleration trends, while the result still shows that more tuning of the motion platform control algorithm is necessary.