Nowadays when somebody talks about self-driving cars, we do not associate them with science fiction anymore. Vehicle automation is not only a rapidly growing industry; the notion of autonomous cars is becoming mainstream. The sudden growth of various driver assistance functions which are being developed, raises the need for quicker and more efficient testing.
However this shift in pace cannot cause a loss of thoroughness in testing. Quite the contrary, as we are trying to substitute the driver, a comprehensive analysis of the vehicle’s behaviour becomes all the more crucial. With the use of conventional test cars, this becomes immensely time and resource consuming. Although real vehicle tests will never completely be replaced, the exhaustive examination of a system this complex is unimaginable without the help of computer simulations.
In the past semester, I had the opportunity to gain insight into the development and testing of driver assistance systems at Robert Bosch Budapest. The department where I have been working is responsible for testing the comfort and safety functions realized by various sensors of a modern car. My focus was the hardware-in-the-loop testing of the front radar.
In this paper, I am going to show the radar based driver assistance functions and walk through the system testing process of a vehicle radar with the help of a simulation framework called CLARA. By the end of my thesis, I would like to have created a comprehensive picture about the state of today’s vehicle automation, and a profound study about testing related challenges, technologies and specialized solutions.