Autonomous driving relies on a variety of sensors, especially on radars, which have unique robustness under heavy rain/fog/snow and poor light conditions. With the rapid increase of the amount of radars used on modern vehicles, where most radars operate in the same frequency band, the risk of radar interference becomes a compelling issue. This article analyses automotive radar interference and proposes several new approaches, which combine industrial and academic expertise, toward the path of interference-free autonomous driving. INTRODUCTION AND MOTIVATION Radar is becoming the standard equipment in all modern cars, supporting, e.g., cruise control and collision avoidance in most weather conditions whilst providing high-resolution detections on the order of centimeters in the millimeter-wave (mmWave) band. The next generation of Advanced Driver Assistance (ADAS) and Autonomous Drive (AD) vehicles will have a multitude of radars covering multiple safety and comfort applications like crash-avoidance, self-parking, in-cabin monitoring, cooperative driving, collective situational awareness and more. Since automotive radar transmissions are uncoordinated, there is a non-negligible probability of interference among vehicles, as shown in Fig. 1. While current automotive radars are already impacted by interference to some extent, it is today unlikely to get issues noticeable to the customer as the state-of-the-art automotive radars are continuously updated and improved on multiple system levels. However, the mutual interference problem is expected to become more challenging, unless properly handled, as more vehicles are equipped with a larger number of radars providing 360 • situational awareness at various distances to enable more advanced future ADAS and AD functionalities.