Skip to Main Content
In a road collision avoidance system, it is vital to detect with high reliability the distance and approaching speed to an obstacle on the lane in which the vehicle is running. Although laser radar is already used in Japan and microwave radar (24 GHz) in the USA, in order to detect such a distance and approaching speed, there are some problems for both of them-vulnerability to rain, mist, etc. for laser radar applications, and the bulky size, etc. for microwave applications. Millimeter-wave radar systems with excellent environmental resistance and the potential in size reduction have been developed as means to solve such problems, but they have not been used in practice due to the insufficient detection performance, difficulty in on-vehicle installation, cost and so on. This paper discusses a method to improve the detection performance using a new algorithm for digital signal processing and a beam steering method for curved roads, by which remarkable improvements over the conventional system are attained against the false detections of unnecessary objects and failures in detecting necessary objects.