Doppler Principle Based Situational Awareness Method for Autonomous Driving

Situational awareness is a critical connection in autonomous driving, which is now based on a combination of technologies. Camera, lidar, radar and ultrasonic sensors are the most commonly used modality for this task. Among them, lidar has the highest accuracy within 200 m and becomes the main input module for environmental perception. Nevertheless, lidar has defects of large size, high cost, high energy consumption, and accuracy is affected by the weather environment. To overcome the challenge, this paper proposes a situational awareness method for autonomous driving based on the Doppler principle, which is a help and supplement to the onboard radar. Specifically, five 24 GHz speed sensors are used to measure the Doppler speed of the vehicle in relative motion on the horizontal road surface, and the three-dimensional position coordinates of the vehicle and the two-dimensional speed of the horizontal plane can be calculated by a joint observation. Another 24 GHz speed sensor assists in distinguishing each target, judging the vehicle model, and predicting the possible movement of nearby vehicles. Experiment results illustrate that the measurement errors of the prototype system designed according to this paper are in line with experiment expectations, which proves the feasibility of Doppler radar for situational awareness.