Exploiting Temporal Relations on Radar Perception for Autonomous Driving
This work addresses the cost-effective but low-precision radar perception problem for autonomous vehicles, representing an incremental improvement through temporal modeling.
The paper tackles the problem of low angular resolution and precision in radar-based object recognition for autonomous driving by exploiting temporal information from successive radar frames. The proposed temporal relational layer that models object relations across frames shows superiority over baseline approaches in both object detection and multiple object tracking.
We consider the object recognition problem in autonomous driving using automotive radar sensors. Comparing to Lidar sensors, radar is cost-effective and robust in all-weather conditions for perception in autonomous driving. However, radar signals suffer from low angular resolution and precision in recognizing surrounding objects. To enhance the capacity of automotive radar, in this work, we exploit the temporal information from successive ego-centric bird-eye-view radar image frames for radar object recognition. We leverage the consistency of an object's existence and attributes (size, orientation, etc.), and propose a temporal relational layer to explicitly model the relations between objects within successive radar images. In both object detection and multiple object tracking, we show the superiority of our method compared to several baseline approaches.