A perception centred self-driving system without HD Maps
This addresses the issue of labor-intensive map recording and accuracy degradation in new or changing environments for autonomous driving systems, though it appears incremental as it builds on existing perception methods.
The paper tackles the scalability problem in self-driving systems by proposing a new localization method that does not rely on HD Maps, GPS, or IMU, instead using only driving-related features like lane lines, with a new line detector tested on multiple datasets.
Building a fully autonomous self-driving system has been discussed for more than 20 years yet remains unsolved. Previous systems have limited ability to scale. Their localization subsystem needs labor-intensive map recording for running in a new area, and the accuracy decreases after the changes occur in the environment. In this paper, a new localization method is proposed to solve the scalability problems, with a new method for detecting and making sense of diverse traffic lines. Like the way human drives, a self-driving system should not rely on an exact position to travel in most scenarios. As a result, without HD Maps, GPS or IMU, the proposed localization subsystem relies only on detecting driving-related features around (like lane lines, stop lines, and merging lane lines). For spotting and reasoning all these features, a new line detector is proposed and tested against multiple datasets.