Towards Learning-based Localization for Autonomous Driving

Abstract

We present a robust and precise localization system that achieves centimeter-level localization accuracy in varied city scenes. Our system adaptively uses information from complementary sensors such as GNSS, LiDAR, and IMU to achieve high localization accuracy and robustness in various challenging scenes, including urban downtown, highway, tunnel and so on. Moving forward, we introduce our latest work in exploring learning-based localization system. It leverages the help of the deep neural network structures to establish a learning-based approach that achieves centimeter-level localization accuracy, comparable to prior state-of-the-art systems with hand-crafted pipelines.

Date
Mar 28, 2019 9:00 AM
Location
MIPR 2019, San Jose California, U.S.A.