Temporal model improves outdoor LiDAR sensor localization accuracy
Temporal-Aware Fusion for Robust Outdoor LiDAR Localization
Computer Vision and Pattern Recognition
Summary
Knowing exactly where a sensor is in the world is really important for things like self-driving cars. Many current methods figure out sensor location one measurement at a time, which can get confused if the scene changes or looks similar in spots. The authors present a method that looks at multiple sensor readings over time to get a better, more stable estimate of the sensor’s position. Their approach uses smart ways to guess how points match across scans and combines these guesses while considering uncertainty, leading to more reliable location estimates in outdoor environments. Tests show this method works better than others on well-known datasets.
What this means in practice
- •For autonomous vehicle engineers: Improve self-driving car localization by using multiple consecutive LiDAR scans to enhance position accuracy in changing outdoor environments.
- •For robotics system developers: Build outdoor robots that maintain better global positioning by integrating temporal consistency and uncertainty in LiDAR sensor data processing.
Authors
Minghang Zhu, Zhijing Wang, Yuxin Guo, Chen Liu, Yongshu Huang, Wen Li, Sheng Ao, Cheng Wang
Abstract
LiDAR relocalization aims to estimate the global 6-DoF pose of a sensor in the environment. However, existing regression-based approaches often encounter limitations in dynamic or ambiguous scenarios, as they typically prioritize single-frame inference, leaving the potential of spatio-temporal consistency across scans not fully explored. In this paper, we propose a Temporal-aware Localization framework (TempLoc) designed to enhance the robustness of outdoor localization by effectively modeling sequential consistency. Specifically, a Global Coordinate Estimation module is first introduced to predict point-wise global coordinates and associated uncertainties for each LiDAR scan. A Prior Coordinate Generation module is then presented to estimate inter-frame point correspondences by the attention mechanism. Lastly, an Uncertainty-Guided Coordinate Fusion module is deployed to integrate both predictions of point correspondence in an end-to-end fashion, yielding a more temporally consistent and accurate global 6-DoF pose. Experimental results on the NCLT and Oxford RobotCar benchmarks show that our TempLoc outperforms state-of-the-art methods by a large margin, demonstrating the effectiveness of temporal-aware correspondence modeling in LiDAR relocalization.