Attention-based Vehicle Self-Localization with HD Feature Maps

TL;DR

This paper introduces an attention-based neural network approach for vehicle self-localization using HD maps, improving accuracy and generalization while reducing data collection costs.

Contribution

The paper proposes a novel attention mechanism for matching point measurements to landmarks, and a simulation framework for training data generation in vehicle localization.

Findings

Achieves superior localization accuracy on multiple datasets.

Demonstrates strong generalization to unseen environments.

Reduces data collection costs with simulation-based training.

Abstract

We present a vehicle self-localization method using point-based deep neural networks. Our approach processes measurements and point features, i.e. landmarks, from a high-definition digital map to infer the vehicle's pose. To learn the best association and incorporate local information between the point sets, we propose an attention mechanism that matches the measurements to the corresponding landmarks. Finally, we use this representation for the point-cloud registration and the subsequent pose regression task. Furthermore, we introduce a training simulation framework that artificially generates measurements and landmarks to facilitate the deployment process and reduce the cost of creating extensive datasets from real-world data. We evaluate our method on our dataset, as well as an adapted version of the Kitti odometry dataset, where we achieve superior performance compared to related approaches; and additionally show dominant generalization capabilities.