What Really Matters for Robust Multi-Sensor HD Map Construction?

High-definition (HD) map construction methods are crucial for providing precise and comprehensive static environmental information, which is essential for autonomous driving systems. While Camera-LiDAR fusion techniques have shown promising results by integrating data from both modalities, existing approaches primarily focus on improving model accuracy, often neglecting the robustness of perception models—a critical aspect for real-world applications. In this paper, we explore strategies to enhance the robustness of multi-modal fusion methods for HD map construction while maintaining high accuracy. We propose three key components: data augmentation, a novel multi-modal fusion module, and a modality dropout training strategy. These components are evaluated on a challenging dataset containing 13 types of multi-sensor corruption. Experimental results demonstrate that our proposed modules significantly enhance the robustness of baseline methods. Furthermore, our approach achieves state-of-the-art performance on the clean validation set of the NuScenes dataset. Our findings provide valuable insights for developing more robust and reliable HD map construction models, advancing their applicability in real-world autonomous driving scenarios. Project website: https://robomap-123.github.io/.

Authors

• Xiaoshuai Hao
• Yuting Zhao
• Yuheng Ji
• Luanyuan Dai
• Peng Hao 0003
• Dingzhe Li
• Shuai Cheng 0002
• Rong Yin 0001

Keywords

• Training
• Accuracy
• Data augmentation
• Robot sensing systems
• Robustness
• Data models
• Autonomous vehicles
• Intelligent robots
• Map Construction
• Robust Function
• Robust Construction
• Robust Method
• Training Strategy
• Fusion Method
• Set Of Datasets
• Clear Set
• Multimodal Methods
• Improve Model Accuracy
• Types Of Corruption
• Transformer
• Point Cloud
• Real-world Scenarios
• Camera Images
• Average Precision
• Path Planning
• Sensor Failure
• Camera Data
• Transformation Module
• Lidar Data
• Resilience Scores
• Bird’s Eye
• 3D Object Detection
• LiDAR Sensor
• Corrupted Data
• Sensor Noise