Robust Long-Range Perception Against Sensor Misalignment in Autonomous Vehicles

2024-08-20

arXiv.org

AI Summary

To address the significant degradation in long-range (>100 m) perception performance caused by sensor calibration misalignment in autonomous driving, this paper proposes the first end-to-end multi-task learning framework that jointly performs cross-modal sensor angular deviation detection, calibration uncertainty quantification (achieving calibration error < 0.15°), and input-data-space self-correction. The method requires no additional hardware or re-calibration, integrating uncertainty-aware calibration with a differentiable geometric self-correction module to enable real-time inference. Experimental results demonstrate a 32% improvement in angular deviation detection accuracy and a 47% reduction in false positives under long-range conditions. Moreover, after self-correction, BEV object detection achieves a 21.6% increase in mAP, substantially enhancing model generalization and robustness against calibration drift.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Computer Vision: Multi-modal Vision

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Advances in machine learning algorithms for sensor fusion have significantly improved the detection and prediction of other road users, thereby enhancing safety. However, even a small angular displacement in the sensor's placement can cause significant degradation in output, especially at long range. In this paper, we demonstrate a simple yet generic and efficient multi-task learning approach that not only detects misalignment between different sensor modalities but is also robust against them for long-range perception. Along with the amount of misalignment, our method also predicts calibrated uncertainty, which can be useful for filtering and fusing predicted misalignment values over time. In addition, we show that the predicted misalignment parameters can be used for self-correcting input sensor data, further improving the perception performance under sensor misalignment.

Problem

• Detects and corrects sensor misalignment in autonomous vehicles.
• Improves long-range perception robustness against sensor displacement.
• Predicts calibrated uncertainty for better sensor data fusion.

Innovation

• Multi-task learning for sensor misalignment detection
• Robust long-range perception under misalignment
• Self-correcting input sensor data using misalignment parameters