Title
CCTR: Calibrating Trajectory Prediction for Uncertainty-Aware Motion Planning in Autonomous Driving
Author
Cao, Chengtai (City University of Hong Kong)
Chen, Xinhong (City University of Hong Kong)
Wang, J. (TU Delft Microwave Sensing, Signals & Systems; City University of Hong Kong)
Song, Q. (TU Delft Embedded Systems)
Tan, Rui (Nanyang Technological University)
Li, Yung Hui (Hon Hai Research Institute)
Date
2024
Abstract
Autonomous driving systems rely on precise trajectory prediction for safe and efficient motion planning. Despite considerable efforts to enhance prediction accuracy, inherent uncertainties persist due to data noise and incomplete observations. Many strategies entail formalizing prediction outcomes into distributions and utilizing variance to represent uncertainty. However, our experimental investigation reveals that existing trajectory prediction models yield unreliable uncertainty estimates, necessitating additional customized calibration processes. On the other hand, directly applying current calibration techniques to prediction outputs may yield suboptimal results due to using a universal scaler for all predictions and neglecting informative data cues. In this paper, we propose Customized Calibration Temperature with Regularizer (CCTR), a generic framework that calibrates the output distribution. Specifically, CCTR 1) employs a calibration-based regularizer to align output variance with the discrepancy between prediction and ground truth and 2) generates a tailor-made temperature scaler for each prediction using a post-processing network guided by context and historical information. Extensive evaluation involving multiple prediction and planning methods demonstrates the superiority of CCTR over existing calibration algorithms and uncertainty-aware methods, with significant improvements of 11%-22% in calibration quality and 17%-46% in motion planning.
Subject
General
To reference this document use:
http://resolver.tudelft.nl/uuid:5948fcda-e5af-4898-8b88-9c1b3e10b28a
DOI
https://doi.org/10.1609/aaai.v38i19.30085
Embargo date
2024-09-24
ISSN
2159-5399
Source
Proceedings of the AAAI Conference on Artificial Intelligence, 38 (19), 20949-20957
Event
38th AAAI Conference on Artificial Intelligence, AAAI 2024, 2024-02-20 → 2024-02-27, Vancouver, Canada
Bibliographical note
Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.
Part of collection
Institutional Repository
Document type
journal article
Rights
© 2024 Chengtai Cao, Xinhong Chen, J. Wang, Q. Song, Rui Tan, Yung Hui Li