Improving Explainable Object-induced Model through Uncertainty for Automated Vehicles
Authors:
- Shihong LingSLShihong Ling
- Yue WanYWYue Wan
- Xiaowei JiaXJXiaowei Jia
- Na DuNDNa Du
Preprint
Abstract
The rapid evolution of automated vehicles (AVs) has the potential to provide safer, more efficient, and comfortable travel options. However, these systems face challenges regarding reliability in complex driving scenarios. Recent explainable AV architectures neglect crucial information related to inherent uncertainties while providing explanations for actions. To overcome such challenges, our study builds upon the "object-induced" model approach that prioritizes the role of objects in scenes for decision-making and integrates uncertainty assessment into the decision-making process using an evidential deep learning paradigm with a Beta prior. Additionally, we explore several advanced training strategies guided by uncertainty, including uncertainty-guided data reweighting and augmentation. Leveraging the BDD-OIA dataset, our findings underscore that the model, through these enhancements, not only offers a clearer comprehension of AV decisions and their underlying reasoning but also surpasses existing baselines across a broad range of scenarios.Keywords
All Related Versions
- Version 1, 2024-02-23, ArXiv
- Published version: , 443.
This publication has 24 references indexed in Scilit:
- Explainability of Deep Vision-Based Autonomous Driving Systems: Review and ChallengesInternational Journal of Computer Vision, 2022
- Driving behavior explanation with multi-level fusionPattern Recognition, 2022
- Aleatoric and epistemic uncertainty in machine learning: an introduction to concepts and methodsMachine Learning, 2021
- Information Aware max-norm Dirichlet networks for predictive uncertainty estimationNeural Networks, 2021
- Look who’s talking now: Implications of AV’s explanations on driver’s trust, AV preference, anxiety and mental workloadTransportation Research Part C: Emerging Technologies, 2019
- Stop explaining black box machine learning models for high stakes decisions and use interpretable models insteadNature Machine Intelligence, 2019
- The MBPEP: a deep ensemble pruning algorithm providing high quality uncertainty predictionApplied Intelligence, 2019
- Textual Explanations for Self-Driving Vehicles2018
- Sentiment analysis: Bayesian Ensemble LearningDecision Support Systems, 2014
- Why did my car just do that? Explaining semi-autonomous driving actions to improve driver understanding, trust, and performanceInternational Journal on Interactive Design and Manufacturing (IJIDeM), 2014