Uncertainty-Aware Autonomous Vehicles: Predicting the Road Ahead

2025-10-26

AI Summary

Autonomous driving perception systems often exhibit overconfidence in rare or ambiguous scenarios, posing significant safety risks—necessitating rigorous uncertainty quantification. This paper introduces the first integration of Random Set Neural Networks (RS-NNs) into an autonomous driving software stack, leveraging their output belief functions to enable fine-grained, category-level uncertainty estimation for road layout classification—outperforming conventional CNNs and Bayesian approaches in both accuracy and calibration. We design a ROS-based, uncertainty-driven closed-loop control framework that dynamically adapts vehicle speed based on real-time uncertainty estimates. Extensive validation on a real-world autonomous racing platform demonstrates that RS-NNs achieve superior classification accuracy and uncertainty calibration across diverse complex road conditions. Crucially, the system enables aggressive maneuvering under high-confidence predictions while triggering proactive speed reduction and hazard avoidance under low-confidence conditions—thereby substantially enhancing overall robustness and operational safety.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Multiagent Systems: Multiagent Systems under Uncertainty
• Reasoning under Uncertainty: Uncertainty Representations

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Autonomous Vehicle (AV) perception systems have advanced rapidly in recent years, providing vehicles with the ability to accurately interpret their environment. Perception systems remain susceptible to errors caused by overly-confident predictions in the case of rare events or out-of-sample data. This study equips an autonomous vehicle with the ability to 'know when it is uncertain', using an uncertainty-aware image classifier as part of the AV software stack. Specifically, the study exploits the ability of Random-Set Neural Networks (RS-NNs) to explicitly quantify prediction uncertainty. Unlike traditional CNNs or Bayesian methods, RS-NNs predict belief functions over sets of classes, allowing the system to identify and signal uncertainty clearly in novel or ambiguous scenarios. The system is tested in a real-world autonomous racing vehicle software stack, with the RS-NN classifying the layout of the road ahead and providing the associated uncertainty of the prediction. Performance of the RS-NN under a range of road conditions is compared against traditional CNN and Bayesian neural networks, with the RS-NN achieving significantly higher accuracy and superior uncertainty calibration. This integration of RS-NNs into Robot Operating System (ROS)-based vehicle control pipeline demonstrates that predictive uncertainty can dynamically modulate vehicle speed, maintaining high-speed performance under confident predictions while proactively improving safety through speed reductions in uncertain scenarios. These results demonstrate the potential of uncertainty-aware neural networks - in particular RS-NNs - as a practical solution for safer and more robust autonomous driving.

Problem

• Addresses overconfidence in autonomous vehicle perception systems
• Quantifies prediction uncertainty using Random-Set Neural Networks
• Enables dynamic speed control based on road condition uncertainty

Innovation

• Random-Set Neural Networks quantify prediction uncertainty
• Belief functions identify uncertainty in novel scenarios
• Uncertainty dynamically modulates vehicle speed for safety