SafePath: Conformal Prediction for Safe LLM-Based Autonomous Navigation
Preprint
Abstract
Large Language Models (LLMs) show growing promise in autonomous driving by reasoning over complex traffic scenarios to generate path plans. However, their tendencies toward overconfidence, and hallucinations raise critical safety concerns. We introduce SafePath, a modular framework that augments LLM-based path planning with formal safety guarantees using conformal prediction. SafePath operates in three stages. In the first stage, we use an LLM that generates a set of diverse candidate paths, exploring possible trajectories based on agent behaviors and environmental cues. In the second stage, SafePath filters out high-risk trajectories while guaranteeing that at least one safe option is included with a user-defined probability, through a multiple-choice question-answering formulation that integrates conformal prediction. In the final stage, our approach selects the path with the lowest expected collision risk when uncertainty is low or delegates control to a human when uncertainty is high. We theoretically prove that SafePath guarantees a safe trajectory with a user-defined probability, and we show how its human delegation rate can be tuned to balance autonomy and safety. Extensive experiments on nuScenes and Highway-env show that SafePath reduces planning uncertainty by 77\% and collision rates by up to 70\%, demonstrating effectiveness in making LLM-driven path planning more safer.Keywords
All Related Versions
This publication has 13 references indexed in Scilit:
- Probabilistically Correct Language-Based Multi-Robot Planning Using Conformal PredictionIEEE Robotics and Automation Letters, 2024
- Safe Planning in Dynamic Environments Using Conformal PredictionIEEE Robotics and Automation Letters, 2023
- Uncertainty-Aware Model-Based Offline Reinforcement Learning for Automated DrivingIEEE Robotics and Automation Letters, 2023
- Algorithmic Learning in a Random World2022
- Deep Learning for Safe Autonomous Driving: Current Challenges and Future DirectionsIEEE Transactions on Intelligent Transportation Systems, 2020
- A Rule-Based Cooperative Merging Strategy for Connected and Automated VehiclesIEEE Transactions on Intelligent Transportation Systems, 2019
- Predictive Maneuver Planning for an Autonomous Vehicle in Public Highway TrafficIEEE Transactions on Intelligent Transportation Systems, 2018
- Least Ambiguous Set-Valued Classifiers With Bounded Error LevelsJournal of the American Statistical Association, 2018
- If, When, and How to Perform Lane Change Maneuvers on HighwaysIEEE Intelligent Transportation Systems Magazine, 2016
- Efficient Sampling-Based Motion Planning for On-Road Autonomous DrivingIEEE Transactions on Intelligent Transportation Systems, 2015