Safe and Non-Conservative Contingency Planning for Autonomous Vehicles via Online Learning-Based Reachable Set Barriers

2025-09-09

AI Summary

Addressing the challenge of simultaneously ensuring safety and efficiency for autonomous driving in dynamic, uncertain environments—particularly under unpredictable human driving behaviors and perception uncertainties—this paper proposes a real-time emergency trajectory planning framework integrating event-triggered online learning with forward reachable set (FRS) barriers. Our method innovatively combines control-intent set modeling, incremental FRS updating, and barrier function constraints to guarantee invariant safety without requiring precise trajectory predictions, thereby overcoming the conventional trade-off between conservatism and safety. We employ a consensus alternating direction method of multipliers (ADMM) for efficient optimization. Extensive validation in high-fidelity highway and urban simulations, as well as real-vehicle experiments, demonstrates significant improvements in traffic throughput and ride comfort, while maintaining safety and planning feasibility under complex uncertainties.

Technology Category

• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Search and Optimization: Learning to Search

Application Category

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

Abstract

Autonomous vehicles must navigate dynamically uncertain environments while balancing the safety and driving efficiency. This challenge is exacerbated by the unpredictable nature of surrounding human-driven vehicles (HVs) and perception inaccuracies, which require planners to adapt to evolving uncertainties while maintaining safe trajectories. Overly conservative planners degrade driving efficiency, while deterministic approaches may encounter serious issues and risks of failure when faced with sudden and unexpected maneuvers. To address these issues, we propose a real-time contingency trajectory optimization framework in this paper. By employing event-triggered online learning of HV control-intent sets, our method dynamically quantifies multi-modal HV uncertainties and refines the forward reachable set (FRS) incrementally. Crucially, we enforce invariant safety through FRS-based barrier constraints that ensure safety without reliance on accurate trajectory prediction of HVs. These constraints are embedded in contingency trajectory optimization and solved efficiently through consensus alternative direction method of multipliers (ADMM). The system continuously adapts to the uncertainties in HV behaviors, preserving feasibility and safety without resorting to excessive conservatism. High-fidelity simulations on highway and urban scenarios, as well as a series of real-world experiments demonstrate significant improvements in driving efficiency and passenger comfort while maintaining safety under uncertainty. The project page is available at https://pathetiue.github.io/frscp.github.io/.

Problem

• Balancing safety and efficiency in autonomous vehicle navigation
• Addressing unpredictable human-driven vehicle behaviors and perception errors
• Avoiding overly conservative planning while ensuring collision-free trajectories

Innovation

• Online learning-based reachable set barriers
• Event-triggered HV intent set quantification
• Consensus ADMM for contingency trajectory optimization