Risk-Aware Autonomous Driving with Linear Temporal Logic Specifications

2024-09-15

AI Summary

This work addresses the challenge of jointly optimizing and trading off multi-dimensional risks—collision, traffic violation, and casualty—in autonomous driving. We propose a novel linear temporal logic (LTL) modeling paradigm that incorporates temporal ordering and severity-weighted semantics, enabling explicit embedding of quantitative risk measures into LTL interpretations. To our knowledge, this is the first approach supporting decidable synthesis of safety and co-safety specifications under risk-aware constraints. By integrating occupancy measure theory, we formulate a unified linear programming (LP) optimization framework that jointly minimizes heterogeneous risk metrics subject to LTL-defined behavioral constraints. Extensive evaluation across three representative scenarios in the CARLA simulator demonstrates that our method significantly reduces the composite risk score while simultaneously ensuring regulatory compliance and collision avoidance. The resulting risk-aware decisions exhibit superior safety guarantees and decision rationality compared to state-of-the-art baseline methods.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories
• Reasoning under Uncertainty: Sequential Decision Making

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

Humans naturally balance the risks of different concerns while driving, including traffic rule violations, minor accidents, and fatalities. However, achieving the same behavior in autonomous systems remains an open problem. This paper extends a risk metric that has been verified in human-like driving studies to encompass more complex driving scenarios specified by linear temporal logic (LTL) that go beyond just collision risks. This extension incorporates the timing and severity of events into LTL specifications, thereby reflecting a human-like risk awareness. Without sacrificing expressivity for traffic rules, we adopt LTL specifications composed of safety and co-safety formulas, allowing the control synthesis problem to be reformulated as a reachability problem. By leveraging occupation measures, we formulate a linear programming (LP) problem for this LTL-based risk metric. Consequently, the synthesized policy balances different types of risks, including not only collision risks but also traffic rule violations. The effectiveness of the proposed approach is validated by three typical traffic scenarios in the Carla simulator.

Problem

• Balancing risks in autonomous driving systems
• Extending risk metrics for complex driving scenarios
• Formulating LTL-based risk metrics using linear programming

Innovation

• Extends risk metric with LTL for complex scenarios
• Formulates LP problem using occupation measures
• Validates approach with Carla simulator scenarios