Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Risk-Aware Autonomous Driving with Linear Temporal Logic Specifications

Shuhao Qi, Zengjie Zhang, Zhiyong Sun, Sofie Haesaert

TL;DR

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 that go beyond just collision risks, thereby reflecting a human-like risk awareness.

Abstract

Human drivers naturally balance the risks of different concerns while driving, including traffic rule violations, minor accidents, and fatalities. However, achieving the same behavior in autonomous driving 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 further formulate a linear programming (LP) problem for this LTL-based risk metric. Consequently, the synthesized policy balances different types of driving risks, including both collision risks and traffic rule violations. The effectiveness of the proposed approach is validated by three typical traffic scenarios in Carla simulator.

Risk-Aware Autonomous Driving with Linear Temporal Logic Specifications

TL;DR

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 that go beyond just collision risks, thereby reflecting a human-like risk awareness.

Abstract

Human drivers naturally balance the risks of different concerns while driving, including traffic rule violations, minor accidents, and fatalities. However, achieving the same behavior in autonomous driving 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 further formulate a linear programming (LP) problem for this LTL-based risk metric. Consequently, the synthesized policy balances different types of driving risks, including both collision risks and traffic rule violations. The effectiveness of the proposed approach is validated by three typical traffic scenarios in Carla simulator.
Paper Structure (16 sections, 16 equations, 8 figures, 1 table)

This paper contains 16 sections, 16 equations, 8 figures, 1 table.

Table of Contents

  1. Introduction
  2. Preliminary and Problem Statement
  3. Finite-State Model for Vehicles and Environments
  4. Specifications: Linear Temporal Logic (LTL)
  5. Risk-Aware Control Synthesis Problem
  6. human-like risk metric
  7. Product MDP and Reachability Problem
  8. Limitations of Event Probability
  9. Human-like Metric Design
  10. Risk-aware control
  11. Linear Programming Problem
  12. Implementation Framework
  13. Pedestrian Crossing Example
  14. Improved risk-aware control and experimental study
  15. Experimental Study
  16. ...and 1 more sections

Figures (8)

  • Figure 1: Unprotected turn in an intersection
  • Figure 2: Sketch for $\mathbb{P}_\pi$ with three traces in different colors.
  • Figure 3: Sketch for $\mathcal{R}_{\bar{\pi}}$. Red and blue indicate two types of events with different severity levels. The grey arrows illustrate decreasing awareness levels over the horizon due to discounting, while the colored arrows represent the corresponding risk values.
  • Figure 4: Risk field extracted from discounted occupation measures. The higher layer is the risk field, while the lower layer is a map of a reach-avoid problem, where the red rectangle is an obstacle and the green rectangle is the target.
  • Figure 5: The overall framework of control synthesis.
  • ...and 3 more figures

Theorems & Definitions (6)

  • Definition 1
  • Definition 2
  • Definition 3
  • Definition 4
  • Definition 5
  • Remark 1