Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Fault Detection and Human Intervention in Vehicle Platooning: A Multi-Model Framework

Farid Mafi, Mohammad Pirani

TL;DR

Addresses simultaneous fault detection and driver-state identification in vehicle platoons under communication failures using tail-vehicle measurements. Develops a transfer-function–based framework applicable to predecessor-following and symmetric bidirectional topologies, integrating driver behavior models to capture disturbance propagation. Introduces a blending-based identification to reduce the model count while preserving accuracy, with formal notes on uniqueness; validated in CarSim/Simulink simulations showing accurate fault-location and driver-state identification. Collectively, the work advances safe, real-time, tail-vehicle–driven fault diagnosis for human-in-the-loop platooning.

Abstract

Vehicle platooning has been a promising solution for improving traffic efficiency and throughput. However, a failure in a single vehicle, including communication loss with neighboring vehicles, can significantly disrupt platoon performance and potentially trigger cascading effects. Similar to modern autonomous vehicles, platoon systems require human drivers to take control during failures, leading to scenarios where vehicles are operated by drivers with diverse driving styles. This paper presents a novel multi-model approach for simultaneously identifying signal drop locations and driver attitudes in vehicular platoons using only tail vehicle measurements. The proposed method distinguishes between attentive and distracted driver behaviors by analyzing the propagation patterns of disturbances through the platoon system. Beyond its application in platooning, our methodology for detecting driver behavior using a multi-model approach provides a novel framework for human driver identification. To enhance computational efficiency for real-time applications, we introduce a blending-based identification method utilizing chosen models and weighted interpolation, significantly reducing the number of required models while maintaining detection accuracy. The effectiveness of our approach is validated through high-fidelity CarSim/Simulink environment simulations. Results demonstrate that the proposed method can accurately identify both the location of signal drops and the corresponding driver behavior. This approach minimizes the complexity and cost of fault detection while ensuring accuracy and reliability.

Fault Detection and Human Intervention in Vehicle Platooning: A Multi-Model Framework

TL;DR

Addresses simultaneous fault detection and driver-state identification in vehicle platoons under communication failures using tail-vehicle measurements. Develops a transfer-function–based framework applicable to predecessor-following and symmetric bidirectional topologies, integrating driver behavior models to capture disturbance propagation. Introduces a blending-based identification to reduce the model count while preserving accuracy, with formal notes on uniqueness; validated in CarSim/Simulink simulations showing accurate fault-location and driver-state identification. Collectively, the work advances safe, real-time, tail-vehicle–driven fault diagnosis for human-in-the-loop platooning.

Abstract

Vehicle platooning has been a promising solution for improving traffic efficiency and throughput. However, a failure in a single vehicle, including communication loss with neighboring vehicles, can significantly disrupt platoon performance and potentially trigger cascading effects. Similar to modern autonomous vehicles, platoon systems require human drivers to take control during failures, leading to scenarios where vehicles are operated by drivers with diverse driving styles. This paper presents a novel multi-model approach for simultaneously identifying signal drop locations and driver attitudes in vehicular platoons using only tail vehicle measurements. The proposed method distinguishes between attentive and distracted driver behaviors by analyzing the propagation patterns of disturbances through the platoon system. Beyond its application in platooning, our methodology for detecting driver behavior using a multi-model approach provides a novel framework for human driver identification. To enhance computational efficiency for real-time applications, we introduce a blending-based identification method utilizing chosen models and weighted interpolation, significantly reducing the number of required models while maintaining detection accuracy. The effectiveness of our approach is validated through high-fidelity CarSim/Simulink environment simulations. Results demonstrate that the proposed method can accurately identify both the location of signal drops and the corresponding driver behavior. This approach minimizes the complexity and cost of fault detection while ensuring accuracy and reliability.
Paper Structure (38 sections, 85 equations, 11 figures, 1 table)

This paper contains 38 sections, 85 equations, 11 figures, 1 table.

Table of Contents

  1. Introduction
  2. Motivation
  3. Literature Review
  4. Contributions
  5. Problem Formulation
  6. System Dynamics
  7. Multi-Model Identification Framework
  8. The Proposed Fault Detection Approach
  9. Multi-Model Detection Method
  10. First-to-Last Transfer Function Analysis
  11. Transfer Function Analysis and Driver Behavior
  12. A Blending Multi-Model Approach
  13. Uniqueness of the Identification Solution
  14. Simulation Results
  15. Conclusion
  16. ...and 23 more sections

Figures (11)

  • Figure 1: One-dimensional platoon formation with a virtual reference vehicle and N following vehicles: a) Initial Formation b) After Failure.
  • Figure 2: Multi-model scheme for simultaneous identification of signal drop location and driver behavior in vehicular platoons
  • Figure 3: Proposed Blending-Based Fault Detection Approach.
  • Figure 4: (a) Platoon of Vehicles in CarSim (b) Platoon after Fault Occurrence.
  • Figure 5: Lead Vehicle Velocity Profile with Fault Occurrences.
  • ...and 6 more figures