Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Learning from Disengagements: An Analysis of Safety Driver Interventions during Remote Driving

Ole Hans, Jürgen Adamy

TL;DR

This study analyzes Safety Driver interventions during Remote Driving System operation in a real-world urban ODD to understand how RD driving experience affects disengagements. Using data from over 14,000 km driven by 25 RDs, it characterizes 183 SD interventions, identifies four key disengagement scenarios, and demonstrates a pronounced learning curve where SD interventions drop sharply within the first 400 km and then plateau. The analysis employs multi-factor classification by experience, scenario type, and speed, with statistical tests (e.g., chi-square, Pearson and Spearman correlations) to show significant improvements in performance with experience. The findings inform targeted, experience-based RD training and system-design considerations to enhance safety, controllability, and efficiency of RDS in urban environments, and point to simulator-assisted training and hazard-focused assessment as avenues for future work.

Abstract

This study investigates disengagements of Remote Driving Systems (RDS) based on interventions by an in-vehicle Safety Drivers (SD) in real-world Operational Design Domains (ODD) with a focus on Remote Driver (RD) performance during their driving training. Based on an analysis of over 14,000 km on remote driving data, the relationship between the driving experience of 25 RD and the frequency of disengagements is systematically investigated. The results show that the number of SD interventions decreases significantly within the first 400 km of driving experience, which illustrates a clear learning curve of the RD. In addition, the most common causes for 183 disengagements analyzed are identified and categorized, whereby four main scenarios for SD interventions were identified and illustrated. The results emphasize the need for experience-based and targeted training programs aimed at developing basic driving skills early on, thereby increasing the safety, controllability and efficiency of RDS, especially in complex urban environment ODDs.

Learning from Disengagements: An Analysis of Safety Driver Interventions during Remote Driving

TL;DR

This study analyzes Safety Driver interventions during Remote Driving System operation in a real-world urban ODD to understand how RD driving experience affects disengagements. Using data from over 14,000 km driven by 25 RDs, it characterizes 183 SD interventions, identifies four key disengagement scenarios, and demonstrates a pronounced learning curve where SD interventions drop sharply within the first 400 km and then plateau. The analysis employs multi-factor classification by experience, scenario type, and speed, with statistical tests (e.g., chi-square, Pearson and Spearman correlations) to show significant improvements in performance with experience. The findings inform targeted, experience-based RD training and system-design considerations to enhance safety, controllability, and efficiency of RDS in urban environments, and point to simulator-assisted training and hazard-focused assessment as avenues for future work.

Abstract

This study investigates disengagements of Remote Driving Systems (RDS) based on interventions by an in-vehicle Safety Drivers (SD) in real-world Operational Design Domains (ODD) with a focus on Remote Driver (RD) performance during their driving training. Based on an analysis of over 14,000 km on remote driving data, the relationship between the driving experience of 25 RD and the frequency of disengagements is systematically investigated. The results show that the number of SD interventions decreases significantly within the first 400 km of driving experience, which illustrates a clear learning curve of the RD. In addition, the most common causes for 183 disengagements analyzed are identified and categorized, whereby four main scenarios for SD interventions were identified and illustrated. The results emphasize the need for experience-based and targeted training programs aimed at developing basic driving skills early on, thereby increasing the safety, controllability and efficiency of RDS, especially in complex urban environment ODDs.

Paper Structure

This paper contains 21 sections, 6 figures, 1 table.

Table of Contents

  1. Introduction
  2. Related Work
  3. Remote Driving
  4. Disengagements
  5. Remote Driving System Model
  6. Remote Driving System Human-Machine-Interface
  7. Remote Driving System Operational Design Domain
  8. Remote Driver Training
  9. Data Preparation and Filtering
  10. Evaluation
  11. Evaluation of average driving speed
  12. Number of Disengagements over Driving Experience
  13. Disengagement Classification
  14. Distribution of Disengagement Reasons
  15. Analysis of Disengagement Reasons
  16. ...and 6 more sections

Figures (6)

  • Figure 1: Vay Remote Control Station (RCS) within an operations center in Las Vegas, Nevada.
  • Figure 2: Human Machine Interface (HMI) of the Remote Control Station (RCS) showing a left turn maneuver.
  • Figure 3: Extract of the geographical part of the Remote Driving System (RDS) Operational Design Domain (ODD) and distribution of the remotely driven distance of more than 14,000 km in Las Vegas, Nevada, US.
  • Figure 4: Evolution of the count of Safety Driver (SD) intervention over remote driving experience.
  • Figure 5: Count per disengagement reasons for different remote driving experience levels.
  • ...and 1 more figures