Managing Driving Modes in Automated Driving Systems
archive: archived pipeline: cataloged verified
Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)
Summary
This paper addresses the critical challenge of managing driving modes in semiautomated vehicles (SAE Levels 3 and 4), which are projected to dominate roadways for decades before fully automated systems become viable. The core problem involves the "Request-to-Intervene" (RtI) decision, where an Automated Driving System (ADS) must determine when to transfer control to a human driver due to exceeding its Operational Design Domain (ODD). The authors identify a fundamental dilemma: the ADS must balance the risk of transferring control to a potentially distracted or unprepared driver against the risk of retaining control in conditions where its capabilities are insufficient. To resolve this, the study proposes an integrated modeling framework that simultaneously considers ODD supervision, environment monitoring, driver-state monitoring, trajectory planning, and driver-intervention performance assessment. The methodology leverages decision analysis and Bayesian forecasting to create probabilistic models for core ADS management functions. These models allow the system to explicitly represent state uncertainty and forecast future states based on sensor data collected up to time $t$. The framework includes specific modules for monitoring the environment and driver state, planning trajectories, and assessing the driver’s performance during interventions. By utilizing state-space models, the ADS can predict departures from safe operating conditions and schedule decisions several steps ahead. The authors develop a suite of control algorithms based on expected utility maximization to manage transitions between automated, manual, and emergency modes. These algorithms operate on a "management by exception" principle, issuing early warnings and RtI commands only when necessary, thereby minimizing unnecessary interruptions while ensuring safety. The efficacy of the proposed framework is evaluated through a simulated case study. The simulation tests the algorithms' ability to reason about uncertain environments and output deterministic best responses for mode transitions. The results demonstrate that the Bayesian approach effectively reduces error propagation in ADS operations by accounting for dynamic features of the driving environment. The empirical evaluation highlights specific tradeoffs associated with control algorithm parameterization, such as the timing of warnings relative to driver reaction times. Furthermore, the study uncovers decision-support dilemmas that require further ethical and regulatory analysis, particularly regarding the optimal balance between proactive caution and operational efficiency. The significance of this work lies in its comprehensive approach to RtI management, moving beyond isolated studies of human-machine interfaces to provide a unified statistical and decision-theoretic foundation. By integrating multiple monitoring and assessment modules, the framework offers a robust method for handling the heterogeneous traffic mix expected in the coming decades. The findings provide actionable insights for ADS developers on how to implement probabilistic models for safer mode transitions. Additionally, the paper underscores the need for continued research into the ethical implications of automated decision-making in safety-critical scenarios, contributing to the broader discourse on the deployment of semiautomated transportation systems.
Provenance
The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | Crossref | — | — | 1 | 2026-08-09 |
| archive | success | unpaywall | — | — | 2 | 2026-08-09 |
| extract | success | pdftotext | — | — | 4 | 2026-08-10 |
| clean | success | clean | — | — | 2 | 2026-08-10 |
| chunk | success | chunk | — | — | 2 | 2026-08-10 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 2 | 2026-08-10 |
| promote | success | — | — | — | 1 | 2026-08-09 |
| summarize | success | llm | qwen3.6-27b-nvidia | summ-v5 | 2 | 2026-08-10 |
| tag | success | vector_similarity | — | — | 17 | 2026-08-11 |
| verify | partial | — | — | — | 2 | 2026-08-10 |
Summary generated by qwen3.6-27b-nvidia on 2026-08-10; verification: verified_with_issues.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.
Information type
What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).
- Theoretical Contribution: computational model, conceptual framework, theory or model