Cooperation between driver and automated driving system: Implementation and evaluation

Guo, Chunshi; Sentouh, Chouki; Popieul, Jean‐Christophe; Haué, Jean-Baptiste; Langlois, Sabine; Loeillet, Jean-Jacques; Soualmi, Boussaad; That, Thomas Nguyen · 2017 · OpenAlex-citations

DOI: 10.1016/j.trf.2017.04.006

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 challenge of shared control authority between human drivers and automated driving systems (ADS), specifically focusing on SAE Level 3 automation. While existing research often prioritizes takeover scenarios, this study proposes a "driver-vehicle cooperation" paradigm to manage interference during complex driving tasks. The authors aim to improve driving performance by exploiting human-automation synergy, allowing drivers to intervene when they disagree with the system’s conservative or socially inadequate maneuvers. The specific use case selected for implementation and evaluation is highway merging management, a scenario characterized by strong social interactions and potential conflicts between merging and mainline vehicles. The study implements a "deliberative maneuver cooperation" principle using a hierarchical finite state machine (HFSM) for maneuver planning. In this framework, the ADS displays its intended maneuver (pass or yield) and plausible alternatives to the driver via a Head-Up Display (HUD) and simulated augmented reality. The driver can select an alternative using capacitive buttons if the system’s risk assessment permits. The system retains final authority, executing the driver’s choice only if it remains feasible. To evaluate this design, the authors conducted a user study with 22 participants using a driving simulator. The experiment employed a two-phase procedure: Phase 1 assessed the intuitiveness of the Human-Machine Interface (HMI) and cooperation logic without prior instruction, while Phase 2 evaluated cooperation performance and driving metrics after instruction. Scenarios varied by traffic density (fluid vs. congested) and merging behavior (nominal, hesitant, forced). Results indicate mixed success regarding interface intuitiveness. The simulated augmented reality tracking the merging vehicle was rated as the easiest element to understand, while the HUD symbols and button color codes caused confusion, with many participants misinterpreting arrows as acceleration indicators rather than maneuver intentions. In Phase 1, only 21% of button presses in fluid traffic correctly selected an available alternative (PAA), whereas 50% pressed the button for the system’s intended maneuver, suggesting users expected confirmation rather than intervention. However, comprehension improved in congested traffic (53% PAA). In Phase 2, where users were instructed, the study measured driving performance metrics such as interaction duration, speed maintenance, and acceleration variation. The findings demonstrate that driving simulation is effective for prototyping and evaluating interaction designs for ADS, providing data on both user perception and system performance. The significance of this work lies in its contribution to the design of cooperative automation systems that share authority with drivers. By validating a specific cooperation principle and HMI design through empirical testing, the paper highlights the importance of clear visual feedback and intuitive command structures in human-automation interaction. The study underscores that while the cooperation paradigm offers potential benefits for handling complex social driving situations, the HMI must be carefully designed to ensure users correctly interpret system intentions and available alternatives. This approach supports the development of more usable and socially adaptive automated driving 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.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-25; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• automation
• automation surprise
• situational awareness
• in vehicle coaching
• ehmi external hmi
• odd communication

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).

• Applied Guidance: design guidelines
• Theoretical Contribution: conceptual framework, computational model