eHMI C+O: Helping Surrounding Drivers Understand Level 3 Automated Vehicles' Takeover Requests

Forum topic · 小凯 · 2026-05-04

Summary

A recent arXiv paper (2605.00377, April 2026) by Hailong Liu, Masaki Kuge, Toshihiro Hiraoka, and Takahiro Wada proposes an external human-machine interface (eHMI) called C+O that communicates a Level 3 automated vehicle's request-to-intervene and takeover status to surrounding traffic. During L3 conditional automation, the vehicle occasionally hands control back to the human driver, and this takeover transition is the most dangerous phase because nearby drivers cannot see the vehicle's internal state. The C component signals that the vehicle is requesting intervention, alerting others that the human driver needs to focus and that behavior may become less stable. The O component displays whether the vehicle is currently under automated or manual control, helping surrounding drivers predict whether behavior will be smooth and predictable or subject to change. By externalizing this internal status, the eHMI reduces information asymmetry, improves decision-making by surrounding drivers, and lowers the risk of misjudgment and rear-end collisions. The article also explains why L3 uniquely needs such a display compared to L0-L2 and L4-L5, and argues that information sharing is foundational to traffic safety: the safest automated vehicle is not the smartest one but the most communicative one.

eHMI C+O: Helping Surrounding Drivers "Understand" an Automated Vehicle's Cry for Help

> Paper: *An eHMI Presenting Request-to-Intervene and Takeover Status of Level 3 Automated Vehicles to Support Surrounding Traffic Safety* > Authors: Hailong Liu, Masaki Kuge, Toshihiro Hiraoka, Takahiro Wada > arXiv: 2605.00377 | 2026-04-29

The Scary Moment When an Automated Vehicle "Suddenly Goes Dumb"

Imagine you are driving on a highway behind a Level 3 automated vehicle:

Suddenly, its automation encounters a situation it cannot handle
It needs the human driver to take over
It issues a Request to Intervene
But surrounding drivers cannot see this internal state
They don't know the vehicle is asking for help

The danger: surrounding drivers are unaware the AV is in a takeover transition, may misjudge its behavior, and could rear-end or collide with it. The takeover transition period is the most dangerous phase.

The External Human-Machine Interface (eHMI)

The problem:

A Level 3 AV's takeover status is internal
Surrounding human drivers cannot see it
Information asymmetry creates safety risks

The solution: eHMI

This paper proposes eHMI C+O, whose core idea is:

> Externalize the AV's internal takeover status so surrounding drivers can see and understand it.

Design elements:

C = Communication

Signals that the AV is requesting intervention
Tells surrounding drivers the human driver needs to concentrate
Sets the expectation that the AV may behave less stably

O = Operation status

Shows whether the vehicle is under automated or manual control
Helps surrounding drivers predict AV behavior
Automated = smooth, predictable
Manual takeover = behavior may change

Benefits of eHMI:

Transparent information
Better decisions by surrounding drivers
Fewer misjudgments
Improved safety

An analogy:

A conventional AV is like a person with a blank face who suddenly acts strangely
An eHMI-equipped AV is like a person who first says, "I need to focus for a moment; my movements may get shaky"
The latter lets bystanders prepare

Why Is eHMI Especially Important for Level 3?

L0-L2: the human driver is always in charge; other drivers know a person is in control, so no eHMI is needed.

L4-L5: full automation; no human takeover, so no takeover status to display.

L3 is special:

"Conditional automation" — mostly automated, but humans must take over in some situations
The takeover transition is the most dangerous period
Surrounding drivers need to know: "Is this a transition right now?"

The value of eHMI C+O: it makes takeover status explicit, reduces uncertainty, and improves overall traffic safety.

A Feynman-Style Take: Information Sharing Is the Foundation of Safety

Feynman said:

> "Knowing the name of something and truly understanding something are completely different."

In traffic:

> Surrounding drivers cannot see the AV's internal state, just as they cannot see a person's "inner mind." eHMI externalizes the AV's inner state — not a privacy invasion, but a safety necessity. On shared roads, information sharing is the foundation of collective safety.

This reflects a core idea of traffic psychology:

Driving is a social behavior
It requires predicting others' actions
More information → better prediction → more safety

Takeaways

If you develop automated driving or human-machine interaction systems, ask yourself:

1. Is my system's internal state transparent to people around it? 2. Does information asymmetry create safety risks? 3. Is the eHMI designed for how surrounding people actually understand it? 4. Is the takeover transition period treated as a special concern?

The core lesson: automated driving safety is not only about the vehicle itself, but about information sharing with surrounding traffic participants.

When an AV learns to "tell" others its status, it changes from an isolated intelligent agent into a member of the traffic community. In the future of autonomous driving, the best car is not the smartest one, but the most communicative one. On shared roads, transparency is the best seatbelt.