WorldIndustrialWeekly General: Euro NCAP Pressure Pushes In-Cabin Safety From Steering Columns to Central Radar-Camera Systems

WorldIndustrialWeekly General — The cabin monitoring conversation reported from BARCELONA, Spain, is no longer about a single camera bolted to a steering column. It is about system architecture. Industrial readers should care because supplier design choices, sensing cost, and safety certification are converging in a way that resembles other capital-intensive product categories: the winning design is the one that satisfies five-star expectations without scattering multiple optical subsystems throughout the product.

Case context: AutoSens Europe 2026 and the shift to central sensing

At AutoSens Europe 2026, Davide Maggi, project officer at the European Commission’s Joint Research Centre (JRC), described automakers merging separate sensing tasks into unified central systems. The Joint Research Centre perspective matters because the same bodies that shape regulation are also watching how human-factors principles can be embedded into safety assessment. Maggi’s point was architectural: a central device can cover both the driver and occupants, while multiple dedicated cameras create cost and integration pressure.

Davide Maggi during a panel at AutoSens Europe 2026 (Source: AutoSens)

Steering-column monitoring gave way to central optical planning

For nearly a decade, driver distraction countermeasures were commonly implemented with optical sensors behind the steering wheel. That placement gives algorithms a direct view of eyes and face, which is useful for detecting blinking, gaze shifts, and fatigue-related behavior. Maggi told EE Times that the early driver monitoring field was largely built around that layout.

“Ten years ago, when we started seeing the first driver monitoring systems, they were all steering column-based—most of them, except for Tesla, which was the only one adding an RGB camera in the center top,” Maggi said.

Companies now want to streamline interior electronics and reduce component counts, so a dedicated steering-column-only architecture is losing ground. Rather than aiming one device only at the driver, the direction is toward centralized optical sensors near the rearview mirror or inside the center dashboard display, covering driver and passengers simultaneously.

“It seems like everyone is moving away from a driver monitoring system just facing the driver and pointing toward a camera in the center,” Maggi said. He added that a single centrally mounted device “can also provide both driver monitoring and occupant monitoring across the entire front and rear cabin.”

Data points and citations that anchor the change

• Conference basis: AutoSens Europe 2026, BARCELONA, Spain.
• Institutional basis: Davide Maggi, European Commission’s Joint Research Centre (JRC); also described as the Joint Research Centre, with the JRC referenced in the regulatory discussion.
• Rating basis: European New Car Assessment Programme, known as Euro NCAP, with expanded protocols for 2026 and further changes described for 2029.
• Engineering basis: a shift from a two-camera concept toward central camera and radar combinations.
• Interview source: Maggi told EE Times during coverage tied to AutoSens.
• Exception cited in the history: Tesla was described as the early outlier using an RGB camera in the center top.

Optical tradeoffs: central coverage is not free

Combining driver monitoring and occupant monitoring in one camera creates technical compromises. Once the optical sensor moves from the steering column to the vehicle center, the viewing angle becomes more oblique. Subtle facial movements become harder to track when the driver turns away. Maggi described the problem with a spatial metaphor: pulling coverage from one side can expose another.

“Having the camera up there and moving from a two-camera system—one for DMS and one for occupant monitoring—to one that does both has its own challenges,” Maggi said. “The blanket is this size: If you pull it from one side, some other part will go uncovered .”

Central mounting also makes optical algorithms vulnerable to line-of-sight obstruction. If the driver moves his head sideways, the system may lose accuracy in tracking where the driver is looking. “In this case, if you put the camera up there, whenever you lose accuracy in tracking where the driver is looking when he moves his head to this side, basically half of the face is gone,” Maggi said.

Everyday accessories can worsen the problem. “with the headwear, it could be getting tricky, because if you’re wearing a hat or anything, some blockage might occur .”

Physical limits remain for passengers hidden from view, including an infant sleeping under a blanket or positioned below the camera’s line of sight.

Hybrid architecture: central cameras paired with radar

To overcome the blind spots of one central camera without adding secondary optical assemblies, suppliers are pairing camera functions with in-cabin radar. Short-range millimeter-wave radar and ultra-wideband sensing can detect subtle motion, including the breathing pattern associated with a sleeping child, through fabrics and other materials. The architecture changes the role of the camera from sole witness to one sensor within a fused perception system.

“It’s interesting that now it’s going back to having one-size-fits-all from a technology side,” Maggi said about the move toward centralized cabin architectures.

“And the new part is that most of them—actually, I think all of them—are also embedding radar monitoring systems to detect what the camera cannot see .”

Pairing one central camera with radars lets Tier 1 automotive suppliers support both driver monitoring and child presence detection. At the conference, system developers demonstrated integration approaches that combined visual video feeds with real-time radar point-cloud data.

Maggi highlighted cost advantages for volume vehicle production. “I’ve seen some of them already providing a high-level vision combining camera and radar; others just provide a visual of what they get from the point-cloud data,” he said. “It’s fascinating. It’s supposed to be a lot cheaper than mounting cameras to monitor everything .”

Euro NCAP pressure shapes engineering choices

The acceleration toward merged in-cabin sensing is closely tied to updated European vehicle safety assessment criteria. Euro NCAP has introduced expanded protocols for 2026 and outlined further changes for 2029 that evaluate driver state monitoring, driver impairment detection, and child presence detection.

At the Joint Research Centre, Maggi’s work involves embedding human factors principles into European regulatory frameworks, and he emphasized how ratings influence engineering decisions.

“For sure, cost is one thing,” Maggi said. “I don’t know what influence that Euro NCAP has had in these choices, because now in Euro NCAP you can see that for 2026 and 2029 you have more requirements for driver monitoring and occupant monitoring especially, which was not there some years ago .”

Euro NCAP ratings matter to European car buyers when selecting new vehicles. That market influence can make automakers prioritize meeting test criteria efficiently over deploying elaborate, proprietary monitoring hardware.

“I don’t know if manufacturers now see that there’s no profitability in having a high-end driver monitoring system,” Maggi noted. “Because the only thing that matters is actually getting five stars on Euro NCAP.”

Regulatory boundary: assistance and automation are separated

European authorities are setting clearer regulatory boundaries between driver-assistance features and fully automated driving systems. Maggi said the separation is intended to prevent driver confusion about vehicle capabilities. European regulations strictly separate driver control assistance systems, or DCAS, from automated driving systems, or ADS.

“Regulatory-wise, the first thing that we tried to do is to separate levels of automation as they are defining a level, completely discarding the SAE [levels of driving automation] definition and really going into the details of what is an ADS and how it should be regulated, versus what is an assistant system,” Maggi explained.

Under these guidelines, automated systems cannot abruptly transfer control into driver-assistance modes. “DCAS regulation and ADS regulation are two very separate things,” Maggi said. He added that automated regulations specify “that there cannot be a transition from a Level 4 system to a Level 2 system, to avoid any more confusion happening with the driver not knowing the capability of the system in the vehicle .”

Common misconceptions and corrected readings

Misconception	What the AutoSens reporting indicates
Central sensing simply makes driver monitoring more reliable.	It can make the architecture simpler, but central mounting creates optical tradeoffs: more oblique views, line-of-sight issues, and sensitivity to headwear or blocked faces.
Radar is a secondary feature that only supports parking or collision warnings.	Here, radar is used inside the cabin to detect what the camera cannot see, including subtle breathing signals associated with a child hidden by fabric.
A high-end monitoring system is automatically the most commercially rational design.	The reporting links automaker behavior to efficient compliance, noting that what matters commercially may be getting five stars on Euro NCAP rather than building a more elaborate proprietary system.
Driver assistance and automated driving can move between each other as a product feature.	European regulation separates DCAS and ADS, and the discussion says there cannot be a transition from a Level 4 system to a Level 2 system, because it confuses the driver about system capability.

Action suggestions for product, operations, and supply-chain teams

1. Review sensing architecture as a product-level decision. For teams outside automotive, the lesson is that safety functions often become architecture decisions: centralize where it reduces cost and integration burden, but map every coverage gap before committing.
2. Check rating and protocol horizons. Treat the Euro NCAP discussion around 2026 and 2029 as a compliance signal that can move engineering priorities before final product launch dates.
3. Stress-test optical placement under realistic conditions. Ask suppliers how the camera handles oblique angles, head movement, headwear, blankets, and passengers outside the line of sight.
4. Ask for fused camera and radar evidence. If radar is added, expect demonstrations using point-cloud data and integration of visual and motion information, not only a spec-sheet mention.
5. Keep assistance and automation claims separate in documentation. Use regulatory language that distinguishes DCAS and ADS, and avoid implying transitions between levels that are not permitted.

FAQ

Why are steering-column cameras being replaced?

Steering-column cameras were a common early solution for driver monitoring because they gave a direct view of the driver’s face. The shift is happening because automakers want to reduce component counts and cover both driver and occupants from one central architecture, while still meeting five-star expectations.

What is the main optical risk of moving the camera to the center?

The angle becomes more oblique, so facial tracking can weaken when the driver turns. The system may also lose part of the face or be blocked by headwear. These are the technical tradeoffs behind central camera placement.

Why add radar if a camera is already installed?

Radar helps detect signals that a camera cannot see, especially when a person is partially covered or outside the camera’s line of sight. It can support child presence detection through motion and breathing-related detection.

How much does Euro NCAP influence the engineering change?

Cost matters, but Euro NCAP protocols for 2026 and 2029 add driver monitoring and occupant monitoring requirements that were not there some years ago. The market relevance of five stars gives automakers a reason to optimize for efficient compliance.

Does this mean automated driving and assistance will be merged?

No. The regulatory discussion described is the opposite: DCAS and ADS are separated, and automated regulations do not allow a transition from a Level 4 system to a Level 2 system that could confuse the driver.