Understanding User Requirements for Creating Sensor-Powered Smart Car Cabins Through Retrofitting

This paper investigates the potential of retrofitting to enhance sensor-powered smart car cabins by identifying limitations in built-in systems through interviews and defining user requirements via participatory design, ultimately offering design recommendations for future retrofit solutions.

The Gist

Imagine buying a new smartphone. You get it with a camera, a microphone, and a fingerprint scanner. But what if you hate the camera's location? What if you want a better microphone for your specific voice, or you want to take your favorite sensors and put them in your other car?

Right now, that's impossible. Car manufacturers install sensors (like cameras, microphones, and motion detectors) deep inside the car's dashboard and seats. They are hidden, permanent, and you can't change them. If you don't like how they work, you're stuck with them, or you have to buy a whole new car.

This paper is about a radical idea: What if we could "retrofit" our cars with our own sensors, just like we do with our smart homes?

Here is the breakdown of the research, explained simply with some creative analogies.

The Problem: The "One-Size-Fits-All" Car

The researchers talked to 18 drivers and passengers to find out what was annoying them about modern "smart" cars. They found five main headaches:

1. The "Hotel Room" Problem: You love the fatigue-detection system in your own car. But when you rent a car for a business trip, that feature is gone. You feel unsafe because your "digital safety blanket" didn't travel with you.
   • Analogy: It's like having a favorite pillow at home, but every hotel gives you a different, lumpy one. You can't bring your own pillow.
2. The "Bundled Junk" Problem: You buy a car with a voice assistant you hate, or a camera that watches you when you don't want it to. You can't turn it off or remove it; it's built into the car's DNA.
   • Analogy: It's like buying a TV that comes with a remote you can't throw away, even if you prefer using your phone to change channels.
3. The "Bad Angle" Problem: The car's microphone is in the front, but you are in the back seat with your kids. You have to shout to be heard. The sensors are placed for the "average" person, not for your specific family.
4. The "Outdated Tech" Problem: Car technology moves fast. A new sensor might come out next year that makes driving safer. But because your sensors are glued into the dashboard, you can't upgrade them without buying a new car.
5. The "Broken Toy" Problem: If a sensor breaks, you can't just swap it out like a lightbulb. You have to take the whole car to a mechanic, which is expensive and time-consuming.

The Solution: The "LEGO" Approach

The researchers propose Retrofitting. Instead of the car manufacturer deciding everything, you get to be the architect.

Imagine your car interior is a giant LEGO board.

• The Sensors: Instead of being hidden, they are little "blocks" (cameras, microphones, air quality sensors) that you can buy separately.
• The Placement: You can stick them anywhere you want using Velcro or magnets. Need a microphone closer to the back seat? Stick it there. Want a camera to watch the road but not your face? Move it.
• The Portability: When you switch cars (or go to a rental), you just pop your "LEGO blocks" out of your old car and stick them into the new one. Your smart experience travels with you.

What People Wanted (The User Study)

The researchers then held a workshop with 15 people, giving them 3D-printed fake sensors to play with inside a real car. They asked, "How would you build your perfect smart car?"

Here is what they learned:

• Don't make "All-in-One" bricks: People didn't want one giant box that did everything. They wanted separate, small sensors. Why? Because if one breaks, you only replace that one piece, not the whole system.
• Make it "Plug-and-Play": People were worried they wouldn't have the technical skills to install these. They wanted sensors that just "click" into place, like a USB drive, without needing wires or screwdrivers.
• The "Magic Map": When you have sensors everywhere, how do you know which one is which? Participants wanted a digital map on their phone or the car screen that shows exactly where every sensor is and what it's doing.
• The "Storage Case": When you take the sensors out of the car, where do they go? People imagined a special carrying case (like a charging case for wireless earbuds) that holds your sensors safely and tells you if you forgot to pack one.
• The "Privacy Switch": In a car with friends, one person might want a camera, but another might hate it. The system needs to be clear about who is watching whom, and let people easily turn off sensors they don't like.

The Future: The Self-Driving Car

The researchers also looked ahead to Autonomous Vehicles (AVs)—cars with no steering wheels.

• The "Living Room" on Wheels: In a self-driving car, you aren't driving. You might be eating, sleeping, or working. This means you might need different sensors, like ones that track your lunch or your sleep quality.
• The "Smart Home" Merge: The line between your house and your car will blur. You might bring your home's air-quality sensor into your self-driving car for a long commute. The car becomes just another room in your "smart ecosystem."

The Bottom Line

This paper argues that cars shouldn't be static boxes with fixed technology. They should be customizable, portable, and user-friendly.

By letting users "retrofit" their cars with their own sensors, we can:

1. Fix the "One-Size-Fits-All" issue by letting people choose what they need.
2. Save money by upgrading just the sensors, not the whole car.
3. Keep your "smart life" consistent whether you are in your own car, a rental, or a self-driving taxi.

It's about giving the power back to the driver and passenger, turning the car from a factory product into a personal, adaptable space.

Technical Summary

Here is a detailed technical summary of the paper "Understanding User Requirements for Creating Sensor-Powered Smart Car Cabins Through Retrofitting."

1. Problem Statement

Modern vehicles are increasingly equipped with manufacturer-installed "smart cabin" sensors (e.g., for climate control, gesture recognition, driver monitoring, and voice input). However, the authors identify a critical gap: these sensors are embedded, non-configurable, and lack portability. Users face five primary challenges with current built-in systems:

• Lack of Portability: Smart cabin experiences cannot be transferred between different vehicles (e.g., rental cars or family vehicles), forcing users to relearn interaction models.
• One-Size-Fits-All Limitations: Manufacturers bundle sensors that users may not need (e.g., mandatory voice assistants) while omitting features specific to individual needs.
• Inflexible Customization: Users cannot adjust sensor placement to suit specific roles (e.g., rear-seat passengers needing better microphone access) or personal preferences (e.g., left-handed gesture control).
• Upgrade Obstacles: Hardware upgrades are difficult, expensive, and require specialized tools, unlike software updates.
• Repair Difficulties: Embedded sensors are hard to access, repair, or disable if they malfunction or trigger false positives.

The paper proposes retrofitting (using aftermarket, user-selected sensors) as a solution to decouple smart cabin experiences from the vehicle chassis, allowing for personalization, portability, and easier maintenance.

2. Methodology

The authors conducted a two-phase mixed-method study involving 33 participants (18 in Phase 1, 15 in Phase 2) with diverse backgrounds (engineering, CS, and non-technical).

• Phase 1: Semi-Structured Interviews (18 Participants)

  • Goal: Identify challenges with built-in sensors and opportunities for retrofitting.
  • Procedure: Participants were introduced to common sensor types via 3D visualization and then engaged in scenario-based interviews covering four contexts: Rental Cars, Purchasing New Cars, Shared Family Cars, and Multi-Purpose (Ride-hailing) Cars.
  • Analysis: Open and axial coding were used to identify themes regarding user dissatisfaction and needs.

• Phase 2: Probe-Based Participatory Design (15 Participants)

  • Goal: Explore user expectations for retrofitting mechanisms and co-design solutions.
  • Procedure:
    • In-Car Familiarization: Participants used 3D-printed sensor mockups (cameras, microphones, radars, etc.) with Velcro attachments to physically place sensors on a real vehicle (2019 VW Arteon).
    • Co-Design Sessions: Using discussion cards, groups brainstormed solutions for sensor installation, removal, migration, and multi-user scenarios.
  • Analysis: Qualitative analysis of design activities and discussions to derive user requirements and design implications.

3. Key Contributions

1. Novel Conceptualization: Introduces retrofitting as a viable strategy to complement manufacturer-installed sensors, shifting the paradigm from fixed, embedded systems to modular, user-configurable ecosystems.
2. Empirical Insights: Provides a comprehensive taxonomy of user challenges with built-in sensors (Portability, Customization, Upgradability, Repairability) and maps specific retrofitting opportunities to address them.
3. Design Recommendations: Offers a set of actionable design implications for future smart cabin systems, focusing on modularity, installation workflows, and experience migration.

4. Key Results and Findings

A. User Challenges with Built-in Sensors (Phase 1)

• Portability (C1): Users desire "smart home" consistency across vehicles but cannot transfer their preferred sensor setups to rentals or other cars.
• Mismatched Needs (C2): Users are frustrated by sensors they cannot opt out of (e.g., two-handed steering detection) and the inability to add missing features.
• Lack of Customization (C3): Fixed sensor locations (e.g., microphones in the front) fail to serve rear passengers or specific user demographics (e.g., left-handed users).
• Hardware Lock-in (C4 & C5): Upgrading hardware is nearly impossible for end-users, and repairing embedded sensors requires dealership visits, unlike modular smart home devices.

B. User Expectations for Retrofitting (Phase 2)

• Modularity over Integration: Participants strongly preferred individual, standalone sensors over integrated "all-in-one" modules. They argued that single-unit modules limit placement flexibility and make partial upgrades/repairs difficult.
• Installation & Removal:
  • Ease: Installation must be "plug-and-play," requiring no tools or wiring.
  • Guidance: Users need visual cues (color coding, AR guidance) to ensure correct placement for optimal performance.
  • Feedback: Immediate system feedback is required to confirm successful installation or identify errors.
  • Removal: Users need clear software interfaces to identify which sensor corresponds to which function before removal, along with warnings about the consequences of disabling a sensor.
• Storage: Dedicated, organized storage cases (similar to earbud charging cases) are needed to prevent loss and track sensor inventory.
• Experience Migration: Users want to transfer their "feeling" of a smart cabin (e.g., comfort, efficiency) rather than just specific sensors. Systems should support selective migration (transferring only essential features) and "copy-paste" restoration of configurations.
• Multi-User Scenarios: In shared vehicles, mechanisms are needed to manage privacy, data ownership, and conflicting preferences when multiple users bring their own sensors.

C. Autonomous Vehicle (AV) Implications

• Flexibility: AVs allow sensor placement in areas previously restricted for driver visibility (e.g., windshields).
• New Activities: AVs enable activities like eating or napping, suggesting a need for sensors typically found in homes (e.g., dietary tracking) to be integrated into the cabin ecosystem.
• Dynamic Configurations: Sensor setups may change frequently based on the passenger group, requiring systems that are transparent and easily reconfigurable.

5. Significance and Design Implications

The paper argues that retrofitting transforms the car cabin from a static product into a dynamic, user-centric platform. The authors propose the following design guidelines for future development:

1. Prioritize Flexibility: Move away from integrated modules toward standalone, modular sensors that can be placed on various surfaces (ceiling, seats, doors) to suit ergonomic and privacy needs.
2. Simplify Installation Workflows: Develop non-invasive attachment mechanisms (e.g., magnetic, suction, electroadhesion) and intuitive guidance systems (AR, visual cues) to lower the technical barrier for users.
3. Enable Experience-Centric Migration: Design software that allows users to save and transfer "experiences" (configurations) rather than just hardware lists, supporting selective transfer and easy restoration.
4. Ensure Privacy and Transparency: In multi-user scenarios, provide clear visibility into which sensors are active, who owns the data, and how to disable specific sensors to respect privacy boundaries.
5. Bridge Ecosystems: Facilitate the integration of smart home sensors into AVs, creating a seamless ecosystem where users can carry their preferred sensing environment from home to car and between different vehicles.

Conclusion:
This research establishes a foundational understanding of how retrofitting can address the rigidity of current smart car cabins. It highlights that while technical challenges (power delivery, attachment stability) exist, the primary barrier is usability and design. By focusing on modularity, ease of use, and experience portability, the automotive industry can evolve toward a more adaptable and personalized smart mobility future.