Delivery Director in Automotive at Intellias
In 2020, sophisticated-car infotainment system designs are no longer impressive. Smartphone integration, voice assistance, recognition of driver distraction, assisted personalized driving, connected in-car payments can be easily taken for granted. With the rapid growth of automotive technologies, OEMs are being challenged to create HMIs that offer usability and ensure safety while standing out from the competition. Still, there are general rules that can be helpful to those interested in designing an HMI for connected cars and who are approaching this topic for the first time.
Since the main task of an HMI designer is to create a product that meets user requirements, profound research is the first step of any HMI development lifecycle. You can collect information on users with the help of different methods and tools: reviews on similar products, surveys, field studies, questionnaires, and more.
When conducting research, it is critical not only to define the drivers’ needs but also to identify their mental models — their understanding of what the system contains, how it works, and why it works this way. They help users automate interaction while reducing cognitive resources.
If mental models coincide with an infotainment system design, it becomes more efficient and the driver won’t be distracted from the road.
It is especially important in the car-sharing context when a driver doesn’t have much time to get accustomed to the system.
User observations (watching and listening to users as they interact with different HMIs) can help identify these implicit constructs. However, to receive genuine feedback, you need to ensure the testing environment is as close to real conditions as possible. For example, you may consider removing the researcher from the vehicle and recruiting a family member to sit in. This can make the driver’s reactions more natural.
Once the research is collected, it’s time to create key user personas — fictional, generalized representations of potential customers. This approach will help to customize and personalize each product.
Automotive HMI experts differentiate three criteria to evaluate the cognitive ergonomic quality of an HMI in automotive: usability, distraction, and user acceptance.
The task of HMI designers is to ensure high usability, low distraction, and high user acceptance. And the HMI design principles listed below can help achieve the perfect balance of these criteria.
Being particular when choosing features for an in-car infotainment system is critical for high user acceptance rates. All HMIs contain the following elements:
It’s important to first focus on the features that ensure safety and the successful completion of the driver’s primary task: to get from point A to point B. Only after that, you should proceed with additional functionality.
Notably, some of the features that were once considered “additional” are now becoming must-haves. For example, making/receiving calls, navigation, and music/podcasts/audiobooks are insisted on by drivers.
The challenge is in the balance between functional complexity and simplicity of user interaction. Too many features not only decrease user acceptance rates but also negatively affect usability.
Conduct additional user research to define what to add and what to omit.
A lot of text on displays may distract the user from driving, which means that no label should go beyond one line. This rule, however, doesn’t apply to text that should be read in non-motion situations.
In any case, all text on screens should be precise and comprehensive.
It will help users to be 100% certain of what to do next, which particularly applies to critical situations that require immediate action. Visualization and graphical representation drastically improve the ease of use. The icons replacing labels can be a simple solution, but making apprehensible and straightforward icons is an art.
Also, when “drafting” the design of a head unit and other screens, don’t underestimate the font size. Given that the average distance from the eye to the main display is about 60 cm/24 inches, the font size shouldn’t be less than 5.3/6 mm.
Speaking of interaction with the HMI, it is critical to minimize the number of “taps” required to complete each task. A minimal amount of calls to action can decrease distraction and prevent motion sickness, which is more likely to happen when a driver attempts to perform too many tasks while driving.
With good designed HMI, a brief glance is enough for users to understand what needs to be done. According to the (American) National Highway Traffic Safety Administration, a driver should be able to perform the required actions in a series of eight, 1.5-second glances.
To ensure each action is completed successfully, it is also crucial to choose interaction types that are easier to perform over more complex ones. For example, single tap options, scrolling, flicking left/right/up/down, etc. are less risky. When it comes to scrolling, you should avoid the free-scrolling function: all scrolling actions should have a snapping effect.
On top of that, it is necessary to balance hard and soft interactions. Hard interactions are actions deliberately performed by a driver, while soft interactions are those performed by the system without input. The right combination of interactions can decrease the distractions of drivers significantly.
As drivers interact with the HMI, they usually have tree ways to input data: voice commands, request typing, or handwriting. To make sure all variants work in a perfect way, it’s necessary to eliminate obstacles that may prevent data recognition.
Ambient noise levels while driving may affect speech interpretation. When it comes to typing requests, vehicle vibration, speed limits, high level of distraction, and low level of input convenience for a left-handed driver make it difficult to hit the keyboard. As for handwriting, it seems to have less obstacles than other methods, especially if implemented correctly. Today, it’s possible to choose destination, change car settings, and even switch over favorite songs with the twist of a finger.
With the natural language processing and convolutional neural network, handwriting recognition is possible within the transcription of finger-drawn words into a digital text. Moreover, neural networks help to implement a multi-character recognition (MCR) model for more accurate interpretation. Classical machine learning approaches also allow to apply word suggestion and word correction models that improve word detection.
Despite today’s level of technology in the automotive industry, the amount of attention a driver gives to the road should be just as high as driving manually. The challenge is, the higher the reliability is on a system, the lower the level of a driver’s attention.
“It is impossible for even a highly motivated human being to maintain effective visual attention towards a source of information on which very little happens, for more than about half an hour.”
This challenge should be greatly considered when designing an automotive HMI. Luckily, modern technologies offer multiple strategies for counteracting the effects of being unobtrusively out of control. For example, the Mercedes E allows keeping hands off the wheel for about 30 seconds before it displays red hands on the steering icon. If no action taken, the system starts using sound warnings, which become more intrusive while the car reduces speed and eventually stops.
In terms of safety in the automotive context, dark mode is preferable as it reduces distraction. However, it doesn’t have to be the only option. The light mode improves readability in many cases, such as using the navigation map. Besides, people with astigmatism find it harder to discern white text on a black background.
When it comes to differing color elements on screens, prioritize safety over aesthetics. Subtle tones are not so important. Instead, “traditional” colors will ensure intuitive user experience, while high contrasts will make text more visible. These rules should particularly be taken into account when considering a head-unit display design.
With the advent of autonomous vehicles and connected cars, stricter requirements are being imposed on the design of human-machine interfaces. To stand out among the competition, modern HMIs should have intuitive navigation, not distract from driving, and offer more value than similar products.
Though it sounds complicated, the combination of profound user research, a balanced approach to functionality, compliance with design principles, and other suggested tips are vital in creating the perfect HMI for car-sharing solutions and connected cars.