Tag: Advanced Research

 

Every year, 1.3 million people die in road crashes around the world. More than 20% of these fatalities are estimated to be alcohol related. For as long as cars have existed, drunk driving has been a complex problem in need of comprehensive solutions.  

From ignition interlocks to roadside alcohol tests – different strategies for mitigating drunk driving have been deployed for over a century. Yet none of them have managed to effectively prevent a still very large number of lives lost to accidents caused by intoxicated driving.  

Critical research is being conducted by government bodies, the automotive industry, technology companies and academia to determine more effective approaches. On June 13, 2022, Smart Eye gathered industry experts and leading researchers for a live panel discussion on the current state of intoxicated driving research. You can access the recording of that discussion here: https://www.www.smarteye.se/blogs/advancing-road-safety-the-state-of-alcohol-intoxication-research/  

These are our top five insights from the discussion: 

1. In-vehicle sensors may be the best alternative to challenge today’s 100-year-old standard

Today’s prevalent measure to determine alcohol intoxication – Blood Alcohol Concentration (BAC) – was developed nearly 100 years ago. Despite the major societal and technological changes the world has undergone since then, our current legislation and policies are still tied to this century-old standard.  

To this day, breathalyzers and other traditional methods for detecting alcohol intoxication remain important but are insufficient for effectively improving road safety. Roadside tests are dependent on limited police resources, while a mandated implementation of alcolocks in vehicles has been met with resistance from car manufacturers and car owners alike.  

In order to work around the limitations of these traditional methods, researchers are exploring new ways to detect impaired and drunk driving. Through in-vehicle systems that use cameras and other sensors for analyzing the driver’s physical and mental state, intoxication can be detected using technology that is already installed in today’s cars – such as Driver Monitoring Systems (DMS). Rather than measuring exact blood alcohol level, these systems would combine the output from multiple sensors and, by fusing the data, offer a comprehensive assessment of whether a driver is fit to drive or not.   

2. Identifying the reasons behind driver impairment may not be necessary to improve road safety

In-vehicle intoxication detection technologies, using multiple sensors and machine learning, are still in the early phases of development. It will take extensive research and data collection before they are able to accurately detect alcohol intoxication every time. But until then, these systems can still be effective in preventing drunk driving.  

Driver drowsiness and driver distraction can be as much of a traffic risk as an intoxicated driver. While we want in-vehicle sensors to be able to differentiate between different types of driving impairment in the long-term, it may not be absolutely necessary at this early stage. As long as the technology can indicate that the driver is in a less-than-optimal state, we can take measures to improve road safety.  

Once an inattentive, possibly intoxicated, driver is detected, the system can choose to activate a range of driver support functions, like autobrake or lane support systems, at an earlier point than it normally would. As the technology evolves and becomes more and more reliable, the system would be able to interfere in more drastic ways – such as preventing the vehicle from starting with an undisputedly drunk driver behind the wheel.  

3. The core technology needed to detect intoxicated driving is already being installed in cars all over the world 

In-vehicle systems with the ability to detect an unfit driver is in no way a new idea. In fact, this technology is already installed in millions of cars on the roads today. Driven by regulation and legislation, the implementation of Driver Monitoring Systems (DMS), developed for identifying distraction or drowsiness in drivers, will soon be mandatory in new vehicles all over the world.  

Advanced, camera-based Driver Monitoring Systems are fast becoming more and more common in everything from personal cars to trucks and buses. By adding new algorithms and features to existing systems in vehicles, it will soon be possible to also use these systems to detect alcohol intoxication and other types of impairment. This is bound to drastically speed up the process of getting intoxication detection software into vehicles everywhere.    

4. Diverse data is essential for developing effective intoxication detection systems 

Future intoxication detection systems that use computer vision and machine learning will need to be developed using not only very large amounts of data – but also very diverse data. To avoid data and algorithmic bias, it is critically important that we train and validate these algorithms with equal representation of different genders, ethnicities, age groups,  and other ways people present themselves.  

When developing intoxication detection systems even more consideration is needed. There are certain medical conditions or medications that can cause people to appear intoxicated even if they aren’t. In the future, intoxication detection systems will need to be able to identify this to avoid the risk of bias against people with illness. This requires the developers of these technologies to gather data from a broad swath of conditions and merge that data with other sensor inputs, to come to an accurate understanding of a driver’s condition.   

 5. Why intoxicated driving is exceptionally difficult to study – and incredibly important

The more advanced in-vehicle technologies for detecting driver impairment are based on machine learning and computer vision. As with any AI-based technology, large amounts of data are key for training the algorithms and improving performance. As important as this process is, collecting data of drunk drivers in moving vehicles is a very complicated task.  

In an on-going study called Fit 2 Drive, Smart Eye and the Swedish National Road and Transport Research Institute (VTI) have so far been able to collect data from more than 30 participants while gradually increasing their blood alcohol concentration. At the same time, the participants drove a car equipped with multiple sensors on an enclosed race track. This is the first collection of data to offer visual information from the driver, giving valuable insight into how different people behave at different levels of intoxication.  

In Sweden, driving while intoxicated is illegal even in very controlled situations on an enclosed test track. In order to conduct the study, Smart Eye and VTI had to get special permission from the Swedish government. But despite the complicated processes of collecting data from drunk drivers, these types of studies are absolutely necessary for future research on driver impairment.  

To learn more about how Smart Eye and VTI managed to collect data from intoxicated drivers, watch the video below: 

And, if you would like to listen to the panel discussion in its entirety, you can access that recording here: https://www.www.smarteye.se/blogs/advancing-road-safety-the-state-of-alcohol-intoxication-research/  

What makes a good driver?

We can all improve our driving skills in different ways. But what happens when you put a driver into a situation far beyond what most of us are used to? One way to answer this question is to examine high-performance driving.

Driving at the level of speed that is considered normal on a racetrack can only be described as extreme driving conditions to everyday drivers. High-performance drivers are constantly put into scenarios where they must make complex decisions quickly and with very little information. To add even more pressure to the driver in this situation, these decisions have huge consequences, whether they result in loss of time or, even worse, a wrecked vehicle and an injured driver.

According to racecar driving instructors, the main bottleneck to going faster while maintaining safety lies in the driver’s ability to effectively perceive events, make snap decisions, and aptly control the vehicle. This not only requires years of training, but focused training that shapes the driver’s cognitive skills in a very specific way.

But how do trainers target the correct events and experiences to shape the driver’s cognitive skills, like anticipation and attention, to decrease the delay between perception, decision and response? And how can researchers effectively measure the learning process in drivers as they learn?

In this study, we put Smart Eye’s eye tracking technology and iMotions’s biometric analysis platform to the ultimate test: turning the physiological signals of race car drivers in extreme conditions into powerful insights that help them become better drivers.

The Bondurant study: How does our technology hold up under extreme conditions? 

In September 2020, Smart Eye’s Senior Sales Engineer Aaron Galbraith traveled to Arizona for a pilot data collection with a performance driving instructor while on site at the Bondurant High Performance Driving School.

The Bondurant High Performance Driving school (now renamed the Radford Racing school) specializes in instruction for high-performance driving but is also a way for the everyday driver to enhance their skills. Located in Chandler, Arizona, the racetrack contains multiple challenging driving situations in a desert environment.

To get the desired results out of the study, the client had defined a number of requirements:

1. Smart Eye’s system had to be able to determine whether the student driver was looking out the windshield, out the driver’s side window, or out the passenger side window.
2. The system also had to have the ability to record and replay where the student driver was looking during difficult turns and maneuvers. This would let the trainer provide feedback to help the student understand what they were doing wrong and assist the student in correcting bad driving behavior on the spot.
3. After the training session, the client wanted to be able to obtain video of the drivers’ faces to include in their report.
4. Lastly, the client wanted to be able to tie the vehicle data, including acceleration, braking, steer angle sensors and more, to other biometric data, like gaze or emotion, to compare how the student was feeling to how the vehicle was handled in that moment.

Collecting data in a race car: Unique challenges and creative solutions 

To provide our client with the most precise eye tracking data possible, we used our most advanced remote eye tracker: Smart Eye Pro.

Smart Eye Pro features the best combined head box, field of view and gaze accuracy on the market. It is also an incredibly flexible system that is known to deliver exact results no matter the circumstances. But would it be able to hold up in conditions as extreme as on the Bondurant High Performance Driving School racetrack? On site, we were faced with some unique challenges that really put Smart Eye Pro to the test:

Mounting the cameras 

Smart Eye Pro uses multiple cameras that are freely placed in the vehicle’s interior. However, in this study, the training instructors weren’t allowed to make any permanent changes to the vehicle. For example, this meant the cameras could not be mounted to the vehicle by drilling holes in the dashboard or the console area. The solution became two-sided tape, which was used to stick the cameras to the vehicle’s interior surfaces. While not as foolproof as a solid mounting solution, the two-sided tape actually held up incredibly well considering the high amount of stress the system was being put under.

In this situation, the flexibility of the system was of great benefit as the Smart Eye Pro cameras can be remotely placed in any environment. The system is also non-intrusive, which means subjects don’t have to wear any type of glasses or head gear to be tracked. Because of this, the cameras could be repositioned to find the best possible view of the driver’s head and eyes, while ensuring a very realistic driving experience for the driver.

Vehicle vibration 

It’s hard to find a more challenging environment for a system to be placed in than a race car. Switching between extreme acceleration, deceleration, braking and cornering – the vehicle is literally being put through the paces. Given all these potential problems, we were very pleased to see how well Smart Eye Pro’s camera calibration held up.

The Bondurant study showed us that even though Smart Eye Pro is an advanced research-grade system, it isn’t limited to static research environments, like a lab or stationary vehicle simulator.  

The number of cameras 

How many cameras would be required to capture the driver’s gaze?

One of the benefits of Smart Eye Pro is that it’s scalable. This means the number of cameras required depends on each, specific use case.

For this study, we decided to use Smart Eye’s traditional automotive set-up: three cameras on the dashboard and one camera next to the center console. This provides very good coverage on the windshields, side mirrors, instrument cluster, and center console. However, we could have obtained similar gaze tracking results by reducing the number of cameras to the three on the dashboard, since Bondurant was not interested in tracking gaze on the center console region.

Dramatic differences in lighting 

Smart Eye Pro can operate in any lighting condition. Since the system only uses the light produced from its flashes, external or ambient light is generally not a problem.

But while on the Bondurant track, the difference in lighting between the interior and exterior of the vehicle was so extreme we were forced to get creative.

The original plan was to use an over-the-shoulder scene camera, but because of the contrast in lighting, this wasn’t possible. Instead, we repurposed the camera as a forward-facing responding camera in iMotions, and we were then able to use the Affectiva database in iMotions to detect the driver’s emotions. But this is something we will get back to in a little while.

Extreme temperatures 

During our visit, the Arizona August sun was creating temperatures reaching as high as 119 degrees Fahrenheit (48 degrees Celsius). Even so, the two-sided tape mounting solution held up from one day to the next. And since the vehicle cabin was air conditioned, the interior temperature was kept well within normal operating boundaries for the system.

Data analysis: How to actually generate insight from the data gathered in the field 

Despite challenging circumstances, we were able to collect very valuable data from the Bondurant racetrack. But how do we analyze this data to gain powerful insights from the training session? 

To find out how drivers’ biometrics can help us understand in which ways signals like attention control and emotional reactions affect driving, keep reading on the Affectiva blog.  

Are you curious about Smart Eye Pro, the iMotions platform or Affectiva’s Emotion AI? Click on the links to find out more or order a demo of the products:

Smart Eye Pro: https://www.www.smarteye.se/research-instruments/se-pro/

The iMotions platform: https://imotions.com/platform/

Affectiva’s Emotion AI: https://www.affectiva.com/experience-it/