Master's thesis in collaboration with Smart Eye AB

Smart Eye is the global leader in Human Insight AI, technology that understands, supports and predicts human behavior in complex environments. Together with our subsidiaries Affectiva and iMotions, we are bridging the gap between humans and machines for a safe and sustainable future. Today, our technology is embedded in next-generation vehicles, leading the way towards human-centric mobility through Driver Monitoring Systems and Interior Sensing solutions. Our Research Instruments offer unparalleled insights into automotive, aviation, assistive technology, behavioral science and many more fields. Our subsidiary Affectiva is humanizing technology by pioneering Emotion AI, helping companies gain a deeper understanding of how consumers engage with their content, products and services, in automotive, media & entertainment, market research and beyond. Our subsidiary iMotions provides the world’s leading biosensor software platform, that synchronizes data streams in real time from multiple sensors. Smart Eye was founded in 1999 and is headquartered in Sweden with offices in the US, UK, Germany, Denmark, Egypt, Japan, Singapore and China. A publicly traded company since 2016, our customers include NASA, Nissan, Boeing, Honeywell, Volvo, GM, BMW, Geely, Harvard University, over 1,300 research organizations around the world, 70% of the world’s largest advertisers and 28% of the Fortune Global 500 companies. Visit www.smarteye.ai for more information. Smart Eye is leading the global market for Driver Monitoring Systems. Monitoring the driver behavior is safety critical both for normal driving as well as for autonomous driving. We are now offering a master's thesis collaboration regarding context-aware driver state monitoring in real traffic. Problem Description Driver monitoring systems (DMS) are becoming standard equipment in new cars, and in July 2026 they became mandatory in all new cars in the European Union. Most current functions judge the driver largely in isolation: eyes off the road for too long, or eyelids closing too often, triggers a warning regardless of what happens outside the vehicle. This may be too coarse. A two-second glance away is probably acceptable while the car is standing in a queue, but potentially critical when approaching an intersection. Driving situations differ in how much they require of the driver: how much is going on around the vehicle, how quickly it can change, and how much margin there is for error. Here we call this driving demand. The relevant question is then not only whether the driver is looking at the road, but whether their attention matches the driving demand of the situation. Aim and research questions The aim is to make driver monitoring context-aware by relating driver attention to the driving situation. Using Smart Eye driver monitoring data together with vehicle, map and traffic data, the thesis will estimate the driving demand of the current and upcoming situation, describe how the glance behavior of attentive drivers varies with the driving context, and derive what this implies for when and how an adaptive HMI should intervene. The first objective is to develop a model for estimating the current and upcoming driving demand from vehicle, map and surrounding traffic data. The first objective is to develop a model for estimating the current and upcoming driving demand from vehicle, map and surrounding traffic data.

• RQ1: Which context features best explain situational driving demand, and how well can driving demand be predicted a few seconds ahead?

The second objective is to describe how normal attention patterns depend on context.

• RQ2: How do glance behavior and eyes-off-road durations vary with road type, speed, traffic and assistance-system status in attentive driving?

The third objective is to translate the findings into input for adaptive interaction.

• RQ3: What do the results imply for when and how an adaptive HMI should intervene?

Smart Eye has a test vehicle with their driver monitoring cameras and other sensors installed. The thesis will use this vehicle to collect on-road data and study driver state in the context of the driving situation. The scope may need to be narrowed after the first data review. The thesis is part of AutHUMate, a Horizon Europe project on safe human–technology interaction in the vehicles of the coming decade. In this master’s thesis project stakeholders are Smart Eye and Chalmers. A remuneration will be paid out to the student from Smart Eye after the final thesis has been approved. Location: Smart Eye AB, Headquarters (Gothenburg) Must-have-skills

• Programming skills in Python, R or MATLAB. Experience in data processing, data analysis

and statistical methods is advantageous. Who are you? We welcome all types of personalities and believe in a true mix of people to create a dynamic work environment. We are open to single applicants or pairs of two Master thesis students.

• Ability to work independently in an environment that changes rapidly.
• A positive, enthusiastic, open, entrepreneurial attitude and a willingness to work closely with customers as well as other functions within Smart Eye to achieve the growth.
• Strong analytical, creative skills. Independent, Self-driven, Thorough, Curious and with a strong interest in new technologies.
• Communication skills with fluency in both Swedish and English. Other languages are a merit.

Interviews are held on a continuous basis, so we highly recommend that you submit your application at your earliest convenience. Welcome with your application. Contact For Questions Jordanka Kovaceva, jordanka.kovaceva@chalmers.se and Martin Due, martin.due@smarteye.se