Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Bachelor/Master Thesis: Development of a ML algorithm for the detection of driving under the influence of alcohol
In Switzerland, official reports state that 6.5 % of all car accidents are connected to driving under the influence of alcohol. In this thesis, you will explore this topic and develop an approach to mitigate those risks by applying machine learning (ML) models on an unique dataset.
The improvement of vehicle technologies has equipped cars with an intelligence that enables many new applications. Accordingly, car driving has never been safer, yearly reports show a continuous decrease in accident reports. Many contribute these improvements to new driving assistance systems and higher safety standards that were recently developed.
However, alcohol remains a major accident source and estimates see even higher numbers than the ones officially reported. Yet, there is no technology in sight to mitigate these risks. It is therefore worth investigating how driving under the influence of alcohol can be prevented. In your thesis, you will use an unique dataset collected by us in a driving simulator to develop a drunk driving detection algorithm using state-of-the-art machine learning techniques.
We already collected a multisensor dataset with 30 drivers that is during your thesis available. The participants conducted in several scenarios driving tasks drunk and sober. The following sensors are available (we search currently only students for the ones in bold):
- **Vehicle data**
- **Eye tracking data**
- Audio data
- Video data
- Vital sensor data
The improvement of vehicle technologies has equipped cars with an intelligence that enables many new applications. Accordingly, car driving has never been safer, yearly reports show a continuous decrease in accident reports. Many contribute these improvements to new driving assistance systems and higher safety standards that were recently developed.
However, alcohol remains a major accident source and estimates see even higher numbers than the ones officially reported. Yet, there is no technology in sight to mitigate these risks. It is therefore worth investigating how driving under the influence of alcohol can be prevented. In your thesis, you will use an unique dataset collected by us in a driving simulator to develop a drunk driving detection algorithm using state-of-the-art machine learning techniques.
We already collected a multisensor dataset with 30 drivers that is during your thesis available. The participants conducted in several scenarios driving tasks drunk and sober. The following sensors are available (we search currently only students for the ones in bold):
- **Vehicle data**
- **Eye tracking data**
- Audio data
- Video data
- Vital sensor data
We are interested in understanding how we can most accurately and understandable detect that a driver is driving drunk. Therefore, you should use state-of-the-art machine learning approaches to build robust classifieres. You should start with focussing on conventional approaches (e.g., logistic regression, tree-based models) before using more advanced neuronal networks (e.g., RNN or CNN models). Furthermore, we expect you to explain why your model outperforms other approaches and thus you should use explainable AI approaches (e.g., statistical approaches or SHAP values) to interpret your classifier.
During the thesis, you will work closely with us and will receive a dedicated supervision. We are highly interested to invest time in your thesis as the topic above counts into our research.
We are interested in understanding how we can most accurately and understandable detect that a driver is driving drunk. Therefore, you should use state-of-the-art machine learning approaches to build robust classifieres. You should start with focussing on conventional approaches (e.g., logistic regression, tree-based models) before using more advanced neuronal networks (e.g., RNN or CNN models). Furthermore, we expect you to explain why your model outperforms other approaches and thus you should use explainable AI approaches (e.g., statistical approaches or SHAP values) to interpret your classifier.
During the thesis, you will work closely with us and will receive a dedicated supervision. We are highly interested to invest time in your thesis as the topic above counts into our research.
Kevin Koch (kevinkoch@ethz.ch).
Please contact us with your CV, a short statement of motivation, and your current transcripts of records (bachelor & master).
Kevin Koch (kevinkoch@ethz.ch).
Please contact us with your CV, a short statement of motivation, and your current transcripts of records (bachelor & master).