Innovative Machine Learning Algorithm meets Carrera for auto piloting
Results of this project are explained in the final report and a short demo video was made by the students, see video below:
- Sponsored by: PwC
- Project Lead: Dr. Ricardo Acevedo Cabra
- Scientific Lead: M.Sc. Oliver Kobsik, M. Sc. Stephan Bautz, M.Sc. Philipp Düpree, B.A. Manuel Kuhlin.
- TUM Co-Mentor: Cristina Cipriani
- Term: Winter semester 2022
The PwC network spans the globe and employs nearly 300,000 people. With 21 locations and 13.000 employees in Germany alone, we are also well-established in this country while we are constantly on the lookout for new colleagues who are enthusiastic about innovation. Every day our company is confronted with new challenges and issues, which we always address with our values in mind. We act with integrity, we make a difference, we work together, we care for others, and we reimagine the possible day by day with our global PwC network. In our subcluster 'CDO' in Financial Services Technology Consulting, we advise our clients - multinational banks, insurance companies, and finance-oriented corporates - on the digital transformation of their projects. This includes the areas of Data & Analytics, Big Data, Artificial Intelligence, Blockchain Technologies, and Credit Risk.
About the project
Are you interested in combining theoretical machine learning with a practical use case, getting your hands dirty, and physically seeing your results?
Problem statement: Insurance companies provide new data-driven products based on IoT data like “pay-as-you-drive” car insurance models. This brings new complexity to their IT enterprise systems and also requires new skills to analyze the data. We transferred the “pay-how-you-drive” business model from the real world to a slot car racetrack (“Carrera Bahn”). The Carrera Bahn has been equipped with light barriers, the cars are modified with acceleration sensors and the track can be monitored by camera. Those three sensors are used to calculate driver scores using a cloud-based environment. Since the Carrera track with all sensors is provided to the students, the data can be generated easily and AI-specific. In case of any problems with the data acquisition, PwC has a backup data set ready for students to create the basic solution.
Objective: Our goal is to analyze the perfect movement of the cars on track with physical sensors (optical sensor, acceleration sensor, and image sensor) as a real-time data stream input by using the data obtained to set up and train an AI to drive the fastest lap without disqualifying (e.g. flying out of the racetrack). Start configuring the racetrack in such a way that the sensors collect meaningful data used for training, validation, and testing the related AI. In the meantime, with the guidance of our experts, find a solution to set up a machine learning algorithm. Start with the technical part by implementing the ML algorithm using state-of-the-art tools. Get hands-on feedback to optimize your solution and define new boundaries.
Accepted students to this project should attend (unless they have proven knowledge) online workshops at the LRZ from TBA. More information will be provided to students accepted to this project.