Working Student - Generative AI-supported Testing Methods for Automated Driving Development

Complexity and scale of automotive systems are increasing, particularly in the area of automated driving and advanced driving assistance systems. The increased complexity of such safety-critical systems demands for advanced testing and verification methods to ensure safety under an abundance of operating conditions and traffic scenarios that may be encountered.

Recently, generative AI has shown promising results for content generation, such as for text (ChatGPT), image (stable diffusion), or video generation (OpenAI SORA). In our R&D work, we would like to explore the possibilities of using generative AI to create traffic scenarios learned from recorded traffic data.

We are seeking for a talented and driven Working Student (m/f/d) to join the Systems Engineering R&D team and support the development of testing methods by using recent results from generative AI for trajectory and scenario generation. In your role, you gain hands-on experience on the training and validation of generative AI models, and contribute to our R&D testing platform for automated driving.

Responsibilities:

• Implementation of a generative AI model for trajectory/scenario generation
• Definition of relevant features in the traffic scene
• Training and validation of the scenario generation model
• Integration of the methods into our existing testing workflow
• Extension of the generation model by a control mechanism to guide the generation procedure towards user-defined targets
• Benchmark the testing campaign to extract critical test case

Requirements

• Pursuing a B.Sc/M.Sc degree in Computer Science, Electrical Engineering, or a related field.
• Strong programming skills, ideally in Python.
• Familiarity with machine learning training frameworks (e.g. pytorch) is a big plus.
• Familiarity with autonomous vehicle technology and testing methodologies is a plus.
• Fluent level of English or German is required.
• Team player with intercultural competencies as well as independent working style with good self-organization capabilities.