Prof. Dr. h.c. Hasso Plattner


Application of Causal Inference in Automotive Production

Deep Learning (DL) is the area of Machine Learning (ML) that applies artificial Neural Networks (NNs) to a broad family of tasks in different domains. The Internet of Things (IoT) in its essence is the inter-networking of physical entities (agents), each being a combination of sensors, intelligence and actuators. DL can be seen as a facilitator for IoT from multiple point of views, including both the injection of NNs into the agents and the use of NNs for orchestrating the cooperation and collaboration of the agents.

This project will look into the orchestration of things in car production processes. Sensors in production gather data. Actuators react intelligently in order to reach the production goals. Robots are agents that consists of sensors, actuators and intelligence. The more robots collaborate, the more sensors and actuators need to be taken into account. Production parameters like material thickness, densities and dimensions also play a role. Also, it is required that an intelligent orchestration works properly in all production states - i.e., smoothly, material delayed, on hold, et cetera.

The project will be a joint effort of HPI and Porsche AG/MHP, represented by Porsche Digital Lab Berlin (PDLB).