The goal of this project is to design an estimator for the mass, and the location of the center of mass, of a vehicle. Both are important parameters of the vehicle’s dynamics and are thus needed for an optimal control design. However, due to the variable load of a vehicle, these parameters cannot be measured offline, but have to be estimated online while the vehicle is operating. Severaly model-based approaches have been proposed to this end, but none of them has proven to be be fully satisfactory. The fundamental question posed in this project is whether better estimates of the mass, and the center of mass, can be generated based on the available data streams by the use of a deep learning approach. The approach shall be tested on a model car, which features an embedded computing platform (NVIDIA Jetson TX2).

Team & Contact

• Prof. Dr. Georg Schildbach (georg.schildbach@uni-luebeck.de)
• Jasper Pflughaupt