The push towards automated and autonomous driving is partly enabled by the recent advances in machine learning. Various algorithms are entrusted with perception, planning and control of vehicles to perform the desired driving task. However, supervising the state of the vehicle, i.e. the driving dynamics, is mandatory in such a scenario. To improve accuracy and reduce computation effort, a deep learning approach is chosen to model the driving dynamics of a vehicle using the Handling Roadway test bench. Different rear wheel steering controls are used to realize subtle differences in the vehicle dynamics. The resulting data is used to train different neural networks that are able to predict the driving dynamics of each configuration. Using an independent test data set, it can be shown that the neural networks are able to differentiate between the various rear wheel steering controls. These results enable the implementation of an extensive observer module. Finally, possible developments to further improve the performance are being discussed.
Session:
Vehicle Dynamics
|
| 15:00 - 15:30