Abstract: A system receives identification of a machine learning model, stored by a vehicle, that is ready for training. The system determines, based on a configuration file associated with the model, whether batch-based or live training is to be used to train the model and collects data for training, defined by a configuration file associated with the model. The system calls a learning as a service vehicle process to load a kernel and configures the kernel based on configuration data defined in the configuration file. The system receives notification from the learning as a service process that the training is complete. Additionally, the system validates a model, responsive to the notification including indication that training was successful, using a vehicle model validation process to test the model with live data before deployment by background execution of the model and saves a copy of the model for deployment responsive to successful validation.

Abstract: This disclosure describes systems and using dual vehicle digital twins for model-based learning and remotely-operated vehicles. An example method may include receiving, by a processor associated with a first digital twin stored locally on a computing system of a vehicle or a second digital twin stored remotely on a first remote system, data relating to a first sub-system of the vehicle and a second sub-system of the vehicle. The example method may also include sending the data to the first digital twin or the second digital twin, wherein sending the data includes synchronizing with the first digital twin or the second digital twin.

Abstract: Routes may be presented to a vehicle operator with information about predicted autonomous driving levels. A road network route to a destination is received. The road network route is partitioned into segments. Indicia of the segments are provided to an inference engine. The inference engine predicts levels of autonomous driving for the respective segments generated by the inference engine based on the indicia of the segments. Based on the predicted levels of autonomous driving for the respective segments, a ratio of a level of autonomous driving for the road network route is computed. The ratio corresponds to a proportion of time and/or distance for the road network route during which a driver-system is predicted to be engaged at the level of autonomous driving. A user interface is displayed, which includes a graphic representation of the road network route and a graphic indication of the ratio.