A method of generating vehicle control data includes: storing, with a storage device, relationship prescription data; operating, with an execution device, an operable portion of an internal combustion engine; acquiring, with the execution device, a detection value from a sensor that detects the state of the vehicle; calculating, with the execution device, a reward; and updating, with the execution device, the relationship prescription data using update mapping determined in advance, the update mapping using the state of the vehicle based on the detection value, an operation amount used to operate the operable portion, and the reward corresponding to the operation as arguments, and returning the relationship prescription data which have been updated such that an expected profit for the reward calculated when the operable portion is operated in accordance with the relationship prescription data increases.

A controller includes a processor programmed to determine, for a vehicle, a first control input based on input data and first reference parameters. The processor is further programmed to operate the vehicle according to the first control input. Based on operating data of the vehicle for an operating condition, the processor determines a second control input for the vehicle. Operating the vehicle according to the second control input reduces a cost of operating the vehicle relative to operating the vehicle according to the first control input. The processor is further programmed to determine, based on the second control input, second reference parameters. The controller generates a third control input based on the second reference parameters and the input data. A cost of operating the vehicle according to the third control input is reduced relative to the cost of operating the vehicle based on the first control input.

Enclosed are embodiments for sampling driving scenarios for training machine learning models. In an embodiment, a method comprises: assigning, using at least one processor, a set of initial physical states to a set of objects in a map for a set of simulated driving scenarios, wherein the set of initial physical states are assigned according to one or more outputs of a random number generator; generating, using the at least one processor, the set of simulated driving scenarios in the map using the initial physical states of the objects in the set of objects; selecting, using the at least one processor, samples of the simulated driving scenarios; training, using the at least one processor, a machine learning model using the selected samples; and operating, using a control circuit, a vehicle in an environment using the trained machine learning model.

This disclosure is related to a method, a computer program code, and an apparatus for training a convolutional neural network for an autonomous driving system. The disclosure is further related to a convolutional neural network, to an autonomous driving system comprising a neural network, and to an autonomous or semi-autonomous vehicle comprising such an autonomous driving system. For training the convolutional neural network, in a first step real-world driving data are selected as training data. Furthermore, synthetic driving data are generated as training data. The convolutional neural network is then trained on the selected real-world driving data and the generated synthetic driving data using a genetic algorithm.

A driver assist analysis system includes a processing system and a database that stores vehicle data, vehicle operational data, vehicle accident data, and environmental data related to the configuration and operation of a plurality of vehicles with driver assist systems or features. The driver assist analysis system also includes one or more analysis engines that execute on the processing system to determine one or more driving anomalies (e.g., accidents or poor driving operation) based on the vehicle operational data, and that correlate or determine a relationship between the driving anomalies and the operation of the driver assist systems or features. The driver assist analysis system then determines an effectiveness of operation of one or more of the driver assist systems or features as a statistical measure based on the determined relationship and the driver assist analysis system notifies a user or receiver, such as a manufacturer or an insurance company of the statistical measure.

A universal modeling method is provided for a motor vehicle, the universal modeling method including: providing an input signal set, the input signal set comprising those signals of respective sensors of the motor vehicle which can be relevant for the control of devices of the motor vehicle; selecting a set of modeling signals from the input signal set as a function of a system architecture of the motor vehicle; and determining an output signal set by way of a time-discrete selective state space model modeling function taking into account the set of modeling signals. In this case, the output signal set functions as a signal set for controlling corresponding actuators of the devices of the motor vehicle.

A method for determining a trajectory of an autonomous vehicle includes a first phase that is carried out while the trajectory of the vehicle is controlled manually. The first phase includes calculating a first theoretical trajectory of the vehicle, measuring a trajectory actually followed by the vehicle, and calculating a correction factor. Calculating the correction factor includes a comparison of the first theoretical trajectory with the trajectory actually followed. The method also includes a second phase that is carried out while the trajectory of the vehicle is controlled autonomously. The second phase includes calculating a second theoretical trajectory of the vehicle, and calculating a customised trajectory of the vehicle. Calculating the customised trajectory is based on the second theoretical trajectory and on the correction factor.

A method for reducing exhaust gas emissions of a drive system of a vehicle including an internal combustion engine, including generating first driving profiles using a computer-implemented machine learning system, the statistical distribution of the first driving profiles being a function of a statistical distribution of second driving profiles measured during real driving operation, calculating respective exhaust gas emissions for the first driving profiles using a computer-implemented modeling of the vehicle or the drive system, adapting the drive system as a function of at least one of the calculated exhaust gas emissions, the adaptation taking place as a function of a level or of a profile of the calculated exhaust gas emissions and of a statistical frequency of the corresponding first driving profile, the statistical frequency of the corresponding first driving profile being ascertained with the aid of the statistical distribution of the first driving profiles.

The present application discloses a method for determining an automatic driving feature, an apparatus, a device, a medium and a program product, and relates to automatic driving technology in the field of artificial intelligence. A specific solution is: acquiring scenario information of a plurality of driving scenarios and driving behavior information of an automatic driving system in each of the driving scenarios, where the driving behavior information includes a decision made by the automatic driving system and an execution result corresponding to the decision; determining an automatic driving feature of the automatic driving system according to the scenario information of the plurality of driving scenarios and respective driving behavior information. The automatic driving feature determined through the above process can represent a characteristic of an automatic driving strategy adopted by the automatic driving system.

A method and an apparatus for automated driving, a device, and a computer-readable storage medium are provided. The method for automated driving includes: obtaining data associated with a parameter and an external environment of a vehicle, the data being collected according to a collection policy for a target scenario for the vehicle; generating an automated driving model for the target scenario based on the obtained data; and updating the collection policy by testing the automated driving model.