Robert Boschgmbh

Carsten Hasberg

Philipp Rasp

Jan-Hendrik Pauls

Oliver Pink

Piyapat Saranrittichai

Stefan Nordbruch

Tayyab Naseer

Danny Hiendriana

Muhammad Sheraz Khan

Ali Alawieh

Fabian Dominik Reister

Valentin Frommherz

Christoph Schroeder

Sybille Eisele

Karsten Muehlmann

Armin Ruehle

Tobias Strauss

Jan Rohde

Martin Mueller

Maxim Dolgov

Thomas Michalke

Peter Christian Abeling

Florian Hauler

Peter Sautter

Michael Helmle

Jeannine Schwarzkopf

Anke Svensson

Timo Nachstedt

Artur Koch

Andre-Marcel Hellmund

Ross Kidson

A method for training an artificial neural network uses training data that include first image data of a first image and second image data of a second image of an infrastructure. The first image includes a first feature, and the second image includes a second feature corresponding to the first image. The training data include a relative desired translation and a relative desired rotation between the first feature and the second feature. The training includes extracting the first feature from the first image and extracting the second feature from the second image using the artificial neural network. The extracted first feature is represented by first feature data having a first volume of data. The extracted second feature is represented by second feature data having a second volume of data. The training further includes ascertaining a relative translation and a relative rotation between the extracted first feature and the extracted second.

Method for training an artificial neural network

A method for providing operating data for an at least semiautomated vehicle. The method includes: sensorially detecting surroundings data with the aid of a surroundings sensor system of the vehicle; conveying the sensorially detected surroundings data to a cloud and conveying a request to the cloud for localization data; receiving the localization data from the cloud; and ascertaining the operating data for use in a cooperative driving maneuver of the vehicle with at least one further road user from the localization data received from the cloud and from the sensorially detected surroundings data; and initiating the cooperative driving maneuver using the ascertained operating data.

Method for providing operating data for an at least semi-automated vehicle

A method and a device for determining a highly precise position of a vehicle. The method includes a step of sensing environment data values, which represent an environment of the vehicle, the environment encompassing multiple environment features which have at least one regular structure, and the environment data values at least encompassing the at least one regular structure; a step of carrying out a comparison of the environment data values with a map; a step of determining the highly precise position of the vehicle as a function of the comparison; and a step of supplying a signal on the basis of the highly precise position.

Method and device for determining a highly precise position of a vehicle

A method is described for verifying a digital map of a higher-level automated vehicle (HAV), in particular a highly automated vehicle, including the steps: S1 providing a digital map, preferably a highly accurate digital map: S2 determining an instantaneous reference position and localizing the reference position in the digital map; S3 establishing at least one actual feature property of a feature in the surroundings of the reference position, the establishment being carried out with the aid of at least one information source; S4 comparing the actual feature property to a setpoint feature property of the feature and ascertaining at least one difference value as the result of the comparison. A corresponding device and a computer program are also described.

Method for verifying a digital map of a higher-level automated vehicle, corresponding device and computer program

A method for certifying map elements for safety-critical driving functions by a control unit, at least one observation variable of at least one mapping step being ascertained by at least one map element after an implementation of the mapping step by a monitoring function, and being compared with a setpoint value of the observation variable, at least one result value being calculated based on a comparison of the observation variable with the setpoint value of the observation variable for the at least one mapping step by the monitoring function, the at least one result value being stored as a certificate and being linked with the at least one map element. A control unit, a computer program as well as a machine-readable memory medium are also described.

Optimized subdivision of digital maps into map sections

A method for providing data for creating a digital map. The method includes: detecting surroundings sensor data of the surroundings during a measuring run of a physical system, preferably a vehicle, the surroundings sensor data capturing the surroundings in an at least partially overlapping manner, first surroundings sensor data including three-dimensional information, and second surroundings sensor data including two-dimensional information; extracting, with the aid of a first neural network situated in the physical system, at least one defined object from the first and second surroundings sensor data into first extracted data; and extracting, with the aid of a second neural network situated in the physical system, characteristic features including descriptors from the first extracted data into second extracted data, the descriptors being provided for a defined alignment of the second extracted data in a map creation process.

Method and device for providing data for creating a digital map

A method for operating a higher-level automated vehicle (HAV), in particular a highly automated vehicle, is provided, including: S1 for providing a digital map, which may be a highly accurate digital map, in a driver assistance system of the HAV; S2 for determining an instantaneous vehicle position and localizing the vehicle position in the digital map; S3 for providing an expected setpoint traffic density at the vehicle position; S4 for ascertaining an instantaneous actual traffic density in the surroundings of the HAV; S5 for comparing the actual traffic density to the setpoint traffic density and ascertaining a difference value as the result of the comparison; S6 for checking the vehicle position of the HAV for plausibility at least partially based on the difference value and/or S7 for updating the digital map at least partially based on the difference value. Also described are a corresponding driver assistance system and a computer program.

Method for operating a higher-level automated vehicle (HAV), in particular a highly automated vehicle

A method for determining a localization pose of an at least partially automated mobile platform, the mobile platform being equipped to generate ground images of an area surrounding the mobile platform, and being equipped to receive aerial images of the area surrounding the mobile platform from an aerial-image system. The method includes: providing a digital ground image of the area surrounding the mobile platform; receiving an aerial image of the area surrounding the mobile platform; generating the localization pose of the mobile platform with the aid of a trained convolutional neural network, which has a first trained encoder convolutional-neural-network part and a second trained encoder convolutional-neural-network part.

Method for determining a localization pose of an at least partially automated mobile platform

A method is described for the certification by a control unit of map elements for safety-critical driving functions. At least one observation variable of at least one mapping step of at least one map element is ascertained after an implementation of the mapping step via a monitoring function and is compared with a setpoint value of the observation variable. At least one result value is calculated based on a comparison of the observation variable with the setpoint value of the observation variable for the at least one mapping step via the monitoring function. The at least one result value is stored as a certificate and is linked to the at least one map element. A control unit, a computer program and a machine-readable memory medium are also described.

Certification of map elements for automated driving functions

A computer-implemented method for supplying radar data. The method includes the following steps: receiving input data, the input data including satellite images; generating radar data using a trained machine learning algorithm, which is applied to the input data; and outputting the generated radar data.

Method and device for supplying radar data

A method for operating a more highly automated vehicle (HAF), in particular a highly automated vehicle, including: S&#x2014;providing a digital map or a highly accurate digital map, in a driver-assistance-system of the HAF; S&#x2014;determining a current vehicle position and locating the vehicle position in the digital map; S&#x2014;providing at least one expected feature property of at least one feature in a surroundings of the HAF; S&#x2014;detecting at least one actual feature property of a feature in the surroundings of the HAF at least partially on the basis of the expected feature property; S&#x2014;comparing the actual feature property with the expected feature property and ascertaining at least one differential value; S&#x2014;checking the plausibility of the actual feature property at least partially on the basis of the differential value. Also described is a corresponding system and a computer program.

Method for operating a more highly automated vehicle (HAF), in particular a highly automated vehicle

A method for fusing state data via a control unit. State data of a first mobile unit and of an object ascertained via a sensor system of the first mobile unit are received. State data of an object ascertained via a sensor system of a second mobile unit and/or state data of the second mobile unit, transmitted via a communication link from the second mobile unit to the first mobile unit, are received. A node is created in a time-position diagram for each set of received state data of the first mobile unit, the second mobile unit, and the objects. A data optimization of the state data ascertained by the first mobile unit and/or by the second mobile unit is carried out. An optimization problem is created based on the optimized state data ascertained by the first mobile unit and the optimized state data received from the second mobile unit.

Graph-based method for the holistic fusion of measured data

A method and device for creating a localization map, including a step of receiving environmental data values, a step of dividing the environmental data values into at least two separate data records, a step of creating a plurality of partial maps, a step of creating the localization map and a step of making the localization map available.

Method and device for creating a localization map

A method and corresponding system and computer program for a highly-automated-vehicle (HAV), the method including: providing a digital-map; determining a present vehicle-position relative to the digital-map; providing at least one setpoint property of at least one feature in an HAV-environment; using at least one sensor to detect at least one actual property of a feature in the HAV-environment based at least in part on the setpoint property; comparing the actual property to the setpoint property and determining at least one difference-value based on the comparison; and verifying the digital-map, the digital-map being classified as not up-to-date if the difference-value reaches/exceeds a specified-threshold-value of a deviation, and being classified as up-to-date if the difference-value remains below the specified-threshold-value of the deviation.

Method for verifying a digital map of a more highly automated vehicle (HAV), especially of a highly automated vehicle

In a method for operating a vehicle, a planning map and a localization map are provided, the vehicle is localized on the localization map, a map corridor indicated in the planning map is selected based on the localization, a sensor corridor is ascertained using a sensor unit of the vehicle, and the map corridor is compared to the sensor corridor. Using a specified threshold value for a deviation between the map corridor and the sensor corridor, it is decided whether the map corridor and the sensor corridor are identical. The map corridor is utilized for operating the vehicle if the map corridor and the sensor corridor are identical, and the sensor corridor is utilized for operating the vehicle if the map corridor and the sensor corridor are not identical.

Growing community of inventors

Ilsfeld-Auenstein, Germany

Carsten Hasberg