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.

A driving support system includes: an acquisition portion configured to acquire visual-recognition position information on a position where a driver of a vehicle visually recognizes a traffic light; an image acquisition portion configured to acquire a forward image ahead of the vehicle; a traffic-light recognition portion configured to recognize a traffic light included in a forward image; and a notification portion configured to notify the driver of warning when the traffic light is not recognized from the forward image, in a case where the vehicle is present at a position based on the visual-recognition position information.

In one embodiment, a method for monitoring a localization function in an autonomous driving vehicle (ADV) can use known static objects as ground truths to determine when the localization function encounter errors. The known static objects are marked on a high definition (HD) map for the real-time driving environment. When the ADV detects one or more known static objects, the ADV can use sensor data, locations of the one or more static objects, and one or more error tolerance parameters to create a localization error tolerance area surrounding a current location of the ADV. The ADV can project the tolerance area on the HD map, performs a localization operation to generate an expected location of the ADV on the HD map, and determines whether the generated location falls within the projected tolerance area. If the generated location falls outside the projected tolerance area, indicating a localization function of the ADV encounter errors, the ADV can generate an alarm to alert a human driver to switch to a manual driving mode. If no human driver is available in the ADV, the ADV can activate a vision-based fail-safe localization procedure.

Aspects of the disclosure provide for controlling an autonomous vehicle. For instance, a trajectory for the autonomous vehicle to traverse in order to follow a route to a destination may be generated. A first error value for a boundary of an object, a second error value for a location of the autonomous vehicle, a third error value for a predicted future location of the object may be received. An uncertainty value for the object may be determined by combining the first error value, the second error value, and the third error value. A lateral gap threshold for the object may be determined based on the uncertainty value. The autonomous vehicle may be controlled in an autonomous driving mode based on the lateral gap threshold for the object.

The invention relates to a method for abstracting a data record, wherein the data record is provided for machine learning at least one function, comprising the following steps: Training a complex neural network structure comprising different neural networks in the at least one function by way of machine learning based on the data record by means of a machine learning control apparatus, wherein the neural networks and the complex neural network structure are optimized with respect to maximum representativity of the data record, providing the trained complex neural network structure in the form of a data-record-specific knowledge module so that knowledge contained in the data record can be further used in a manner compliant with data protection. The invention further relates to an associated device.

In some implementations, a method of verifying operation of a sensor is provided. The method includes causing a sensor to obtain sensor data at a first time, wherein the sensor obtains the sensor data by emitting waves towards a detector. The method also includes determining that the detector has detected the waves at a second time. The method further includes receiving the sensor data from the sensor at a third time. The method further includes verifying operation of the sensor based on at least one of the first time, the second time, or the third time.

Technologies and techniques for the lateral control of a vehicle. Environment data of a vehicle is detected when travelling a route, and stored environment data, detected when travelling the route by a plurality of other vehicles not currently travelling the route, is received. The plausibility of the stored environment data is checked on the basis of the environment data detected. Lateral control of the vehicle is executed on the basis of the environment data checked for plausibility. The present disclosure also relates to an arrangement for the lateral control of a vehicle.

An electronic control device is mounted on a vehicle having a sensor, and includes: a storage unit that stores a control history in which a vehicle control condition and an actual number of measurement times of the vehicle control condition are associated with each other, the vehicle control condition being a condition for performing an arbitrary vehicle control operation for each position; a position specifying unit that specifies a position of the vehicle; a determination unit that determines a current sensor capability that is a current capability of the sensor; and a selection unit that selects, from the storage unit, the vehicle control condition having the largest actual number among the vehicle control conditions that have a position within a predetermined distance from the position specified by the position specifying unit and that can be determined by the current sensor capability.

Systems, methods, and media for factoring localization uncertainty of an ADV into its planning and control process to increase the safety of the ADV. The uncertainty of the localization can be caused by sensor inaccuracy, map matching algorithm inaccuracy, and/or speed uncertainty. The localization uncertainty can have negative impact on trajectory planning and vehicle control. Embodiments described herein are intended to increase the safety of the ADV by considering localization uncertainty in trajectory planning and vehicle control. An exemplary method includes determining a confidence region for an ADV that is automatically driving on a road segment based on localization uncertainty and speed uncertainty; determining that an object is within the confidence region, and a probability of collision with the ADV based on a distance of the object to the ADV; and planning a trajectory based on the probability of collision, and controlling the ADV based on the probability of collision.

Disclosed is a method for managing a torque to be supplied on a shaft of an electric motor including the following steps: creating and initializing a correction data table; acquiring a torque instruction; determining a torque command; measuring the current output from the converter and measuring the current output from the battery; calculating an electrical power on the basis of the two current measurements and of the voltage; determining, on the basis of a map, a corresponding torque on the basis of the electrical power at the motor and of the rotational speed of the motor; reading the information provided by the motor relating to the torque exerted on its shaft; calculating the difference between the information provided by the motor and the torque determined on the basis of the electrical power; and updating the correction data table.