Abstract: Methods and systems of estimating an accuracy of a neural network on out-of-distribution data. In-distribution accuracies of a plurality of machine learning models trained with in-distribution data are determined. The plurality of machine learning models includes a first model, and a remainder of models. In-distribution agreement is determined between (i) an output of the first machine learning model executed with an in-distribution dataset and (ii) outputs of a remainder of the plurality of machine learning models executed with the in-distribution dataset. The machine learning models are also executed with an unlabeled out-of-distribution dataset, and an out-of-distribution agreement is determined. The in-distribution agreement is compared with the out-of-distribution agreement.

Abstract: A method for providing a function via a network of processing units using multiple computing instances. Each computing instance implements the function and is configured, when it is executed, to determine at least one result in response to a call-up of the function. The method includes validating the computing instances, each being checked as to whether the computing instance corresponds to a respective predetermined state; determining a respective level of reliability for each of the processing units; starting execution of successfully validated computing instances in the processing units, which have a level of reliability that is equal/greater than a predetermined minimum level of reliability; receiving a function request from a user; calling up the function in at least a portion of the executed computing instances corresponding to the function request, in order to determine multiple results; determining a response based on the results; and sending the response to the user.

Abstract: A method determines a distance metric for determining a distance to a data point having heterogeneous classes of variables. The method includes providing training records each assigning a label to a data point, the training records partitioned into training data points of a training amount and validation data points of a validation amount, and training a data-based system model with the training amount, such that the data-based system model associates data points with a model output, respectively. The method further includes for each validation data point of the validation amount, determining a quality level of the data-based system model and a distance value to a nearest training data point for each of the heterogeneous classes of variables. The distance value to the nearest training data point is determined separately with respect to a respective class of variables.

Abstract: An electromechanical brake pressure generator for a hydraulic braking system of a vehicle. The brake pressure generator has a spindle drive unit for converting an input-drive-side rotational motion into a translational motion for piston actuation of a hydraulic piston/cylinder unit. A multi-stage gear linkage is disposed between the spindle drive unit and an electric drive motor. The gear linkage encompasses a planetary gearset unit driven by a motor pinion constituting a sun gear, the output-drive-side ring gear of which unit drives the spindle drive unit. A motor shaft of the electric drive motor extends in adjacently parallel fashion next to a spindle drive shaft of the spindle drive unit.

Abstract: An optical system for a virtual retinal display. The system includes: an image source; an image processing device; a projector unit including a light source generating a light beam and a controllable deflection device for the light beam for scanning projection of the image content; a first redirection unit comprising a first mirror and a statically mounted optical unit, the controllable deflection device projecting the image content onto the first mirror, and the first mirror directs the projected image content onto the statically mounted optical unit; a second redirection unit, the optical unit projecting image content onto the second redirection unit, the second redirection unit directing the projected image content onto an eye of a user.

Abstract: A computer-implemented method for training a machine learning network. The method may include receiving an input data, selecting one or more batch samples from the input data, applying a perturbation object onto the one or more batch samples to create a perturbed sample, running the perturbed sample through the machine learning network, updating the perturbation object in response to the function in response to running the perturbed sample, and outputting the perturbation object in response to exceeding a convergence threshold.

Abstract: An indicator device for a rechargeable battery pack, in particular, a rechargeable battery pack of a power tool, for indicating a charge state of the rechargeable battery pack, including at least one charge state indicator unit, which includes at least one indicator element for indicating the charge state of the rechargeable battery pack, and including at least one control line for transmitting at least one electrical and/or electronic signal, in particular, as a function of a charge state of the rechargeable battery pack, from a separate control unit of the rechargeable battery pack, in particular, structurally separated with respect to the charge state indicator unit, in particular, from main control electronics of the rechargeable battery pack to the charge state indicator unit. The charge state indicator unit is configured to short-circuit at least one indicator element as a function of the electrical and/or electronic signals transmitted via the control line.

Abstract: A method for monitoring operation of at least one electrical load for a vehicle. A comparison of an actual value of an operating parameter with four threshold values is performed. A monitoring signal is also generated dependent on a result of the comparison, the monitoring signal being generated with a first signal characteristic if the actual value lies between the first threshold value and the third threshold value or between the second threshold value and the fourth threshold value, or being generated with a second signal characteristic if the actual value lies above the third threshold value or below the fourth threshold value. The first signal characteristic effects activation of at least one safety function that maintains an active operating state of the at least one electrical load. The second signal characteristic effects deactivation of the at least one electrical load.

Abstract: In a method for producing a brake disc for a motor vehicle, a base disc, which is made of cast iron or aluminum in particular, is provided and equipped with an anti-corrosion layer. The anti-corrosion layer is applied using a wet chemical or galvanic method.

Abstract: The invention relates to a valve (10) for regulating oxygen in a fuel cell system (11), comprising: —a valve housing (12), which has a valve seat (13) and a through-opening (14) in the valve seat (13); —a valve shaft (15), which has a valve disk (16) formed thereon for closing the through-opening (14) in the valve seat (13) in a closed state of the valve (10) and for releasing the through-opening (14) in a released state of the valve (10); and—a moving means (17) for linearly moving the valve disk (16) between an opening position in an open state, in which open state the valve disk (16) is spaced apart from the valve seat (13), and a closing position in the closed state, in which closed state the valve disk (16) is positioned on the valve seat (13), and for rotationally moving the valve shaft (15) in order to pivot the valve disk (16) between the opening position and a releasing position in the released state, in which released state at least part of the valve disk (16) is spaced farther apart from the valve

Abstract: A computer-implemented method for verifying messages in a service-oriented communication system of a vehicle, including receiving a message and a signature in a first entity of the service-oriented communication system, the message and the signature being received via the service-oriented communication system; checking if the message corresponds to a dedicated message and the signature corresponds to a signature belonging to the dedicated message; and verifying the message, if the checking turns out positive. A computer-implemented method for generating predetermined messages in a first entity of a service-oriented communication system of a vehicle, and a service-oriented communication system in a vehicle, which is configured, are also described.

Abstract: A radial piston machine includes a rotor within a housing that is rotatable about an axis, a shaft coupled to the rotor, a distributor which surrounds the shaft and cannot rotate with the rotor and that contacts the rotor in an axial direction, a plurality of brake discs including first brake discs and second brake discs arranged side-by-side along the axis, wherein the first brake discs are coupled to the housing for torque transmission, and the second brake discs are coupled to the shaft for torque transmission, and an actuator configured to releasably exert a brake force in the axial direction on the brake discs. The distributor is located axially between the rotor and the actuator, and the brake discs are located axially between the distributor and the actuator. When the brake force is exerted the brake force is supported by the distributor in the axial direction.

Abstract: An electromechanically drivable brake pressure generator for a hydraulic braking system of a vehicle. The electromechanically drivable brake pressure generator includes an electric motor, using which a driving rotational speed is generatable; a planetary gear, which is driven on the input side by the electric motor to convert the driving rotational speed of the electric motor into a slower rotational speed, the planetary gear including an annulus gear, accommodated rotatably fixedly thereto in an annulus gear receptacle; and a hydraulic module, which is connected to an output side of the planetary gear and using which a brake pressure is generatable. A torque support is situated between the annulus gear and the annulus gear receptacle, which extends in an axial direction to the planetary gear and which cooperates with a recess so that the annulus gear and the annulus gear receptacle are held rotatably fixedly with respect to one another.

Abstract: The disclosure relates to a friction brake body for a friction brake of a motor vehicle, in particular a brake disc, wherein the friction brake body comprises a base body made from gray cast iron, and at least one wear resistant layer formed at least in areas on the base body. The wear resistant layer is a laser alloyed or laser dispersed edge layer of the base body and comprises at least one additive.

Abstract: A brake booster assembly having a brake booster body and a pushrod as well as a pushrod-pedal coupler are provided. The pushrod travels along a pushrod axis when being pushed. The brake booster assembly includes a pushrod-pedal coupler. The pushrod-pedal coupler comprises a pushrod-attachment end arranged on the pushrod and defining a longitudinal axis parallel to the pushrod axis. A pedal-attachment end is configured to be attached to a pedal and having a pedal pivot axis around which the pedal pivots when being pushed. The pedal pivot axis extends across the longitudinal axis. A coupler hinge is arranged between the pushrod-attachment end and the pedal-attachment end and connects the pushrod-attachment end with the pedal-attachment end. The coupler hinge has a coupler hinge axis that extends across the longitudinal axis and the pedal pivot axis. The pedal pivot axis and the coupler hinge axis are offset along the longitudinal axis.

Abstract: A method for predicting a future action of an object for a driving assistance system for a highly automated mobile vehicle. At least one sensor signal from at least one vehicle sensor of the vehicle is read in, the sensor signal representing at least one piece of kinematic object information concerning the object that is detected by the vehicle sensor at an instantaneous point in time. A planner signal from a planner of the autonomous driving assistance system is read in, the planner signal representing at least one piece of semantic information concerning the object or the surroundings of the object at a point in time in the past. The kinematic object information is fused with the semantic information to obtain a fusion signal. A prediction signal is determined using the fusion signal, the prediction signal representing the future action of the object.

Abstract: A system for tamper detection of a motion detector. The system includes an electronic controller configured to receive orientation data from an accelerometer of the motion detector. The electronic controller is also configured to filter the orientation data of the accelerometer of the motion detector. The electronic controller is further configured to determine an orientation of the motion detector using the filtered orientation data. The electronic controller is also configured to identify a tamper condition of the motion detector based on the orientation of the motion detector. The electronic controller is further configured to activate an alarm device of the motion detector based on the tamper condition.

Abstract: A method for protecting a computer-based classifier. The method includes determining a first similarity of a plurality of input data sets of a computer-based classifier, comparing the determined first similarity of the plurality of input data sets and a second similarity of a comparison corpus of input data sets and recognizing the possibility of an attack on the computer-based classifier when the first similarity of the plurality of input data sets and the second similarity of the comparison corpus deviate from one another in a predetermined manner.

Abstract: A computer-implemented method for filtering operating scenarios during the operation of a vehicle. The method includes receiving at least one time series of data, storing time-ordered data points of the at least one time series of data in a first memory, generating a trace diagram for the time-ordered data points stored in the first memory, checking whether the trace diagram generated is new in comparison to one or more existing trace diagrams, and storing the time-ordered data points stored in the first memory, in a second memory if the trace diagram generated is new.

Abstract: The present invention obtains a controller and a control method capable of achieving appropriate cornering during adaptive cruise control of a straddle-type vehicle. In the controller and the control method according to the present invention, during the adaptive cruise control in which the straddle-type vehicle is made to travel according to a distance from the straddle-type vehicle to a preceding vehicle, motion of the straddle-type vehicle, and a driver's instruction, at least one of braking force distribution, which is distribution of braking forces generated on wheels of the straddle-type vehicle to the front and rear wheels, and drive power distribution, which is distribution of drive power transmitted to the wheels of the straddle-type vehicle to the front and rear wheels, is controlled on the basis of lateral acceleration of the straddle-type vehicle.