Robert Boschgmbh

Philip Lenz

Alexander Lengsfeld

Marcel Brueckner

Joern Jachalsky

Esther Dietrich

A method for predicting an ego-lane for a vehicle. The method includes: receiving at least one image captured by at last one camera sensor of the vehicle, which depicts a lane that may be used by a vehicle; ascertaining a center line of the lane, which extends through a center of the lane, by implementing a trained neural network on the captured image, the neural network being trained via regression to ascertain a center line of a lane, which extends in a center of the lane, based on captured images of the lane; outputting a plurality of parameters, which describe the center line of the lane, via the neural network; generating the center line based on the parameters of the center line; identifying the center line of the lane as the ego-lane of the vehicle; and providing the ego-lane.

Method for predicting an ego-lane for a vehicle

A method used for deriving a control variable for lateral guidance of a motor vehicle by means of images of at least one camera of the motor vehicle, wherein the at least one camera captures images of a roadway. A more precise and/or resource-saving derivation of the control variable, and thus lateral guidance, of the motor vehicle is achieved by projecting a traffic lane to be followed as well as a travel trajectory of the motor vehicle onto the images, and by a control variable for lateral guidance of the motor vehicle being derived by comparing the traffic lane projected onto the images with the travel trajectory projected onto the images. A motor vehicle in which the method is performed is also described.

Method used for deriving a control variable for lateral guidance of a motor vehicle

A method for determining a value of at least one controller variable for guiding a mobile platform in at least semiautomated fashion is proposed. Images of surroundings of the mobile platform are determined using a plurality of two-dimensionally representing sensor systems. A multiplicity of objects are identified in the images. At least two object features for each of the multiplicity of objects are determined in order to determine the at least one controller variable. An input tensor is generated using the in each case at least two determined object features of the multiplicity of objects for a trained neural network. The value of the controller variable is estimated using the input tensor and the trained neural network.

Method for determining a value of a controller variable

A method for determining a movement of a device relative to at least one object based a digital image sequence of the object recorded from the location of the device. The method includes computing a plurality of optical flow fields from image pairs of the digital image sequence; finding the position of an object in a partial image region in the most current image in each case and assigning the partial image region to the object; forming a plurality of partial optical flow fields from the plurality of optical flow fields; selecting a partial flow fields from the plurality of partial flow fields in accordance with at least one criterion to facilitate the estimation of a change in scale of the object; and estimating the change in scale for the at least one object using the assigned partial image region based on the selected partial flow field.

Method for determining a relative movement using a digital image sequence

A method for determining a degree of overlap of at least one object with at least one lane via a representation of a surrounding environment of a platform as a two-dimensional pixel data field. The method includes allocating at least one lane pixel group to pixels of the two-dimensional pixel data field, which correspondingly represent at least one lane; allocating at least one object pixel group to pixels of the two-dimensional pixel data field, which correspondingly represent at least one object; defining at least one object pixel pair in the two-dimensional pixel data field, which pair characterizes a width of the at least one object pixel group; and comparing the object pixel pair and the lane pixel group in the two-dimensional pixel data field.

Method for determining a degree of overlap of an object with a lane

A method is described for ascertaining a state of a vehicle light of a vehicle. The method includes a step of reading in and a step of evaluating. In the step of reading in, an image signal is read in that includes a camera image including a vehicle area of a detected vehicle. In the step of evaluating, the image signal is evaluated by using an artificial teachable model, in particular an artificial neural network. Here, at least two state categories, each representing a state of the vehicle light of the vehicle, are differentiated. That state category is ascertained as the evaluation result which represents the instantaneous state of the vehicle light of the vehicle.

Method and device for ascertaining a state of a vehicle light of a vehicle

An apparatus for ascertaining, from at least one image, a delimited region and/or a motion path of at least one object includes at least one artificial neural network (ANN) made up of several successive layers. The first layer of the ANN receives as input the at least one image or a part thereof. The second layer supplies as output a boundary line of the delimited region, a linear course of the motion path, or a portion of the boundary line or motion path. The dimensionality of the second layer is lower than the dimensionality of the first layer.

Growing community of inventors

Holle, Germany

Philip Lenz