Abstract: A method for lane and roadway detection uses a multitask DNN architecture including an encoder, a first decoder and a second decoder. The method includes the following steps: providing an input image by an optical detection device, filtering the input image by the encoder, generating a first representation of the lane and/or roadway by the encoder, processing the first representation in the first and second decoders, outputting two different representations from the first and second decoders, combining the two different representations, and outputting identified lanes, lane markings and the roadway.

Abstract: A method for improving prediction results of advanced driver assistance systems of a vehicle includes obtaining map data including information about a road geometry in a proximity of the vehicle. A sensor means is assigned, sensing a surrounding of the vehicle. A virtual sensing means output is generated, corresponding to an output of the sensing means if the sensing means. The generation is based on a mathematical model of the sensing means. The surrounding of the vehicle is sensed, and a sensing means output is generated. The sensing means output is compared to the virtual sensing means output. Parameters of the mathematical model are modified, and the virtual sensing means output are generated and compared to the sensing means output. Map data is combined with information derived from the sensing means output to generate combined information, which is output for further processing.

Abstract: The invention relates to a method for lane detection (S1-S3) and roadway detection (1), wherein a multitask DNN architecture is used for detection and consists of at least one encoder and at least one first and one second decoder, the method comprising the following steps: providing (VS1) an input image by means of an optical detection device, filtering (VS2) the input image by means of the encoder, generating (VS3) a first representation of the lane (S1-S3) and/or roadway (1) by the encoder, forwarding (VS4) the first representation to the first and the second decoder, processing (VS5) the first representation in the first and the second decoder, outputting (VS6) two further different representations of the first and second decoder, combining (VS7) the two representations of the first and the second decoder, outputting (VS8) identified lanes (S1-S3) and lane markings and also the roadway (1).

Abstract: An automotive driver assistance for a vehicle with an improved capability to handle obstructed sensor coverage includes steps of acquiring sensor data on an environment of the vehicle from at least one sensor, of generating an environment representation based on the acquired sensor data and of predicting at least one behavior of the vehicle. The method determines at least one area in the environment of the vehicle wherein for the at least one area either a confidence for the sensor data is below a threshold or no sensor data is available and generates at least one virtual traffic entity in the at least one determined area, wherein the virtual traffic entity is adapted to interact with the at least one predicted behavior of the vehicle. A risk measure for each combination of the at least one virtual traffic entity and the predicted behavior of the vehicle is estimated, the calculated risk measure is evaluated and a controlling action for the vehicle is executed based on the evaluated risk measure.

Abstract: An automotive driver assistance for a vehicle with an improved capability to handle obstructed sensor coverage includes steps of acquiring sensor data on an environment of the vehicle from at least one sensor, of generating an environment representation based on the acquired sensor data and of predicting at least one behavior of the vehicle. The method determines at least one area in the environment of the vehicle wherein for the at least one area either a confidence for the sensor data is below a threshold or no sensor data is available and generates at least one virtual traffic entity in the at least one determined area, wherein the virtual traffic entity is adapted to interact with the at least one predicted behavior of the vehicle. A risk measure for each combination of the at least one virtual traffic entity and the predicted behavior of the vehicle is estimated, the calculated risk measure is evaluated and a controlling action for the vehicle is executed based on the evaluated risk measure.

Abstract: A method supports driving an ego-vehicle including a driver assistance system. A traffic participant or infrastructure element involved in the traffic situation is selected and taken into consideration for traffic scene analysis. A hypothetical future trajectory for the ego-vehicle is predicted by predicting the current state of the ego-vehicle and varied to generate a plurality of ego-trajectory alternatives including the calculated hypothetical future ego-trajectory. A hypothetical future trajectory from another traffic participant gained by predicting the current state of the traffic participant or calculating a hypothetical future state sequence of the infrastructure element is determined. Based upon at least one pair of ego-trajectory plus one other trajectory risk functions over time or along the calculated hypothetical future ego-trajectory alternatives are calculated. One risk function corresponds to one ego-trajectory alternative. The risk functions are combined into a risk map.

Abstract: A method for improving prediction results of advanced driver assistance systems of a vehicle includes obtaining map data including information about a road geometry in a proximity of the vehicle. A sensor means is assigned, sensing a surrounding of the vehicle. A virtual sensing means output is generated, corresponding to an output of the sensing means if the sensing means. The generation is based on a mathematical model of the sensing means. The surrounding of the vehicle is sensed, and a sensing means output is generated. The sensing means output is compared to the virtual sensing means output. Parameters of the mathematical model are modified, and the virtual sensing means output are generated and compared to the sensing means output. Map data is combined with information derived from the sensing means output to generate combined information, which is output for further processing.

Abstract: A method supports driving an ego-vehicle including a driver assistance system. A traffic participant or infrastructure element involved in the traffic situation is selected and taken into consideration for traffic scene analysis. A hypothetical future trajectory for the ego-vehicle is predicted by predicting the current state of the ego-vehicle and varied to generate a plurality of ego-trajectory alternatives including the calculated hypothetical future ego-trajectory. A hypothetical future trajectory from another traffic participant gained by predicting the current state of the traffic participant or calculating a hypothetical future state sequence of the infrastructure element is determined. Based upon at least one pair of ego-trajectory plus one other trajectory risk functions over time or along the calculated hypothetical future ego-trajectory alternatives are calculated. One risk function corresponds to one ego-trajectory alternative. The risk functions are combined into a risk map.