Method and control device for recognizing an object in a surroundings of a vehicle
A method recognizes an object in a surroundings of a vehicle. The vehicle has an ultrasonic sensor for monitoring the surroundings of the vehicle and for the transmission and reception of ultrasonic signals. In the method, sensor data generated by the ultrasonic sensor are received. Classification data that contain features to be classified, are subsequently generated from the sensor data. The classification data and/or the sensor data are hereupon entered into a classifier that has been trained with training data for assigning the classification data and/or the sensor data to object classes. Object information that indicates at least one of the object classes is output.
1 . A method for recognizing an object in a surroundings of a vehicle comprising an ultrasonic sensor for monitoring the surroundings of the vehicle and for transmitting and receiving ultrasonic signals, the method comprising:
receiving passive ultrasonic signals with the ultrasonic sensor, the passive ultrasonic signals do not originate from the ultrasonic sensor;
generating ultrasonic sensor data with the ultrasonic sensor based on the received passive ultrasonic signals;
receiving the ultrasonic sensor data generated by the ultrasonic sensor with a control device of the vehicle;
generating classification data that contain features to be classified based on only the ultrasonic sensor data using an evaluation unit of the control device, the classification data including echo data and/or position data;
entering the classification data and/or the ultrasonic sensor data into a classifier unit of the control device;
assigning the classification data and/or the ultrasonic sensor data to at least one object class to generate object information using the classifier unit, the at least one object class corresponding to the recognized object; and
outputting the object information from the control device,
wherein the passive ultrasonic signals include noises in the surroundings of the vehicle, and
wherein the object information identifies when a road on which the vehicle is driving is a wet road and when the road on which the vehicle is driving is a dry road.
2 . The method according to claim 1 , further comprising:
generating the ultrasonic sensor data with the ultrasonic sensor based on echoes of transmitted ultrasonic signals,
wherein the echo data is based on the echoes of transmitted ultrasonic signals, and
the position data is based on reflection sites at which the echoes of transmitted ultrasonic signals were reflected.
3 . The method according to claim 2 , wherein:
the echo data is further based on a transit distance, an intensity, and/or a plausibility of the echoes of transmitted ultrasonic signals, and
the position data is based on coordinates of the reflection sites calculated by trilateration, multilateration, and/or a plausibility of the reflection sites.
4 . The method according to claim 3 , wherein the at least one object class includes pedestrians, objects that can be driven over, and objects that cannot be driven over.
5 . The method according to claim 1 , further comprising:
receiving further ultrasonic sensor data that are generated by the ultrasonic sensor; and
evaluating the further ultrasonic sensor data together with the object information.
6 . The method according to claim 1 , further comprising:
generating repeatedly the classification data at defined time intervals, recording a temporal progress of the classification data, entering the temporal progress of the classification data into the classifier unit; and/or
receiving repeatedly the ultrasonic sensor data at the defined time intervals, recording a temporal progress of the ultrasonic sensor data, and entering the temporal progress of the ultrasonic sensor data into the classifier unit.
7 . The method according to claim 1 , further comprising:
controlling the vehicle and/or at least a component of the vehicle based on the object information.
8 . The method according to claim 1 , wherein a computer program which, when executed on the control device, carries out the method.
9 . The method according to claim 8 , wherein the computer program is stored on a non-transitory machine-readable storage medium.
10 . The method according to claim 1 , wherein the noises in the surroundings of the vehicle include a rolling sound of tires on a wet road and/or a rolling sound of tires on a dry road.
11 . The method according to claim 1 , wherein:
the noises in the surroundings of the vehicle include a venting sound of a compressed air brake,
wherein the object information further identifies when a truck is nearby.
12 . A method for operating an ultrasonic sensor of a vehicle for monitoring a surroundings of the vehicle, the method comprising:
receiving passive ultrasonic signals with the ultrasonic sensor, the passive ultrasonic signals do not originate from the ultrasonic sensor and include noises in the surroundings of the vehicle, the passive ultrasonic signals stored as passive ultrasonic sensor data;
receiving the passive ultrasonic sensor data with a control device of the vehicle;
generating classification data that contain features to be classified based on only the passive ultrasonic sensor data using an evaluation unit of the control device; and
entering the classification data and/or the passive ultrasonic sensor data into a classifier unit of the control device to identify (i) when a road on which the vehicle is driving is a wet road, and (ii) when the road on which the vehicle is driving is a dry road.
13 . The method according to claim 12 , wherein the noises in the surroundings of the vehicle include a rolling sound of tires on a wet road and/or a rolling sound of tires on a dry road.
14 . The method according to claim 12 , wherein:
the noises in the surroundings of the vehicle include a venting sound of a compressed air brake, and
entering the classification data and/or the passive ultrasonic sensor data into a classifier unit of the control device further identifies when a truck is nearby.