Method for operating a parking assistance system, computer program product, parking assistance system and vehicle

Patent metadata

Field	Value
Publication number	US-12583518-B2
Application number	US-202218284018-A
Country	US
Kind code	B2
Filing date	Mar 24, 2022
Priority date	Mar 25, 2021
Publication date	Mar 24, 2026
Grant date	Mar 24, 2026

Abstract

A method for operating a parking assistance system ( 110 ) for a vehicle ( 100 ) is proposed, which is configured to capture and store a trajectory to be trained, in a training mode (MOD 0 ), and which is configured to follow the stored trajectory by means of the vehicle ( 100 ) in a following mode (MOD 1 ). In order to ascertain whether stored optical features, which are used to orient the vehicle in the following mode, need to be updated, distributions of parameters of the optical features are compared. If the similarity of the compared distributions falls below a predetermined threshold, an update is carried out.

First claim

The invention claimed is:

1 . A method for operating a parking assistance system for a vehicle, comprising:
capturing and storing a trajectory to be trained, in a training mode; and autonomously or semi-autonomously following the stored trajectory by the vehicle in a following mode by controlling a steering apparatus and an automatic gear selection system of the vehicle, wherein the training mode comprises:
A1) manually driving the vehicle along the trajectory,
A2) receiving at least one image of an environment of the vehicle during manual driving,
A3) ascertaining a plurality of optical features in the received image, wherein each optical feature is characterized by at least one parameter, and
A4) storing a data set comprising the ascertained optical features;
wherein the following mode comprises:
B1) receiving at least one current image of the environment of the vehicle while operating the parking assistance system in the following mode,
B2) ascertaining the optical features in the received current image,
B3) ascertaining a first distribution of at least one of the parameters based on the stored data set and ascertaining a second distribution of the parameter based on the ascertained optical features of the current image,
B4) ascertaining a similarity between the first distribution and the second distribution, and
B5) updating the stored data set if the similarity ascertained is less than or equal to a predetermined update threshold;
wherein in the training mode, steps A2)-A4) are carried out for multiple positions along the trajectory to be trained, so that for each of the positions a corresponding data set is stored, wherein in the following mode, steps B3) and B4) are carried out for all stored data sets, and wherein step B5) is carried out when the first distribution and the second distribution have a similarity threshold value above a predetermined similarity threshold.

2 . The method as claimed in claim 1 ,
wherein the parameters of the optical features comprise a respective position of each feature in the image, a classification of the respective feature, a color of the respective feature, a geometric shape of the respective feature, and/or a contrast value of the respective feature.

3 . The method as claimed in claim 1 ,
wherein the similarity between the first and second distribution is ascertained based on a Bhattacharyya distance or a Kullback-Leibler distance.

4 . The method as claimed in claim 1 ,
wherein in the training mode, steps A2)-A4) are carried out for multiple positions along the trajectory to be trained, so that for each of the positions a corresponding data set is stored, and wherein in the following mode, steps B3)-B5) are carried out based on those stored data sets, the corresponding position of which are at a distance from a current position of the vehicle, which is less than or equal to a predetermined distance threshold value.

5 . The method as claimed in claim 1 ,
wherein the similarity threshold value corresponds to a lower similarity than an update threshold value.

6 . The method as claimed in claim 1 ,
wherein step B5) is carried out only for the data set that has a first distribution with a greatest similarity to the second distribution in comparison with all data sets of the trajectory.

7 . The method as claimed in claim 4 ,
wherein, based on a respective time stamp of the images received in the training mode, a first stochastic process of the first distribution of the at least one parameter is ascertained, and wherein based on the respective time stamp of the images received in the following mode, a second stochastic process of the second distribution of the parameter is ascertained, and wherein step B5) is carried out based on a similarity between the first stochastic process and the second stochastic process.

8 . The method as claimed in claim 1 ,
wherein updating the stored data set in step B5) is carried out based on the optical features ascertained in step B2).

9 . The method as claimed in claim 1 ,
wherein updating the stored data set in step B5) comprises replacing the data set with a current data set or replacing at least one optical feature contained in the stored data set or updating at least one parameter of an optical feature contained in the stored data set.

10 . A parking assistance system for a vehicle,
which is configured to capture and store a trajectory to be trained, in a training mode, and which is configured to follow the stored trajectory by the vehicle in a following mode, wherein the parking assistance system comprises:
a reception unit for receiving at least one image of an environment of the vehicle while said vehicle travels along the trajectory to be trained, in the training mode,
a first ascertainment unit for ascertaining a plurality of optical features in the received image, wherein each optical feature is characterized by at least one parameter, and
a storage unit for storing a data set comprising the ascertained optical features,
wherein the reception unit is configured to receive at least one current image of the environment of the vehicle while the vehicle travels along the trajectory in the following mode and the first ascertainment unit is configured to ascertain the optical features in the received current image, and wherein the parking assistance system also comprises:
a second ascertainment unit for ascertaining a first distribution of at least one of the parameters based on the stored data set and for ascertaining a second distribution of the parameter based on the ascertained optical features of the current image,
a comparison unit for ascertaining a similarity between the first distribution and the second distribution, and
an updating unit for updating the stored data set if the ascertained similarity is less than or equal to a predetermined update threshold,
wherein the parking assistance system is configured to control the vehicle to autonomously or semi-autonomously follow the stored trajectory in the following mode by controlling a steering apparatus and an automatic gear selection system of the vehicle, wherein, in the training mode, the reception unit receives the images, the first ascertainment unit ascertains the plurality of optical features, and the storage unit stores the data set for multiple positions along the trajectory to be trained, so that for each of the positions a corresponding data set is stored, wherein, in the following mode, the second ascertainment unit ascertains the first distribution of the at least one of the parameters and the comparison unit ascertains the similarity between the first distribution and the second distribution for all stored data sets, and wherein the updating unit updates the stored data set when the first distribution and the second distribution have a similarity threshold value above a predetermined similarity threshold.

11 . A vehicle having at least one camera for capturing and outputting an image of the environment of the vehicle and having a parking assistance system as claimed in claim 10 .

Assignees

• Valeo Schalter & Sensoren Gmbh

Inventors

• Mohamad Usama

Classifications

• B60W2420/403

  Image sensing, e.g. optical camera · CPC title

• B60W40/02

  related to ambient conditions · CPC title

• B60Y2300/06

  Automatic manoeuvring for parking · CPC title

• G01S2015/932

  for parking operations · CPC title

• G01S2013/9323

  Alternative operation using light waves · CPC title