EP3649522A1

EP3649522A1 - Method for operating a more highly automated vehicle (hav), in particular a highly automated vehicle

Info

Publication number: EP3649522A1
Application number: EP18733188.9A
Authority: EP
Inventors: Carsten Hasberg; Philipp RASP; Danny Hiendriana; Valentin Frommherz; Ali ALAWIEH; Fabian Dominik Reister; Muhammad Sheraz Khan; Jan-Hendrik Pauls
Original assignee: Robert Bosch GmbH
Current assignee: Robert Bosch GmbH
Priority date: 2017-07-07
Filing date: 2018-06-04
Publication date: 2020-05-13
Also published as: DE102017211629A1; US20210139046A1; CN110869869A; WO2019007603A1

Abstract

The invention relates to a method for locating a highly automated vehicle (HAV) in a digital localization map, having the steps of: S1 providing a digital map, preferably a highly precise digital map, in a driver assistance system of the HAV; S2 determining a current vehicle position and locating the vehicle position in the digital map; S3 identifying a route section currently being traversed by the HAV in the digital map, said identification process being carried out at least partly using the current vehicle position and/or using a current change of the current vehicle position; S4 providing at least one comparison trajectory being traversed by at least one additional vehicle along the route section currently being traversed, wherein the additional vehicle has already traveled through the route section currently being traversed and/or the additional vehicle is located ahead of the HAV on the route section currently being traversed; S5 comparing the at least one comparison trajectory with the route section currently being traversed as specified in the digital map and ascertaining a differential value as the result of the comparison; and S6 ascertaining an up-to-dateness of the route section being currently traversed in the digital map at least partly using the differential value. The invention further relates to a corresponding system and to a computer program.

Description

description

title

Method for operating a higher automated vehicle (HAF), in particular a highly automated vehicle

The invention relates to a method for operating a highly automated vehicle (HAF), in particular a highly automated vehicle, and a driver assistance system for controlling a highly automated vehicle (HAF), in particular a highly automated vehicle.

State of the art

In view of an increase in the degree of automation of vehicles ever more complex driver assistance systems are used. For such driver assistance systems and functions, such as highly automated driving or fully automated driving, a large number of sensors in the vehicle is required, which allow an accurate detection of the vehicle environment. In the following, all levels of automation are understood to be more highly automated, which in the sense of the Federal Highway Research Institute (BASt) correspond to automated longitudinal and lateral guidance with increasing system responsibility, eg highly automated and fully automated driving. The prior art discloses a variety of ways to perform a method of operating a highly automated vehicle (HAF). In order to increase the location of the highly automated vehicle (HAF) in a digital map, it is necessary to always be able to guarantee the accuracy of the digital map. A problem in this context, however, is considered to be that relevant information that is stored in the digital map such as information about the entire road construction and / or for example the location of crash barriers, bridges, lane markings and / or traffic signs, in reality extremely short-term can change. If the environment model and digital map show significant deviations, it can be assumed that the map has map errors and can therefore only be used to a limited extent in order to meet the requirement of traffic safety.

In order to control the vehicle more automated in all possible situations, it is therefore necessary to have a largely flawless and reality-compliant digital map available.

It is known that based on various environmental sensors, such as radar sensors, cameras, vehicle dynamics sensors, GPS (Global Positioning System) and digital maps, a representation of the vehicle environment, the so-called environment model, can be constructed by comparing the sensor data or the environment model the validity of a digital map can be validated and, if necessary, increased with the digital map. If the environment model and digital map show significant deviations, it can be assumed that the map is not up-to-date and can only be used to a limited extent.

At this point, there is the problem that the resolution of common sensors in the distance is usually low and therefore the data are subject to a more or less pronounced noise component, which makes a reliable evaluation difficult or even impossible. In the prior art, algorithms for position determination based on the data from environment sensors therefore focus primarily on the perceived with higher security near range.

However, this is a safety deficiency, especially when driving at high speed, since only then a timely response to the often small changes in the route is possible, if sufficiently distant ambient features for card validation can be used. It is also possible to perform certain calculations based on the sensor data with greater accuracy if the features used as a reference are as far away as possible, for example the conclusions about the rotation angles of the sensors used in relation to the alignment of the digital map.

It is therefore an object of the present invention to provide an improved method for operating a higher automated vehicle (HAF), in particular a highly automated vehicle, and an improved one

Driver assistance system for controlling a higher automated vehicle (HAF), in particular to provide a highly automated vehicle with a reliable information on the quality of sensor detections in the remote area is possible, and with the ultimately route changes against a stored in a digital map track status, in short, card errors called, can be detected early and robust, and thus provides an improved validation of a digital map.

Disclosure of the invention

This object is achieved by means of the subject matter of the independent claims. Advantageous embodiments of the invention are the subject of each dependent subclaims.

According to one aspect of the invention, there is provided a method for operating a higher automated vehicle (HAF), in particular a highly automated vehicle, comprising the following steps:

51 provide a digital map, preferably a high accuracy digital map, in a driver assistance system of the HAF;

Determining a current vehicle position and locating the vehicle position in the digital map;

53 Identification of a section of track currently being traveled by the HAF in the digital map, the identification being at least partially wise on the basis of the current vehicle position and / or based on a current change of the current vehicle position takes place;

54 make available at least one driven comparison trajectory of at least one further vehicle along the currently traveled route section, wherein the further vehicle has already traveled the currently traveled route section and / or wherein the further vehicle is on the currently traveled route section in front of the HAF;

55 comparing the at least one comparison trajectory with the currently traveled route section, as indicated in the digital map, and determining a difference value as a result of the comparison; and

56 Determining a topicality of the currently traveled section of track in the digital map at least partially based on the difference value.

According to one embodiment, it is provided that in the event that the difference value exceeds a specified threshold value of a deviation, a request is made to a driver of the HAF to take over the driving task and / or a request is made to a central map server, an update of the digital Map to provide.

According to a further embodiment, it is provided that an information about the height of the difference value and / or the route course is transmitted to the central map server, wherein the central map server transmits this information to further highly automated vehicles, and wherein this transmission preferably takes the form of a map Update of the digital map is done.

Advantageously, step S4 includes determining the at least one driven comparison trajectory using a GPS system integrated into the at least one further vehicle, and / or using the at least one traversed comparison trajectory using at least one suitable comparison trajectory. Neten, in which at least one further vehicle integrated sensor is determined in the context of an odometry calculation.

Preferably, in step S4 several comparison trajectories of several other vehicles are transmitted to the HAF and compared in step S5 with the currently traveled section as indicated in the digital map, wherein the determination of the difference value takes place with the aid of a statistical evaluation of these comparisons. It is advantageous that the at least one comparison trajectory by a

Vehicle-to-vehicle system (V2V) from which at least one further vehicle is transmitted to the HAF and / or that the at least one comparison trajectory is transmitted to a central server computer, wherein the central server computer in particular a vehicle-to-infrastructure system ( V2I) or a cloud system.

In a further embodiment, information from environment sensors of the HAF is used to plausibilize the current vehicle position in the event that the difference value exceeds a defined threshold value of a deviation.

Another object of the present invention is a driver assistance system for controlling a higher automated vehicle (HAF), in particular a highly automated vehicle. The driver assistance system comprises at least one memory module for storing a digital map, preferably a high-precision digital map, wherein the memory module is in particular a memory module or a central server integrated into the HAF. Furthermore, the driver assistance system has a position module for determining a vehicle position of the HAF, an interface for exchanging data with a remote data source, in particular a vehicle-to-vehicle system or a vehicle-to-infrastructure system, and a control device. The position module is preferably a GPS module. Furthermore, the control device is set up to exchange data with the memory module, the position module and the interface and to locate the vehicle position determined by the position module in the digital map. Furthermore, the tax is configured to identify a currently traveled by the HAF route section in the digital map, the identification is at least partially based on the current vehicle position and / or based on a current change in the current vehicle position. According to the invention, it is provided that the interface is set up to receive at least one driven comparison trajectory of at least one further vehicle along the route section currently being traveled. The control device is set up to use the comparison trajectory to compare the at least one comparison trajectory with the currently traveled track section as indicated in the digital map, and to determine a difference value as a result of the comparison.

Advantageously, the control device is set up to determine a comparison trajectory of a section of the route traveled by the HAF and make it available to other vehicles via the interface.

Furthermore, in another embodiment, the control device is set up to determine the at least one driven comparison trajectory using data received via the position module, and / or to set up the at least one driven comparison trajectory using sensor data of at least one suitable sensor in the context of a Determine odometry calculation.

Preferably, the at least one suitable sensor is selected from the group of the following sensors: acceleration sensors, yaw rate sensors, camera sensors, wheel speed sensors, steering angle sensors; and that the control device is adapted to perform the odometry calculation by at least one of the following methods: Inertial Navigation System (INS), visual odometry, vehicle odometry.

A further subject of the present invention is a computer program which comprises a program code for carrying out the method according to the invention when the computer program is executed on a computer. Although the present invention is described below mainly in the context of passenger cars, it is not limited thereto, but can be used with any type of vehicle trucks (HGV) and / or passenger cars (PKW).

Other features, applications and advantages of the invention will become apparent from the following description of the embodiments of the invention, which is illustrated in the figures. It should be noted that the features illustrated are of a descriptive nature only and may be used in combination with features of other developments described above and are not intended to limit the invention in any way.

drawings

The invention will be explained in more detail below with reference to a preferred embodiment, wherein the same reference numerals are used for the same features. The drawings are schematic and show:

1 shows a flowchart of a first embodiment of the method according to the invention;

FIG. 2 shows a schematic representation of the implementation of a second embodiment of the method according to the invention; FIG. and

Fig. 3 is a flowchart of a third embodiment of the method according to the invention.

In step S1 of FIG. 1, a digital map, preferably a high-precision digital map, is provided, which can be done on the device side in a memory module for storing the digital map, wherein the memory module is in particular a memory module integrated in the HAF or a central server. Step S2 involves determining a current vehicle position and locating the vehicle position in the digital map, as is well known in the art. On the device side, this is done according to the invention by means of a position module, the position module preferably having a GPS module.

Module (Global Positioning System) is.

The step identified as S3 in FIG. 1 comprises the identification of a section of the route currently being traveled by the HAF in the digital map, wherein the identification is carried out at least partially on the basis of the current vehicle position and / or on the basis of a current change of the current vehicle position.

In the example of FIG. 2, this currently traveled section comprises two lanes 101, 102 to which two setpoint trajectories 1 10, 1 1 1 are assigned.

This information is stored in the current state of the digital map and is read device-wise by a control device from a memory module in which the digital map is stored. In the example of FIG. 2, the actual route has now changed compared to the state stored in the map. This is characterized by the actual trajectories 1 10 ', 1 1 1'. The deviations between the desired trajectories 1 10, 1 1 1 and the actual trajectories 1 10 ', 1 1 1', for example, be caused by a short-term construction site.

In the prior art, this deviation can be problematic in terms of traffic safety, since a driver assistance system of a more highly automated vehicle may not detect the deviation in time. According to the invention, the provision of at least one driven comparison trajectory of at least one further vehicle along the currently traveled route section is now provided in step S4 of FIG. 1, wherein the further vehicle has already traveled the currently traveled route section and / or the further vehicle is on the currently used section of the route is located in front of the HAF. In this way, the actual trajectories 1 10 ', 11' are now known to the driver assistance system of the HAF. Through the step S5 the comparison of the at least one comparison trajectory with the currently traveled route section, as indicated in the digital map, a difference value can now be determined as the result of the comparison.

In step S6, the determination of an actuality of the currently traveled section of the route in the digital map is now at least partially based on the difference value. In this way, the driver assistance system in the example of Figure 2 is the deviation of the actual trajectories 1 10 ', 1 1 1' of the previously assumed target trajectories 1 10, 1 1 1 before.

If the difference value exceeds a defined threshold value of a deviation, according to an embodiment of the invention, a request can be made to a driver of the HAF to take over the driving task and / or a request can be made to a central map server to make an update of the digital map available. In the example of Figure 2 is in the massive deviation of the actual trajectories 1 10 ', 1 1 1' of the target trajectories 1 10, 1 1 1 assume that the specified threshold value of the deviation is exceeded. The stored in the memory module of the driver assistance system digital map is apparently out of date.

In the example of FIG. 2, this is a route section with two lanes 101, 102. In order to obtain the information that both lanes 101, 102 are affected by the short-term change in the course of the route, in step S4 a plurality of comparison trajectories of a plurality of further vehicles transmitted to the HAF and compared in step S5 with the currently traveled route section, as indicated in the digital map, wherein the determination of the difference value takes place with the aid of a statistical evaluation of these comparisons. In this way, individual lane changes of individual vehicles, for example, from the lane 101 to the lane 102 recognized as such and in the determination of the actual trajectories 1 10 '1 1 1' are filtered out.

FIG. 3 shows a further example of a traffic situation in which the method according to the invention can be used for increasing traffic safety. In this case, the section of road to be traveled on the Traces 101, 102, 103, each with associated desired trajectories 1 10, 1 1 1, 1 12. It can be seen in Figure 3, due to a short-term change in the route, driving on the current section not as deposited in the digital map can be, since a part of the lane 101 is closed to traffic. An actual trajectory 1 10 'extending along the lane 101 therefore has a deviation from the desired trajectory 110 and terminates in the desired trajectory 1 1 1. The desired trajectories running along the lanes 102, 103 are not affected by the traffic change.

The inventive method also detects this situation by accumulation of actual trajectories 1 10 ', 1 1 1', 1 12 'a variety of vehicles and their statistical evaluation, as already explained in connection with Figure 2. An embodiment of the invention provides that the statistical evaluation comprises a classifier, for example a neural network, with which the type of traffic change is detected, for example construction site entrances, laying of individual or all lanes, and / or accidents.

The invention is not limited to the described and illustrated embodiment. Rather, it also encompasses all expert developments within the scope of the invention defined by the claims.

In addition to the described and illustrated embodiments, further embodiments are conceivable which may include further modifications and combinations of features.

Claims

claims

1 . Method for operating a higher automated vehicle (HAF), in particular a highly automated vehicle, comprising the steps:

53 Identification of a section of the route currently being traveled by the HAF in the digital map, the identification taking place at least partially on the basis of the current vehicle position and / or on the basis of a current change in the current vehicle position;

56 Determining a topicality of the currently traveled section of track in the digital map at least partially based on the difference value. A method according to claim 1, characterized in that in the case that the difference value exceeds a predetermined threshold value of a deviation, a request is made to a driver of the HAF to take over the driving task, and / or a request is made to a central map server, an update to provide the digital map.

A method according to claim 2, characterized in that an information about the height of the difference value and / or the course concerning the central map server is transmitted, wherein the central map server transmits this information to other highly automated vehicles, and wherein this transmission preferably in the form of a map Update of the digital card is done.

Method according to one of the preceding claims, characterized in that step S4 includes determining the at least one driven comparison trajectory using a GPS system integrated into the at least one further vehicle, and / or using at least one driven comparison trajectory using at least a suitable, in the at least one further vehicle integrated sensor is determined in the context of an odometry calculation.

Method according to one of the preceding claims, characterized in that in step S4 a plurality of comparison trajectories of a plurality of further vehicles is transmitted to the HAF and compared in step S5 with the currently traveled route section, as indicated in the digital map, wherein the determination of the difference value with the aid of a statistical evaluation of these comparisons.

Method according to one of the preceding claims, characterized in that the at least one comparison trajectory is transmitted by a vehicle-to-vehicle system (V2V) from the at least one further vehicle to the HAF and / or that the at least one comparison trajectory to a central server computer In particular, the central server computer is a vehicle-to-infrastructure (V2I) system or a cloud system. A method according to claim 1, characterized in that in the case that the difference value exceeds a predetermined threshold value of a deviation, information from environment sensors of the HAF are used to make the current vehicle position to plausibility.

Driver assistance system for controlling a higher-automated vehicle, in particular a highly automated vehicle, comprising

A memory module for storing a digital map, preferably a high-precision digital map, wherein the memory module is in particular a memory module integrated in the HAF or a central server;

A position module for determining a vehicle position of the HAF, wherein the position module is preferably a GPS module;

An interface for exchanging data with a remote data source, in particular a vehicle-to-vehicle system or a vehicle-to-infrastructure system; and

A control device that is set up to exchange data with the memory module, the position module and the interface and to locate the vehicle position determined by the position module in the digital map, and is further configured to maintain a section of the road currently being traveled by the HAF identify the digital map, wherein the identification is carried out at least partially based on the current vehicle position and / or based on a current change of the current vehicle position, characterized in that

the interface is set up to receive at least one driven comparison trajectory of at least one further vehicle along the currently traveled route section, wherein the control device is set up to compare the at least one comparison trajectory with the currently traveled route section using the comparison trajectory, as in the digital trajectory Map, and to determine a difference value as a result of the comparison. Driver assistance system according to claim 8, characterized in that the control device is adapted to determine a comparison trajectory of a driven by the HAF track section and to provide over the interface to other vehicles.

0. driver assistance system according to claim 9, characterized in that the control device is adapted to determine the at least one driven comparison trajectory using data received via the position module, and / or is adapted to at least one driven comparison trajectory using sensor data at least of a suitable sensor as part of an odometry calculation.

1 . Driver assistance system according to claim 10, characterized in that the at least one suitable sensor from the group of the following sensors is selected: acceleration sensors, rotation rate sensors, camera sensors, wheel speed sensors, steering angle sensors; and that the control device is adapted to perform the odometry calculation by at least one of the following methods: Inertial Navigation System (INS), visual odometry, vehicle odometry.

2. Computer program comprising program code for carrying out the method according to one of claims 1 to 7, when the computer program is executed on a computer.