WO2020127689A1 - Method for the automated control of a vehicle at a pedestrian crosswalk, control unit - Google Patents

Method for the automated control of a vehicle at a pedestrian crosswalk, control unit Download PDF

Info

Publication number
WO2020127689A1
WO2020127689A1 PCT/EP2019/086244 EP2019086244W WO2020127689A1 WO 2020127689 A1 WO2020127689 A1 WO 2020127689A1 EP 2019086244 W EP2019086244 W EP 2019086244W WO 2020127689 A1 WO2020127689 A1 WO 2020127689A1
Authority
WO
WIPO (PCT)
Prior art keywords
vehicle
pedestrian
crossing
trajectory
time
Prior art date
Application number
PCT/EP2019/086244
Other languages
German (de)
French (fr)
Inventor
Holger Mielenz
Christoph Gustav Keller
Original Assignee
Robert Bosch Gmbh
Daimler Ag
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority to DE102018133157.6A priority Critical patent/DE102018133157A1/en
Priority to DE102018133157.6 priority
Priority to DE102019110071 priority
Priority to DE102019110071.2 priority
Application filed by Robert Bosch Gmbh, Daimler Ag filed Critical Robert Bosch Gmbh
Publication of WO2020127689A1 publication Critical patent/WO2020127689A1/en

Links

Classifications

    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W30/00Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units, or advanced driver assistance systems for ensuring comfort, stability and safety or drive control systems for propelling or retarding the vehicle
    • B60W30/08Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
    • B60W30/095Predicting travel path or likelihood of collision
    • B60W30/0953Predicting travel path or likelihood of collision the prediction being responsive to vehicle dynamic parameters
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W30/00Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units, or advanced driver assistance systems for ensuring comfort, stability and safety or drive control systems for propelling or retarding the vehicle
    • B60W30/08Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
    • B60W30/09Taking automatic action to avoid collision, e.g. braking and steering
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W30/00Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units, or advanced driver assistance systems for ensuring comfort, stability and safety or drive control systems for propelling or retarding the vehicle
    • B60W30/08Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
    • B60W30/095Predicting travel path or likelihood of collision
    • B60W30/0956Predicting travel path or likelihood of collision the prediction being responsive to traffic or environmental parameters
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K9/00Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
    • G06K9/00335Recognising movements or behaviour, e.g. recognition of gestures, dynamic facial expressions; Lip-reading
    • G06K9/00342Recognition of whole body movements, e.g. for sport training
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K9/00Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
    • G06K9/00362Recognising human body or animal bodies, e.g. vehicle occupant, pedestrian; Recognising body parts, e.g. hand
    • G06K9/00369Recognition of whole body, e.g. static pedestrian or occupant recognition

Abstract

The invention relates to a method for the automated performance of a driving function of a vehicle at a pedestrian crosswalk, in which method first a pedestrian trajectory of a pedestrian is predicted and a vehicle trajectory for the crossing of the pedestrian crosswalk is determined. A first time point at which the pedestrian moving on the pedestrian trajectory will reach the pedestrian crosswalk and a second time point at which the vehicle moving on the vehicle trajectory will leave the pedestrian crosswalk are calculated. Subsequently, a crossing decision is made using the first time point and the second time point, and a control signal for a driving function of the vehicle is output on the basis of the crossing decision in such a way that the vehicle moves along the vehicle trajectory.

Description

description

title

Method for automated control of a vehicle on a pedestrian crossing, control device

The invention relates to a method for the automated execution of a driving function of a vehicle at a pedestrian crossing and a control unit for executing the method and a computer program for executing the method and a storage medium for the computer program.

This patent application claims priority from German patent applications DE 10 2018 133 157.6 dated December 20, 2019 and DE 10 2019 110 071.2 dated April 16, 2019, the content of which is incorporated in full by reference into this patent application.

State of the art

Methods are known with which the behavior of a pedestrian when crossing a road can be classified. A pedestrian or a pedestrian environment can be detected using a sensor signal. As a second step, a physical quantity of a connection between the pedestrian and the at least one environment information is determined and the method of the pedestrian is classified on the basis of this quantity in a third step. This makes it possible to make a reliable prediction of a possible pedestrian movement. Such a method is disclosed in the document DE 10 2014 201 159 A1.

It is also known to recognize pedestrians on a crosswalk using environment sensors. For example, camera sensors in a vehicle can do this be installed and images of these camera sensors are evaluated. Such a method is disclosed in DE 10 2014 226 254 A1.

Furthermore, methods for determining the route of a motor vehicle are known, where other vehicles are taken into account in the route determination in order to avoid collisions as far as possible. Such a method is disclosed in US 2012/0072104 A1.

Disclosure of the invention

An object of the invention is to provide a method in which a vehicle trajectory of a vehicle is determined via a pedestrian crossing crossing a roadway and a decision is made to drive over this vehicle trajectory. Another object of the invention is to provide a control device for a vehicle with which the method steps can be carried out.

These objects are achieved with the method and the device of the independent claims. Preferred further developments are given in the claims at under.

A method for automatically executing a driving function of a vehicle at a pedestrian crossing can be carried out in such a way that a pedestrian trajectory of a pedestrian is first predicted. In a further method step, a vehicle trajectory for crossing the pedestrian crossing can be determined. This step can be performed before the pedestrian trajectory prediction or after the pedestrian trajectory prediction. A first point in time at which the pedestrian moving on the pedestrian trajectory will reach the pedestrian crossing and a second point in time at which the vehicle moving on the vehicle trajectory will leave the pedestrian crossing can be calculated. In a further procedural step, a crossing decision can be made on the basis of the pedestrian trajectory and the vehicle trajectory. Alternatively, a crossing decision can be made on the basis of the first point in time and the second point in time. In a last step a control signal for at least one driving function of the vehicle is then output on the basis of the crossing decision in such a way that the vehicle moves along the vehicle trajectory when it is controlled using the control signal.

When predicting the pedestrian trajectory, sensor data can be used to calculate which further route a pedestrian will travel. Among other things, experience can be used.

The crossing decision is positive if it is decided that the vehicle should cross the pedestrian crossing. This is the case if it is possible to cross the pedestrian crossing without endangering or impairing the pedestrian. On the other hand, the crossing decision made is negative if it is decided that the vehicle should not cross the pedestrian crossing. This is the case when it is not possible to cross the pedestrian crossing without endangering or impairing the pedestrian.

When the crossing decision is made on the basis of the pedestrian trajectory and the vehicle trajectory, it can be provided that a positive crossing decision is made if it is determined on the basis of the pedestrian trajectory and the vehicle trajectory that there is no fear of a collision between the pedestrian and the vehicle. Furthermore, the method can also be designed such that a positive crossing decision is only made if a dangerous or uncomfortable situation for the pedestrian can be excluded on the basis of the pedestrian trajectory and the vehicle trajectory.

The method can be carried out in several applications. For example, the method can be carried out when the vehicle is approaching the pedestrian crossing, ie is still in motion. Furthermore, it is possible to carry out the method if the vehicle has come to a standstill in front of the pedestrian and is now to start again to cross the pedestrian crossing. The control signal for at least one driving function of the vehicle can include information about a steering movement, braking and / or acceleration of the vehicle. The method can be used in the context of the different degrees of automation for driver assistance systems defined by the Federal Highway Research Institute. It can be seen before that the driver assistance system assists the driver by taking over either lateral or longitudinal guidance of the vehicle. Furthermore, the method can also be used in partially automated, highly automated or fully automated vehicles, with the driver assistance system taking over both lateral and longitudinal guidance in all degrees of automation. The systems differ in terms of the driver's duty to monitor the driver assistance system. In partially automated vehicles, the driver is obliged to monitor the driver assistance system. In highly automated vehicles, the driver assistance system issues a request to take over the vehicle. In fully automated vehicles, the driver assistance system changes the vehicle to a risk-minimized vehicle state if the driver does not comply with a handover request.

It is also possible that if the vehicle is still moving towards the pedestrian crossing, a negative crossing decision is made, since the vehicle trajectory to be driven and the pedestrian trajectory result in a collision and / or a dangerous and / or uncomfortable situation for the pedestrian would have. In this case, the crossing decision is negative and the at least one driving function of the vehicle is controlled in such a way that the vehicle brakes to let the pedestrian cross the pedestrian crossing and only then crosses the pedestrian crossing itself. The braking can result in a reduction in speed or a complete standstill of the vehicle.

The proposed method can be used for a vehicle with a driver assistance system, which provides functions for the automated execution of driving functions, in which the driver is supported by the system. Alternatively, the proposed method can also be carried out tized vehicles, especially in local public transport.

In one embodiment of the method, a pedestrian crossing is first recognized. In a further step, the pedestrian approaching the pedestrian crossing can be recorded. These two steps can also be carried out simultaneously or in the reverse order. Then who carried out the further process steps.

It is also possible to detect more than one pedestrian, the pedestrians approaching the pedestrian crossing. In this case, a separate pedestrian trajectory is predicted for each pedestrian recorded and the crossing decision is made on the basis of all pedestrian trajectories.

The detection of the pedestrian crossing and the detection of the pedestrian can be carried out as described in document DE 10 2014 226 254 A1. In particular, an environment sensor system with camera sensors can be provided in order to recognize the pedestrian crossing and to detect the pedestrian. The method disclosed in publication DE 10 2014 201 159 A1 can be used to predict the pedestrian trajectories. The content of the documents DE 10 2014 226 254 A1 and DE 10 2014 201 159 A1 relating to these aspects is hereby incorporated in full by reference. The detection can include both the recording and the evaluation of corresponding sensor data.

In one embodiment of the method, the pedestrian crossing can be recognized and / or the pedestrian approaching on the pedestrian crossing can be detected by means of object recognition. In particular, lidar sensors and / or camera sensors can be used. The object recognition can in particular include an evaluation of a camera image on the basis of predetermined patterns.

In one embodiment of the method, a location and a movement vector of the pedestrian are taken into account when predicting the pedestrian trajectories. A speed of the pedestrian can be determined by the motion vector are taken into account, especially if a pedestrian moves faster than walking speed, for example when jogging. Furthermore, it can also be provided to detect changes in the direction of the pedestrian and to predict a new pedestrian trajectory on the basis of a detected change in direction. The latter can be used in particular to recognize that the pedestrian wants to let the vehicle through, for example by pivoting his pedestrian trajectory in such a way that the vehicle has more time to cross the pedestrian crossing or by reducing its speed and thus also more time for the vehicle to cross the pedestrian crossing.

In one embodiment of the method, a first security area around the pedestrian is taken into account when making the crossing decision. As a result, changes in speed of the pedestrian, in particular, which have not been incorporated into the prediction of the pedestrian trajectory, can be intercepted, since appropriate tolerances can thereby be applied. The first security area can include an area around the pedestrian, a size of the area being predetermined by a predetermined first distance from the pedestrian, for example in the range between 30 centimeters and 2 meters, in particular between 50 centimeters and 1 meter. The first safety area then includes the area around the pedestrian, which contains all points that are less than the first distance away from the position of the pedestrian.

In one embodiment of the method, a second security area around the vehicle is taken into account when making the crossing decision. In particular, this makes it possible to compensate for unpredictable unforeseeable factors, such as wet leaves under the drive wheels, for example, when steering the vehicle along the vehicle trajectory, which can lead to a lower possible acceleration. The second security area can comprise an area around the vehicle, a size of the area being predetermined by a predetermined second distance from the pedestrian, for example in the range between 1 meter and 5 meters, in particular between 2.5 meters and 4 meters. The second security area then encompasses the area around the vehicle that contains all points that are less than the second distance from the position of the vehicle. In one embodiment of the method, a dead time and / or an engine torque and / or a coefficient of friction and / or an engine jerk and / or another engine parameter is taken into account when determining the vehicle trajectory. This enables a more precise determination of the vehicle trajectory.

In one embodiment of the method, a first point in time is calculated, the pedestrian moving on the pedestrian trajectory reaching the pedestrian crossing at the first point in time. Furthermore, a second point in time is calculated at which the vehicle moving on the vehicle trajectory leaves the pedestrian crossing. A positive crossing decision is made when the first time is after the second time. If the first point in time is after the second point in time, the pedestrian only reaches the pedestrian crossing when the vehicle has already left the pedestrian crossing. This can ensure that the pedestrian can safely cross the pedestrian crossing after the vehicle has in turn crossed the pedestrian crossing. If the second point in time lies after the first point in time, it can be provided that the vehicle trajectory is changed in such a way that the vehicle brakes before the pedestrian crossing and / or comes to a standstill. If the vehicle is already at the pedestrian crossing, a negative crossing decision means that the vehicle will not start again.

The first safety area around the pedestrian and the second safety area around the vehicle can be taken into account in this embodiment in such a way that a first safety time is subtracted from the first point in time and a second safety time is added to the second point in time and a positive crossing decision is only made if the first point in time minus the first safety time is after the second point in time plus the second safety time.

In one embodiment of the method, the crossing decision made is a positive crossing decision permitting the crossing of the pedestrian crossing if the first point in time is at least one predetermined safety time reserve after the second point in time. The security The time reserve can be, for example, one second, two seconds or five seconds, or a value between one second and five seconds. Furthermore, the safety time reserve can correspond to a sum of the first safety time and the second safety time.

In one embodiment of the method, an intersection of pedestrian trajectory and vehicle trajectory is evaluated in order to make the crossing decision. This evaluation can include, in particular, the point in time at which the vehicle trajectory reaches the intersection and the point in time at which the pedestrian trajectory reaches the intersection.

In one embodiment of the method, a plurality of vehicle trajectories with different driving parameters are determined. Each of the vehicle trajectories determined is evaluated with regard to driving comfort. The crossing decision is made on the basis of the pedestrian trajectory and the multiple vehicle trajectories, wherein the control of the at least one driving function of the vehicle takes place in such a way that the vehicle moves along one of the multiple vehicle trajectories with a predetermined minimum driving comfort. In this case, it can be provided that a positive crossing decision is made when at least one of the vehicle tractors determined enables a positive crossing decision based on the criteria described above. If several of the determined vehicle trajectories enable a positive crossing decision, it can be provided that the vehicle trajectory is selected according to which of the determined vehicle trajectories with a positive crossing decision offers the greatest possible driving comfort. If none of the vehicle trajectories determined with a positive crossing decision have the required minimum driving comfort, it can be provided that the positive crossing decision is revised and the vehicle is not controlled via the pedestrian crossing.

It can be provided that a second point in time is calculated for each vehicle trajectory at which the vehicle moving on the respective vehicle trajectory will leave the pedestrian crossing and the crossing decision is made on the basis of the first point in time and the plurality of second points in time. The different driving parameters can include different accelerations and / or different steering angles, for example for avoidance. The minimum driving comfort can, for example, be defined in such a way that only certain longitudinal and lateral accelerations are permissible for maintaining the minimum driving comfort. It can be provided that an acceleration, in particular a longitudinal acceleration, takes place when the vehicle is started up to a maximum of 1.7 meters per square second. Most people find starting with such an acceleration quick, but safe. For a slower, more comfortable start, it can be seen that the acceleration, in particular the longitudinal acceleration, is a maximum of 1.4 meters per square second. If there is standing space in the vehicle and people are standing in the vehicle, a maximum acceleration of 1.3 meters per square second can be specified. In the latter case, it can be provided by means of sensors, for example cameras, to determine whether standing places of the vehicle are occupied.

In one embodiment of the method, the predefined minimum driving comfort is reduced if a standing time of the vehicle exceeds a predefined value. This can be useful, for example, if, due to the large number of pedestrians crossing the pedestrian crossing, vehicle trajectories suitable for crossing without endangering the pedestrians are calculated, but these are not driven due to the requirement for minimum driving comfort. If the service life is too long and exceeds the specified value, the minimum driving comfort is reduced accordingly so that possible vehicle trajectories can now be driven on. This aspect of the method can also be used several times, with the minimum driving comfort being reduced further and further. For example, vehicle trajectories with a maximum acceleration of 1.4 meters per second square can first be taken into account and then, after a reduction in the minimum driving comfort, vehicle trajectories with a maximum acceleration of 1.7 meters per second square can also be taken into account. The predefined value for the maximum standing time can be determined as a function of a position of the vehicle and a time of day. For example, the specified value can be reduced during rush hour traffic. Rush-hour traffic takes place, for example, in the periods Monday to Friday between 6 a.m. and 9 a.m. and between 4 p.m. and 7 p.m. The specified value for the maximum standing time may be lower on main traffic roads than on secondary roads. For example, the specified value for the maximum standing time can be 30 seconds for a main traffic route during rush hour traffic, 60 seconds for secondary roads during rush hour traffic and 90 seconds outside of rush hour traffic for both main traffic and secondary roads.

It can additionally be provided that a central point determines the utilization of a road network and transmits the predetermined value for the maximum standing time to vehicles that are moving in this road network. The vehicle or a control device of the vehicle can then be set up to receive this predetermined value. It is also possible to receive different values for different positions within the road network.

In one embodiment, a teleoperator of the vehicle can be switched on when a vehicle's standing time exceeds a further predetermined value. This is particularly advantageous for fully automated vehicles, for example in local public transport.

The teleoperator can then possibly intervene manually in the control of the vehicle and thus make a decision regarding the crossing of the pedestrian crossing.

A control unit of a vehicle is set up to carry out one of the processes described.

For this purpose, the control device has in particular an input for at least one sensor signal. Furthermore, the control device has a computing unit for carrying out the method steps and an output for passing on driving information on a system for the automated execution of driving functions.

A computer program comprises instructions that lead to the execution of the method when the computer program is executed on a computer. The computer program includes a machine-readable storage medium.

Embodiments of the invention are explained with reference to the following drawings. The schematic drawing shows:

1 shows a vehicle with a control unit;

2 shows a first traffic situation at a pedestrian crossing;

3 shows a further course of the first traffic situation;

4 shows a further course of the first traffic situation;

5 shows a second traffic situation at a pedestrian crossing;

6 shows a third traffic situation at a pedestrian crossing;

7 shows a further course of the third traffic situation;

8 shows a flowchart of the method; and

9 shows a further flow chart of the method.

1 shows a vehicle 20 with a control unit 22, a sensor 23 and a system 24 for the automated execution of at least one driving function. The control unit 22 is set up to read sensor data from the sensor 23 and to transmit a control signal to the system 24 for the automated execution of at least one driving function. Furthermore, the control unit 22 is set up to carry out a method for automatically controlling a vehicle at a pedestrian crossing. The control unit 22 and the system 24 can thereby also form a unitary component, that is, be out as a unitary component. Furthermore, the control device 22 can comprise a computing unit which is set up to carry out method steps in the context of a computer program. The computer program can be stored on a machine-readable storage medium within the control unit. The system 24 for executing the at least one driving function is set up to control at least one steering movement and / or a speed of the vehicle 20, for example by means of intervention in a steering, accelerating or braking the vehicle 20.

Although only one sensor 23 is shown in FIG. 1, a plurality of sensors 23 can also be assigned to the vehicle 20. The sensors 23 can be set up for pedestrian detection and / or for pedestrian crossing detection and include camera sensors, lidar sensors and / or radar sensors. Furthermore, data from a plurality of sensors 23 can also be fused.

FIG. 2 shows a first traffic situation at a pedestrian crossing 1 of a street 2. A pedestrian 10 approaches the pedestrian crossing 1. A vehicle 20 approaches the pedestrian crossing 1 on street 2 as well. The vehicle 20 has the control unit 22 of FIG. 1 on. Furthermore, the vehicle 20 has a sensor 23 analogous to FIG. 1. The pedestrian crossing 1 can be detected by means of the sensor 23. This can be done, for example, on the basis of a traffic sign 3 which points to the pedestrian crossing 1. Alternatively, this can be done by recognizing a pattern of the pedestrian crossing 1. The pedestrian crossing 1 is shown as a so-called zebra crossing and thus corresponds to the identification of a pedestrian crossing valid in Germany. Alternatively, the pedestrian crossing 1 can also be marked as a solid surface which differs in color from the street 2, by means of two boundary lines crossing the street, by means of two broken boundary lines crossing the street or in such a way that in addition to the marking shown in FIG. 2 Continuous lines across the street are attached. Furthermore, it is possible to attach the strips of the pedestrian crossing 1, which here run parallel to the street 2, at an angle to the street. All of the variants described are intended for pedestrian crossings 1 in the United States of America, depending on State or municipality different. Furthermore, the pedestrian path 1 can also be marked using other markings, which are usual in other parts of the world.

The detection of the pedestrian crossing 1 by means of the sensor 23 can also take place, for example, on the basis of an evaluation of the pattern of the road marking.

The sensor 23 can also detect the pedestrian 10 approaching the pedestrian crossing 1. This can be done for example by means of object detection or by evaluating a movement of a sensor image of the sensor 23. The detection of the pedestrian crossing 1 and the detection of the pedestrian 10 can already be done by the electronics of the sensor 23. Alternatively, the raw data of the sensor 23 can be passed on to the control unit 22, the detection of the pedestrian crossing 1 and the detection of the pedestrian 10 is then carried out by the control unit 22. Furthermore, the control unit 22 is set up to make a prediction of a pedestrian trajectory 11 of the pedestrian 10. In addition, the control unit 22 determines a vehicle trajectory 21 for crossing the pedestrian crossing 1.

It can be provided that when predicting pedestrian trajectory 11, a location and a motion vector of pedestrian 10 are taken into account, and pedestrian trajectory 11 is predicted using these two. When determining the vehicle trajectory 21, it can be provided, as shown in FIG. 2, that the vehicle 20 is still approaching the pedestrian crossing 1, ie has not yet come to a standstill before the pedestrian crossing 1. The control unit 22 is also set up to make a crossing decision in a next method step using the pedestrian trajectory 11 and the vehicle trajectory 21. The crossing decision can be made as a positive crossing decision or a negative crossing decision. If a positive crossing decision is made, the vehicle 20 crosses the pedestrian crossing 1 on the vehicle trajectory 21. If a negative crossing decision is made, the vehicle trajectory 21 of the vehicle is changed such that the vehicle 20 brakes before the pedestrian crossing 1 and / or comes to a standstill is coming. This follows by outputting a control signal to a system for the automated execution of at least one driving function, which for example can correspond to the system 24 of FIG. 1.

Fig. 3 shows a further course of the traffic situation of Fig. 2. Since after evaluating the pedestrian trajectory 11 and the vehicle trajectory 21 was determined by the control unit 22 that a collision between the vehicle 20 and pedestrian 10 is to be feared or a dangerous or uncomfortable situation for the pedestrian 10 would arise if the vehicle simply moved further on the vehicle trajectory 21, the vehicle 20 was braked to a standstill in front of the pedestrian crossing 1. The pedestrian 10 moved on his pedestrian trajectory 11 and entered the pedestrian path 1. A vehicle trajectory 21 for crossing the pedestrian crossing 1 is now calculated in the control unit 22 of the vehicle 20. The vehicle trajectory 21 comes too close to the pedestrian trajectory 11, so that no positive crossing decision can be made in the current traffic situation, since this would still endanger the pedestrian or would still result in a collision.

4 shows the further course of the first traffic situation. The vehicle 20 has stopped in front of the pedestrian crossing 1, the pedestrian 10 has continued to move on the pedestrian crossing 1. The control unit 22 now further calculates a vehicle trajectory 21 on which the vehicle 20 could now cross the pedestrian crossing 1. Since pedestrian 10 would no longer be endangered by this vehicle trajectory 21, there would not be a dangerous situation and / or there would be no fear of a collision, a positive crossing decision is now made and at least one driving function of vehicle 20 based on the positive crossing decision controlled in such a way that the vehicle 20 moves along the vehicle trajectory 21 and thus crosses the pedestrian crossing 1.

Also shown in FIG. 4 is an optional first security area 12 around the pedestrian 10 and an optional second security area 25 around the vehicle 20. It can be provided that the first security area 12 and / or the second security area 25 when the crossing detection occurs. Divorce is considered. The security areas 12, 25 correspond to areas around the pedestrian 10 or the vehicle 20. It can be seen that the security areas 12, 25 are determined by a predetermined distance from the pedestrian 12 or the vehicle 20. When considering whether a collision between pedestrian 10 and vehicle 20 is to be feared, the security areas 12, 25 are also taken into account. In this way, in addition to collisions between pedestrian 10 and vehicle 20, dangerous situations for pedestrian 10 and / or uncomfortable situations for pedestrian 10 can also be avoided.

It can be provided that a dead time and / or an engine torque and / or a friction coefficient and / or an engine jerk and / or another engine parameter are taken into account when determining the vehicle trajectory 21.

The detection of the pedestrian crossing 1 by means of the sensor 23 can take place, for example, by means of object detection of the traffic sign 3. Alternatively or additionally, the pedestrian crossing 1 can be identified by means of a Li-DAR sensor, additional reflectors for the LiDAR signal in front of the pedestrian crossing analogous to the traffic sign 3 can also be provided.

5 shows an alternative further course of the first traffic situation as the second traffic situation. Fig. 5 corresponds partially to Fig. 3, the pedestrian 10 has thus moved to the pedestrian crossing 1 in front of the vehicle 20 be. In contrast to FIG. 3, another pedestrian 13 is also moving towards pedestrian crossing 1 here. The other pedestrian 13 approaching the pedestrian crossing 1 is also detected and a further pedestrian trajectory 14 is predicted. The further pedestrian trajectory 14 also runs over the pedestrian crossing 1. When the crossing decision is made, the control unit 22 now takes into account both the pedestrian trajectory 11 and the further pedestrian trajectory 14. As a result, a positive crossing decision is only made if, in addition to the pedestrian 10, the other pedestrians 13 have crossed the pedestrian crossing 1. In the meantime, should other pedestrians approach pedestrian crossing 1, then these are also recorded and additional pedestrian trajectories for these pedestrians are predicted and taken into account in the crossing decision.

5 also shows further vehicle trajectories 26, 27, 28. In an alternative exemplary embodiment, not only a vehicle trajectory 21, but also the further vehicle trajectories 26, 27, 28 are determined. The driving tool trajectories 21, 26, 27, 28 are illustrated by arrows, with different lengths of the arrows symbolizing a different degree of acceleration. In one embodiment, it is provided that the vehicle trajectories 21, 26, 27, 28 are evaluated with regard to driving comfort. In addition to the pedestrian trajectory 11 and the vehicle trajectory 21, the other vehicle trajectories 26, 27, 28 are also taken into account when making the crossing decision. If a positive crossing decision can be made for one or more of the vehicle trajectory 21, 26, 27, 28, then the one with a certain minimum driving comfort is selected for the vehicle trajectory with a positive crossing decision. If the driving comfort of one of the vehicle trajectories 21, 26, 27, 28 is below the minimum driving comfort and if a positive crossing decision has been made for this vehicle trajectory 21, 26, 27, 28, the vehicle 20 is still not based on this vehicle trajectory 21, 26, 27 , 28 controlled via the pedestrian crossing 1, since the associated vehicle trajectory 21, 26, 27, 28 does not meet the minimum driving comfort.

For example, it may be possible that when evaluating the traffic situation shown in FIG. 5, the further vehicle trajectory 27 with the greatest acceleration would lead to the vehicle 20 crossing the pedestrian walkway 1 after the pedestrian 10 continues to move on his pedestrian trajectory 11 has, could cross before the further pedestrian 13 would be endangered. However, due to the high acceleration of the further vehicle trajectory 27, the control of the automated execution of a driving function is not carried out, since the further vehicle trajectory 27 cannot maintain the minimum driving comfort due to the high acceleration. The other vehicle trajectories 21, 26, 28 lead to a negative crossing decision due to the possible collision with the further pedestrian 13 or endangerment of the further pedestrian 13, so that the vehicle 20 the crossing of the further pedestrian 13 of the pedestrian crossing 1 waits and only then makes a positive crossing decision. A maximum acceleration of the vehicle on the further vehicle trajectory 27 could be, for example, more than 1.7 meters per second square, while a maximum acceleration of 1.7 meters per second square is possible on the other vehicle trajectories 21, 26, 28. Starting with such an acceleration is felt by most people as quick but safe. For a slower, more comfortable start, it can be seen that the acceleration is a maximum of 1.4 meters per square second. If there are 20 standing places in the vehicle and people are standing in the vehicle 20, a maximum acceleration of 1.3 meters per second can be specified. In the latter case, provision can be made to use sensors, for example cameras, to determine whether standing spaces of the vehicle 20 are occupied.

In one exemplary embodiment, it can additionally be provided that the predetermined minimum driving comfort is reduced when a vehicle's service life exceeds a predetermined value. If, for example, the vehicle 20 had already stood in front of the pedestrian crossing 1 for a certain time before the pedestrian 10 reached the pedestrian crossing 1 and the predetermined value for the vehicle's service life had already been exceeded, this would result in the further vehicle trajectory 27 now being reduced due to the reduced Minimum driving comfort is driven. With the further vehicle trajectory 27, the further pedestrian 13 is not endangered or hindered. Consequently, the further vehicle trajectory 27 can be driven on, but leads to a low level of driving comfort for occupants of the vehicle 20. This can be accepted with a correspondingly long service life.

In a further exemplary embodiment, the minimum driving comfort is gradually reduced further, depending on the length of time the vehicle 20 is stationary. The predetermined value for the maximum standing time can be determined as a function of a position of the vehicle 20 and a time of day. For example, the specified value can be reduced during rush hour traffic. Rush-hour traffic takes place, for example, in the periods Monday to Friday between 6 a.m. and 9 a.m. and between 4 p.m. and 7 p.m. The specified value for the maximum standing time may be lower on main traffic roads than on secondary roads. For example, the specified value for the maximum standing time can be 30 seconds for a main traffic route during rush hour traffic, 60 seconds for secondary roads during rush hour traffic and 90 seconds outside of rush hour traffic for both main traffic and secondary roads.

It can additionally be provided that a central point determines the utilization of a road network and transmits the predetermined value for the maximum standing time to vehicles that are moving in this road network. The vehicle 20 or the control unit 22 of the vehicle 20 can then be set up to receive this predetermined value. Different predefined values for different positions within the road network can also be received.

In one exemplary embodiment, a teleoperator is switched on when the vehicle's service life exceeds a further predetermined value.

The teleoperator can then, for example, access driving functions of vehicle 20 from outside and manually instruct system 24 or control unit 22 to carry out certain driving functions.

6 shows a third traffic situation at a pedestrian crossing 1 of a street 2. In contrast to the traffic situation of FIG. 2, here the pedestrian 10 is still further away from the pedestrian crossing 1, so that a positive crossing decision has been made on the basis of the predicted pedestrian trajectory 11 and the vehicle trajectory 21 can and the vehicle 20 can cross the pedestrian crossing 1 before the pedestrian 10 reaches the pedestrian crossing 1.

7 shows the third traffic situation after the vehicle 20 has reached the pedestrian crossing 1, the pedestrian 10 now having arrived on the roadside adjacent to the pedestrian crossing 1.

8 shows a flow chart 30 of the method, a pedestrian trajectory of a pedestrian being predicted in a prediction step 31. In a In the determination step 32, a vehicle trajectory is determined, it being possible for the vehicle trajectory to be determined on the basis of current vehicle and travel parameters. In a decision step 33, a positive or negative crossing decision is made on the basis of the pedestrian trajectory and the vehicle trajectory. In a control step 34, at least one driving function of the vehicle is controlled on the basis of the crossing decision in such a way that the vehicle moves along the vehicle trajectory.

Also shown in FIG. 8 is an optional recognition step 35 in which the pedestrian crossing is recognized. In addition, an optional detection step 36 is shown, in which a pedestrian approaching the pedestrian crossing is detected.

FIG. 9 shows a flow chart 30 of the method, a pedestrian trajectory of a pedestrian being predicted in a prediction step 31. A vehicle trajectory is ascertained in a determination step 32, the vehicle trajectory being able to be ascertained on the basis of current vehicle and travel parameters. In a calculation step 38, a first point in time at which the pedestrian moving on the pedestrian trajectory will reach the pedestrian crossing is calculated. In a further calculation step 39, a second time at which the vehicle moving on the vehicle trajectory is left the pedestrian crossing is calculated. In a decision step 33, a positive or negative crossing decision is made on the basis of the first point in time and the second point in time. In a control step 34, at least one driving function of the vehicle is controlled on the basis of the crossing decision in such a way that the vehicle moves along the vehicle trajectory.

Also shown in FIG. 9 is an optional recognition step 35 in which the pedestrian crossing is recognized. In addition, an optional detection step 36 is shown, in which a pedestrian approaching the pedestrian crossing is detected.

Although the invention has been described in detail by the preferred exemplary embodiments, the invention is not limited to the disclosed examples. limits and other variations thereof can be derived by those skilled in the art without departing from the scope of the invention.

Reference list

1 pedestrian crossing

2 street

3 traffic signs

10 pedestrians

11 pedestrian trajectory

12 first security area

13 other pedestrians

14 more pedestrian trajectories

20 vehicle

21 vehicle trajectory

22 control unit

23 sensor

24 system

25 second security area 26, 27, 28 further vehicle trajectory

30 Flow chart

31 prediction step

32 Investigation step

33 decision step

34 control step

35 recognition step

36 acquisition step

38 calculation step

39 further calculation step

Claims

Expectations
1. A method for the automated execution of a driving function of a vehicle (20) at a pedestrian crossing (1), comprising the following steps:
Prediction (31) of a pedestrian trajectory (11) of a pedestrian (10); Determining (32) a vehicle trajectory (21) for crossing the pedestrian crossing (1);
Calculation (38) of a first point in time at which the pedestrian (10) moving on the pedestrian trajectory (11) will reach the pedestrian crossing (1);
Calculating (39) a second point in time at which the vehicle (20) moving on the vehicle trajectory (21) leaves the pedestrian (1);
Making (33) a crossing decision based on the first time and the second time;
Outputting (34) a control signal for at least one driving function of the vehicle (20) based on the crossing decision in such a way that the vehicle (20) moves along the vehicle trajectory (21).
2. The method as claimed in claim 1, with the additional following steps which are carried out before the prediction (31) of the pedestrian trajectory (11):
Detection (35) of a pedestrian crossing (1);
Detection (36) of the pedestrian (10) approaching the pedestrian crossing (1);
3. The method according to claim 1, wherein a location and a motion vector of the pedestrian (10) are taken into account in the prediction (31) of the pedestrian trajectory (11).
4. The method according to claim 1 to 3, wherein a first security area (12) around the pedestrian (10) at the meeting (33) of the crossing decision is taken into account.
5. The method according to any one of claims 1 to 4, wherein a second security area (25) around the vehicle (20) when meeting (33) the crossing decision is taken into account.
6. The method according to any one of claims 1 to 5, wherein the crossing decision is a positive crossing decision permitting the crossing of the pedestrian crossing (1) if the first point in time is at least a predetermined safety time reserve after the second point in time.
7. The method according to any one of claims 1 to 6, wherein a plurality of vehicle trajectories (21, 26, 27, 28) with different driving parameters, in particular with different accelerations, are determined, the determined vehicle trajectories (21, 26, 27, 28 ) be evaluated with regard to driving comfort, a second point in time being calculated for each of the vehicle trajectories (21, 26, 27, 28) at which the vehicle (21, 26, 27, 28) moving on the respective vehicle trajectory ( 20) the pedestrian crossing (1) is left, the crossing decision being made on the basis of the first point in time and the several second points in time, the control (34) of the at least one driving function of the vehicle (20) taking place in such a way that Vehicle (20) moves along one of the several vehicle trajectories (21, 26, 27, 28) with a predetermined minimum driving comfort.
8. The method according to claim 7, wherein the predetermined minimum driving comfort is reduced when a standing time of the vehicle (20) exceeds a predetermined value.
9. The method according to claim 8, wherein the predetermined value is determined as a function of a position of the vehicle (20) and a time.
10. The method according to any one of claims 7 to 9, wherein a vehicle trajectory (21, 26, 27, 28) with a due to the predetermined minimum driving comfort Acceleration of a maximum of 1.7 meters per square second, in particular of a maximum of 1.4 meters per square second is selected.
11. Control device (22) for a vehicle (20), the control device (22) being set up to carry out one of the methods of claims 1 to 10.
12. Computer program, comprising commands which, when the computer program is executed by a computer, cause the latter to carry out a method according to one of claims 1 to 10.
13. Machine-readable storage medium on which the computer program according to claim 12 is stored.
PCT/EP2019/086244 2018-12-20 2019-12-19 Method for the automated control of a vehicle at a pedestrian crosswalk, control unit WO2020127689A1 (en)

Priority Applications (4)

Application Number Priority Date Filing Date Title
DE102018133157.6A DE102018133157A1 (en) 2018-12-20 2018-12-20 Method for the automated control of a vehicle at a pedestrian crossing, control device
DE102018133157.6 2018-12-20
DE102019110071 2019-04-16
DE102019110071.2 2019-04-16

Publications (1)

Publication Number Publication Date
WO2020127689A1 true WO2020127689A1 (en) 2020-06-25

Family

ID=69143555

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/EP2019/086244 WO2020127689A1 (en) 2018-12-20 2019-12-19 Method for the automated control of a vehicle at a pedestrian crosswalk, control unit

Country Status (1)

Country Link
WO (1) WO2020127689A1 (en)

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20120035846A1 (en) * 2009-04-14 2012-02-09 Hiroshi Sakamoto External environment recognition device for vehicle and vehicle system using same
US20120072104A1 (en) 2009-06-12 2012-03-22 Toyota Jidosha Kabushiki Kaisha Route evaluation device
JP2014093040A (en) * 2012-11-06 2014-05-19 Toyota Motor Corp Collision avoidance support device and collision avoidance support method
DE102014201159A1 (en) 2014-01-23 2015-07-23 Robert Bosch Gmbh Method and device for classifying a behavior of a pedestrian when crossing a roadway of a vehicle and personal protection system of a vehicle
EP3006294A1 (en) * 2013-05-31 2016-04-13 Hitachi Automotive Systems, Ltd. Vehicle control device
DE102014226254A1 (en) 2014-12-17 2016-06-23 Robert Bosch Gmbh Method for operating a motor vehicle, in particular autonomously or partially autonomously driving / driving, signaling device, motor vehicle
US20160335892A1 (en) * 2014-03-10 2016-11-17 Hitachi Automotive Systems, Ltd. System for Avoiding Collision with Multiple Moving Bodies

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20120035846A1 (en) * 2009-04-14 2012-02-09 Hiroshi Sakamoto External environment recognition device for vehicle and vehicle system using same
US20120072104A1 (en) 2009-06-12 2012-03-22 Toyota Jidosha Kabushiki Kaisha Route evaluation device
JP2014093040A (en) * 2012-11-06 2014-05-19 Toyota Motor Corp Collision avoidance support device and collision avoidance support method
EP3006294A1 (en) * 2013-05-31 2016-04-13 Hitachi Automotive Systems, Ltd. Vehicle control device
DE102014201159A1 (en) 2014-01-23 2015-07-23 Robert Bosch Gmbh Method and device for classifying a behavior of a pedestrian when crossing a roadway of a vehicle and personal protection system of a vehicle
US20160335892A1 (en) * 2014-03-10 2016-11-17 Hitachi Automotive Systems, Ltd. System for Avoiding Collision with Multiple Moving Bodies
DE102014226254A1 (en) 2014-12-17 2016-06-23 Robert Bosch Gmbh Method for operating a motor vehicle, in particular autonomously or partially autonomously driving / driving, signaling device, motor vehicle

Similar Documents

Publication Publication Date Title
US10688993B2 (en) Vehicle control system with traffic driving control
US9156473B2 (en) Multi-threshold reaction zone for autonomous vehicle navigation
CN104890595B (en) The device and method of automatic Pilot pot life are predicted in automatic driving vehicle
CN104973071B (en) Persistently establish the device and method and associated vehicle on the boundary of automatic Pilot availability
CN103935361B (en) For autonomous lane changing, process and the valid data flow algorithm surmounting behavior
JP2015140181A (en) Method for minimizing brake intervention based on collision certainty factor
US9308915B2 (en) System and method for warning of a possible collision of a motor vehicle with an object
US8832704B2 (en) In-car-use multi-application execution device
JP2016099713A (en) Automatic driving vehicle system
US8947218B2 (en) Driving support device
US9058746B2 (en) Information processing device associated with vehicle traveling
US8655579B2 (en) Driving assistance device
EP3176046A1 (en) Method and device in a motor vehicle for automated driving
CN106004863B (en) Drive assistance device
US9734390B2 (en) Method and device for classifying a behavior of a pedestrian when crossing a roadway of a vehicle as well as passenger protection system of a vehicle
DE102008060684B4 (en) Method and device for automatic parking of a motor vehicle
CN107207047B (en) Method and device for operating a vehicle
JP4853525B2 (en) Moving region prediction device
EP2657921A1 (en) Method and device for the emergency stopping of a motor vehicle
DE102016104250A1 (en) Gap-based cruise control for an automatic driving system
WO2016094224A1 (en) Autonomous vehicle detection of and response to yield scenarios
DE102013113573A1 (en) Hazardous response decision on a contextual basis
DE102012009297A1 (en) Method for assisting rider when feeding e.g. vehicle, involves proving information, warning and automatic engagement, which results during risk of collision and/or secondary collision with highest priority in priority list
EP2404195B1 (en) Method for automatically detecting a driving maneuver of a motor vehicle and a driver assistance system comprising said method
US8868255B2 (en) Vehicle remote operating system and in-vehicle device

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19832656

Country of ref document: EP

Kind code of ref document: A1