WO2024008483A1

WO2024008483A1 - Method for operating a vehicle, and vehicle

Info

Publication number: WO2024008483A1
Application number: PCT/EP2023/067168
Authority: WO
Inventors: Stefan GLÄSER; Monique ENGEL; Michael Weng
Original assignee: Volkswagen Aktiengesellschaft
Priority date: 2022-07-08
Filing date: 2023-06-23
Publication date: 2024-01-11
Also published as: DE102022207004A1

Abstract

The invention relates to a method for operating a vehicle (50), wherein the presence of an emergency situation (40) involving the vehicle (50) is detected, wherein, after the presence of an emergency situation (40) has been detected: a) a danger zone (20) is determined in the form of at least one future trajectory (22-x) or a trajectory swarm of possible future trajectories (22-x) of the vehicle (50), and wherein b1) information (21) describing the danger zone (20) is communicated to other vehicles (60-1) and/or infrastructure devices (61) in the surroundings of the vehicle (50) by way of a communication device (53), and/or b2) a warning signal (30) is output to warn other road users (60-x) impacted by the danger zone (20).

Description

Method of operating a vehicle and vehicle

The invention relates to a method for operating a vehicle and a vehicle.

Both in manually controlled vehicles and in partially or fully automated vehicles, emergency situations can arise that must be handled appropriately.

From US 2021/0031800 A1 a method is known in which it is possible to provide information to a trajectory sent via V2X that this is an emergency trajectory, which leaves little or no room for negotiation.

The invention is based on the object of creating a method and a vehicle with which emergency situations can be responded to in an improved manner. In particular, other road users should be informed that an emergency situation exists.

The object is achieved according to the invention by a method with the features of patent claim 1 and a vehicle with the features of patent claim 10. Advantageous embodiments of the invention result from the subclaims.

In particular, a method for operating a vehicle is provided, wherein the presence of an emergency situation of the vehicle is recognized, wherein after recognizing the existence of an emergency situation: a) a danger area in the form of at least one future trajectory or a set of possible trajectories future trajectories of the vehicle is determined, and wherein b1) information describing the danger area is transmitted to other vehicles and / or infrastructure facilities in the vicinity of the vehicle by means of a communication device, and / or b2) a warning signal to warn other road users who are affected by the danger area , is output. Furthermore, in particular a vehicle is created, comprising a driver assistance system, wherein the driver assistance system is set up to recognize the existence of an emergency situation of the vehicle and, after recognizing the existence of an emergency situation: a) a danger zone in the form of at least one future trajectory or a set of trajectories of possible future trajectories of the vehicle, and b1) to transmit information describing the danger area to other vehicles and/or infrastructure facilities in the area surrounding the vehicle by means of a communication device of the vehicle, and/or b2) to issue a warning signal to warn others Road users who are affected by the danger zone.

The method and the vehicle make it possible to provide improved warning of those surrounding the vehicle in the event of an emergency situation. Other road users in the area can then react to the emergency situation and, for example, adapt their behavior to the behavior of the vehicle. For this purpose, it is intended that the existence of an emergency situation in the vehicle is recognized. If the existence of an emergency situation has been recognized, a danger zone is determined in the form of at least one future trajectory or a set of trajectories of possible future trajectories of the vehicle. The at least one future trajectory can be determined, for example, by means of a trajectory planner of the driver assistance system and/or obtained from it. Typically, such a trajectory planner determines a large number of possible future trajectories that the vehicle can execute in the current environment, based on recorded sensor data of the environment and an environment model generated therefrom. The danger zone includes in particular those trajectories that the vehicle can implement in the environment up to a predetermined time horizon and/or spatial horizon. In particular, a set of trajectories can be generated from a desired trajectory (e.g. merging into the neighboring lane), a planned trajectory and other possible trajectories. As a rule, the trajectories are linked to a probability of occurrence and/or another evaluation variable with which a cost value is assigned to the trajectories, for example based on a cost function. The cost function can include and/or represent both the costs of the vehicle and the costs of other road users. For example, accelerations, steering angles, rule-compliant or non-compliant behavior (e.g. driving on the road or leaving the road) etc. can be taken into account. Using one of the Using the cost function calculated for a trajectory (total) cost value, the trajectory planner can then select an optimal plan trajectory that should be implemented. Information describing the specific danger zone is transmitted to other vehicles and/or infrastructure facilities in the area surrounding the vehicle using a communication device of the vehicle. In particular, vehicle-to-everything (V2X) and/or car-to-car (C2C) communication is used. In principle, other suitable, in particular wireless, communication methods can also be used, such as WLAN, Bluetooth, mobile communications (LTE, 5G, ...) etc. Alternatively or additionally, a warning signal is used to warn other road users who are affected by the danger zone. issued and/or caused the issue to be issued. Such a warning signal can, for example, be or include an acoustic warning signal (generated, for example, by means of an acoustic signal generator, for example a horn, a loudspeaker, etc.) and/or an optical warning signal (generated, for example, by means of an optical signal generator, for example warning lights, flashing lights, etc.). . In particular, it is provided that, based on the specific danger area, other road users in the danger area are determined in order to specifically warn them.

An advantage of the method and the vehicle is that the area surrounding the vehicle, in particular other road users in the area, can be warned better. By communicating the danger zone, other road users have the direct opportunity to adapt their own behavior and planning to the existing emergency situation of the vehicle. By communicating the danger zone, other road users also gain immediate insight into the type and extent of the vehicle's emergency situation. In the case of a vehicle that is unable to maneuver, the future trajectories transmitted show in particular that the vehicle has little or no scope to change its behavior due to the existing emergency situation. As a result, other road users who are affected by the danger area can then adapt their behavior accordingly. For this purpose, they can, for example, adapt cost functions used in trajectory planning in order to prefer behavior that is adapted to the emergency situation of the vehicle or to evaluate it at lower costs.

When determining the danger zone, it is estimated and/or predicted (predicted) in particular with the help of estimated trajectories of a trajectory planner of the driver assistance system. In particular, vehicle properties (e.g. mass, wheelbase, dimensions, engine size, etc.), vehicle condition (e.g Vehicle age, vehicle load, detected damage, e.g. to the wheels, etc.), driving physics (e.g. coefficients of friction, cornering forces, etc.), road characteristics (e.g. a road surface, curbs, a road course, inclines/declines, etc.) and/or a current road condition (e.g. wetness, slippery ice, etc.) must be taken into account. Map data can also be taken into account, for example to include a road (ahead).

A vehicle is in particular a motor vehicle. In principle, the vehicle can also be another land, rail, water, air or space vehicle, for example an air taxi or a drone.

Parts of the driver assistance system can be designed individually or collectively as a combination of hardware and software, for example as program code that is executed on a microcontroller or microprocessor. However, it can also be provided that parts are designed individually or combined as an application-specific integrated circuit (ASIC) and/or field-programmable gate array (FPGA).

In one embodiment it is provided that at b1) a message is additionally transmitted to other vehicles and/or infrastructure facilities in the vicinity of the vehicle by means of the communication device, which message includes information that the vehicle is in an emergency situation. This can ensure backwards compatibility. Other vehicles, other road users and/or infrastructure facilities that cannot receive trajectories but only simple messages can then still be informed about the existence of the emergency situation. Knowledge of the emergency situation allows, for example, another vehicle to switch to a more defensive driving style and/or reduce speed.

In one embodiment it is provided that after recognizing the existence of the emergency situation, it is checked whether the vehicle can still be brought out of the recognized emergency situation, with measures b1) and / or b2) only being carried out if this is not the case case is. This allows further information, for example from an interior of the vehicle, to be taken into account. For example, the behavior of a driver and/or passengers of the vehicle can be monitored. If the driver and/or a passenger tries, for example, to defuse the emergency situation, the execution of measures b1) and/or b2) can be delayed, for example. Can the If an emergency situation is defused, measures b1) and/or b2) will not be carried out. However, if the emergency situation cannot be defused, measures b1) and/or b2) are carried out.

In one embodiment it is provided that the future trajectories of the set of trajectories are selected such that an occurrence probability assigned to the future trajectories reaches or exceeds a predetermined threshold value or that a cost value assigned to the future trajectories falls below a predetermined threshold value. This allows the number of trajectories to be transmitted to be limited. This is based on the idea that trajectories with a lower probability of occurrence or with high cost values are unlikely to occur and are therefore less relevant for assessing and responding to the emergency situation of the vehicle. In particular, this can reduce the volume of data that has to be transmitted.

In one embodiment it is provided that driving behavior is monitored to detect an emergency situation. For this purpose, in particular the behavior of the vehicle in relation to the surroundings can be evaluated. For example, sensor data from one or more environmental sensors of the vehicle can be evaluated for this purpose. For example, an impending collision can be detected. Alternatively or additionally, the trajectory planner can determine that a planned trajectory is on a collision course with the (estimated or obtained) trajectories of other road users. Further examples are leaving the road, for example leaving the lane can be detected, with it being recognized that this took place without using a turn signal. It can also be recognized that the vehicle is driverless, for example if a human driver can no longer control the vehicle for health reasons or because an automated control/regulation system (self driving system, SDS) has failed. Furthermore, it can also be determined that an emergency stop has occurred, for example if it is or has been determined that the vehicle is or has come to a standstill on a hard shoulder of a motorway or a country road. The sensor and/or control data can then be evaluated, for example, using pattern recognition methods and machine learning and artificial intelligence methods. The pattern recognition methods and/or methods of machine learning and artificial intelligence can be trained, for example, with the help of training data obtained empirically and/or in simulation. In one embodiment it is provided that the driving dynamics of the vehicle are monitored in order to detect an emergency situation. For example, sensor data and/or control data from an anti-lock braking system (ABS), an electronic stability program (ESP) and/or other control and/or regulation systems of the vehicle dynamics control system, such as a lane departure warning system, can be evaluated. The data can then be evaluated, for example, using pattern recognition methods and machine learning and artificial intelligence methods. The pattern recognition methods and/or methods of machine learning and artificial intelligence can be trained, for example, with the help of training data obtained empirically and/or in simulation.

In one embodiment it is provided that a vehicle-driving entity is monitored to detect an emergency situation. For example, vital functions and/or behavior/characteristics of a human driver can be monitored. For this purpose, one or more sensors can be provided, for example an interior camera, steering wheel sensors and/or seat sensors. The sensor data recorded by these sensors can then be evaluated, for example, using pattern recognition methods and machine learning and artificial intelligence methods. For example, a head position, a head tilt, a degree of eyelid opening, a line of sight and/or a temperature of visible skin surfaces can be evaluated. Using a steering wheel sensor, for example, a pulse, skin moisture and/or skin temperature can be recorded, which can then also be evaluated to determine the driver's vital functions. However, the vehicle-driving entity can also be a system for automated control/regulation of longitudinal and lateral guidance of the vehicle (self-driving system, SDS). Such an SDS can then be monitored using control and/or measurement data to ensure the functionality of the SDS. Here too, pattern recognition methods and machine learning and artificial intelligence methods can be used to detect a malfunction. The pattern recognition methods and/or methods of machine learning and artificial intelligence can be trained, for example, with the help of training data obtained empirically and/or in simulation.

In one embodiment it is provided that when determining the danger zone, the behavior of other road users in the area surrounding the vehicle is taken into account. This allows the danger zone to be determined taking into account a possible change in the behavior of these other road users. It can For example, the case may arise that another road user has already recognized the existence of the vehicle's emergency situation and subsequently adapts his own behavior, for example by reducing a speed or using and/or executing a different plan trajectory. In order to take the behavior of other road users into account, sensor data from the vehicle's surrounding sensors are evaluated, for example using pattern recognition methods and/or machine learning and artificial intelligence methods.

In a further embodiment it is provided that in order to take into account the behavior of other vehicles in the environment, current and/or future trajectories of the other vehicles in the area are received and/or queried. This allows the current and planned behavior of the other vehicles to be taken into account in an efficient manner. In particular, this embodiment allows time to be saved, since the current and/or future trajectories of the other vehicles do not have to be estimated in a time-consuming and computationally intensive manner based on sensor data from the vehicle's surroundings sensors, but can be obtained directly from the other vehicles. In particular, both desired and planned trajectories of the other vehicles can be received and/or queried.

Further features regarding the design of the vehicle result from the description of embodiments of the method. The advantages of the vehicle are the same as in the embodiments of the method.

The invention is explained in more detail below using preferred exemplary embodiments with reference to the figures. Show here:

Fig. 1 is a schematic representation of an embodiment of the vehicle;

Fig. 2 is a schematic representation to illustrate an emergency situation and one

danger zone;

Fig. 3 shows a schematic flow diagram of an embodiment of the method for operating a vehicle. 1 shows a schematic representation of an embodiment of the vehicle 50. The vehicle 50 includes a driver assistance system 1. The method described in this disclosure is explained in more detail below using the vehicle 50.

The driver assistance system 1 comprises a computing device 2 and a memory 3. The driver assistance system 1 is set up to recognize the presence of an emergency situation of the vehicle 50 and, after recognizing the existence of an emergency situation: a) a danger zone 20 in the form of at least one future trajectory 22-x (Fig. 2) or a set of trajectories of possible future trajectories 22-x of the vehicle 50, and b1) information 21 describing the danger area 20 by means of a communication device 53 (in particular V2X and / or C2C) of the vehicle 50 to transmit to other vehicles 60-1 and/or infrastructure facilities 61 in the vicinity of the vehicle 50, and/or b2) to cause a warning signal 30 to be issued to warn other road users 60-x who are affected by the danger zone 20.

The warning signal 30 is emitted, for example, by means of an acoustic warning device 54 (e.g. acoustic signal generator) of the vehicle 50 (e.g. a horn or an outside loudspeaker) and/or by means of a visual warning device 55 (e.g. optical signal generator) of the vehicle 50 (e.g. a turn signal and/or or other warning lights).

To detect the emergency situation, the driver assistance system 1 is supplied with, for example, sensor data 10 from environmental sensors 51 of the vehicle 50 and/or control and measurement data 11 from control devices 52 of the vehicle 50. These are then evaluated by the computing device 2 and analyzed, for example, for patterns indicating the existence of an emergency situation. Such patterns can be determined, for example, with the help of empirical test series and/or simulations. Examples of emergency situations include: the driver losing consciousness, an impending collision of the vehicle with an obstacle or another vehicle, the vehicle being unable to maneuver, or an emergency stop. When evaluating, machine learning and artificial intelligence methods can also be used, which were trained with the help of empirical and/or simulation-generated data. The danger zone 20 is determined in particular by means of a trajectory planner (not shown), which can be part of the driver assistance system 1 and/or can be provided by it. The danger area 20 contains a set of trajectories, the set of trajectories containing future trajectories of the vehicle 50, which the vehicle 50 executes with a probability of occurrence assigned to them.

2 shows a schematic representation of a danger area 20 in the surroundings of the vehicle 50 based on an exemplary emergency situation 40. In the example, the emergency situation 40 should be a possible collision of the vehicle 50 with another vehicle 60-1 in the oncoming lane. In the exemplary representation, the danger area 20 includes three trajectories 22-x, each of which is assigned the occurrence probabilities of 40%, 30% and 30% from left to right of the representation. These three trajectories 22-x can be transmitted to the other vehicle 60-1 according to the method. Alternatively, the trajectories 22-x can also be assigned a respective cost value for evaluation.

It can be provided that in b1) a message 22 is additionally transmitted to other vehicles 60-x and/or infrastructure facilities 61 in the vicinity of the vehicle 50 by means of the communication device 53, which message includes information that the vehicle 50 is in an emergency situation. Situation 40 is located.

It can be provided that after recognizing the existence of the emergency situation 40, it is checked whether the vehicle 50 can still be brought out of the recognized emergency situation 40, with measures b1) and / or b2) only being carried out if this is not the case.

It can be provided that the future trajectories 22-x of the set of trajectories are selected in such a way that an occurrence probability assigned to the future trajectories 22-x reaches or exceeds a predetermined threshold value or that a cost value assigned to the future trajectories 22-x reaches a predetermined threshold value falls below. For example, trajectories 22-x with a probability of occurrence below a predetermined threshold of 1%, 5% or 10% cannot be taken into account. Corresponding threshold values can be provided for the cost values. It can be provided that driving behavior is monitored to detect an emergency situation 40.

It can be provided that the driving dynamics of the vehicle 50 are monitored in order to detect an emergency situation 40. For this purpose, the driver assistance system 2 can evaluate in particular the sensor data 10 and control and measurement data 11.

It can be provided that a vehicle-driving entity is monitored to detect an emergency situation 40. For example, interior sensors 56 may be provided that monitor the vital functions of a human driver. If the vehicle-driving entity is an automated driving system (SDS), the system's control data can be monitored.

It can be provided that when determining the danger zone 20, the behavior of other road users 60-x in the surroundings of the vehicle 50 is taken into account. For this purpose, the driver assistance system 2 can evaluate in particular the sensor data 10 and control and measurement data 11.

In a further development, it can be provided that current and/or future trajectories (not shown) of the other vehicles 60-1 in the environment are received and/or queried in order to take into account the behavior of other vehicles 60-1 in the environment.

3 shows a schematic flowchart of an embodiment of the method for operating a vehicle. The method is carried out in particular by means of a vehicle and a driver assistance system according to one of the embodiments described above.

In a measure 100 it is recognized and checked whether the vehicle is in an emergency situation or not. If there is no emergency situation, measure 100 is repeated.

However, if the existence of an emergency situation has been recognized, a danger zone is determined in a measure 101 in the form of at least one future trajectory or a set of trajectories of possible future trajectories of the vehicle. This is done in particular using a trajectory planner, which can also be part of the driver assistance system. The trajectory planner in particular determines several trajectories and estimates a respective probability of occurrence for the trajectories. In a measure 102, information describing the danger zone is transmitted to other vehicles and/or infrastructure facilities in the vicinity of the vehicle by means of a communication device.

Alternatively or additionally, a warning signal is issued in a measure 103 to warn other road users who are affected by the danger area. This can be done, for example, by means of an acoustic signal generator (e.g. horn, loudspeaker, signal horn, etc.) and/or optical signal generator (e.g. turn signals, lights, floor projection, etc.).

As part of measure 102, it can be provided that a message is additionally transmitted by means of the communication device to other vehicles and/or infrastructure facilities in the vicinity of the vehicle, which includes information that the vehicle is in an emergency situation.

Furthermore, it can be provided in a measure 101a that after the existence of the emergency situation has been recognized, it is checked whether the vehicle can still be brought out of the recognized emergency situation, with measures 102 and/or 103 only being carried out if this is not the case the case is.

Further embodiments of the method have already been described above with reference to the vehicle.

Reference symbol list

Driver assistance system computing device memory

Sensor data

Control and measurement data

Danger room

Information -x trajectory warning signal emergency situation vehicle

Surroundings sensor(s) Control unit(s) Communication device Acoustic warning device Visual warning device Interior sensor(s) -x other road users -1 other vehicle

I nfrastruc ture arrang ement 0-103 Measures of the procedure

Claims

Claims Method for operating a vehicle (50), wherein the existence of an emergency situation (40) of the vehicle (50) is detected, wherein after detection of the existence of an emergency situation (40): a) a danger area (20) in the form is determined by at least one future trajectory (22-x) or a set of trajectories of possible future trajectories (22-x) of the vehicle (50), and b1) information (21) describing the danger area (20) by means of a communication device (53 ) is transmitted to other vehicles (60-1) and/or infrastructure facilities (61) in the vicinity of the vehicle (50), and/or b2) a warning signal (30) to warn other road users (60-x) who are coming from the danger area ( 20) are affected is issued. Method according to claim 1, characterized in that at b1) a message is additionally transmitted to other vehicles (50) and/or infrastructure facilities (61) in the vicinity of the vehicle (50) by means of the communication device (53), which message includes information that the vehicle (50) is in an emergency situation (40). Method according to claim 1 or 2, characterized in that after recognizing the existence of the emergency situation (40), it is checked whether the vehicle (50) can still be brought out of the recognized emergency situation (40), the measures b1) and/or b2) only be executed if this is not the case. Method according to one of the preceding claims, characterized in that the future trajectories (22-x) of the set of trajectories are selected such that a probability of occurrence assigned to the future trajectories (22-x) reaches or exceeds a predetermined threshold value or that one of the future trajectories (22-x) each assigned cost value falls below a predetermined threshold value. Method according to one of the preceding claims, characterized in that driving behavior is monitored to detect an emergency situation (40). Method according to one of the preceding claims, characterized in that a driving dynamics of the vehicle (50) is monitored to detect an emergency situation (40). Method according to one of the preceding claims, characterized in that a vehicle-driving entity is monitored to detect an emergency situation (40). Method according to one of the preceding claims, characterized in that when determining the danger zone (20), the behavior of other road users (60-x) in the surroundings of the vehicle (50) is taken into account. Method according to claim 8, characterized in that in order to take into account the behavior of other vehicles (60-1) in the environment, current and/or future trajectories of the other vehicles (60-1) in the area are received and/or queried. Vehicle (50), comprising: a driver assistance system (1), wherein the driver assistance system (1) is set up to detect the existence of an emergency situation (40) of the vehicle (50) and, after detecting the existence of an emergency situation (40 ): a) to determine a danger area (20) in the form of at least one future trajectory (22-x) or a set of trajectories of possible future trajectories (22-x) of the vehicle (50), and b1) the danger area (20) to transmit descriptive information (21) to other vehicles (60-1) and/or infrastructure facilities (61) in the vicinity of the vehicle (50) by means of a communication device (53) of the vehicle (50), and/or b2) issuing a warning signal (30) to warn other road users (60-x) who are affected by the danger area (20).