DE102019005497A1 - Overtaking assistance system for a motor vehicle for determining a lateral distance from an object to be overtaken, and method - Google Patents

Abstract

The invention relates to an overtaking assistance system (12) for a motor vehicle (10), with a detection device (20) for detecting an object (18) to be overtaken in an environment (32) of the motor vehicle (10), with an electronic computing device (22), which determines a lateral distance (24) of the motor vehicle (10) from the object to be overtaken (18) during the overtaking process as a function of at least one decision criterion and the detected object (18) to be overtaken, and an overtaking signal (26) as a function of the determined lateral distance (24) ) on a display device (28) of the overtaking assistance system (12), the electronic computing device (22) having a trained model (30) which is designed to determine the lateral distance (24) and to output the overtaking signal (26). The invention further relates to a method.

Description

The invention relates to an overtaking assistance system for a motor vehicle according to the preamble of claim 1. The invention also relates to a method.

In road traffic, in particular, it is of crucial importance that corresponding overtaking operations by motor vehicles can be carried out safely in relation to other objects in road traffic. For example, the WO 2017/047038 A1 a system and method for providing driving aids for safely overtaking a vehicle. According to one embodiment, an electronic control unit used in a first vehicle is configured to detect a second vehicle in front of the first vehicle. A first position assigned to the first vehicle and a second position assigned to the detected second vehicle are determined for a first time instance. It can be determined whether a side distance between the determined first position and the determined second position is below a predefined threshold distance. A first alarm is triggered when the determined side distance is below the predefined threshold distance.

The object of the present invention is to provide an overtaking assistance system and a method by means of which an overtaking maneuver of an object to be overtaken can be carried out in an improved manner.

This task is solved by an overtaking assistance system and by a method according to the independent patent claims. Advantageous embodiments are specified in the subclaims.

One aspect of the invention relates to an overtaking assistance system for a motor vehicle, with a detection device for detecting an object to be overtaken, with an electronic computing device which, depending on at least one decision criterion and the detected object to be overtaken, determines a lateral distance from the object to be overtaken during the overtaking process and in Depending on the determined lateral distance, an overtaking signal is output on a display device of the overtaking assistance system.

It is provided that the electronic computing device has a taught-in model, which is designed to determine the lateral distance and to output the overtaking signal.

This makes it possible to use the learned model to learn different situations in road traffic and then to use the learned to determine the lateral distance and to output the overtaking signal. It is therefore a learning overtaking assistance system. As a result, an overtaking process can be improved by means of the motor vehicle.

In the present case, in particular, determining means a prediction or a calculation. The learned model can be designed in particular as a neural network.

According to an advantageous embodiment, the taught-in model is designed to determine the lateral distance on the basis of overtaking processes carried out in the past by the motor vehicle and / or on the basis of a predetermined overtaking model and to output the overtaking signal.

It is also advantageous if the electronic computing device is designed to output the overtaking signal when the specific lateral distance exceeds and / or falls below a predetermined threshold value distance, for example as a result of the neural network.

Furthermore, it has proven to be advantageous if the electronic computing device is designed to at least one parameter characterizing the environment and / or at least one parameter characterizing the motor vehicle and / or at least one parameter characterizing the object to be overhauled and / or a driver of the motor vehicle to take characterizing parameters into account by means of the learned model.

In a further advantageous embodiment, the electronic computing device is designed to continuously determine the lateral distance.
Another aspect of the invention relates to a motor vehicle with an overtaking assistance system. The motor vehicle is designed in particular as a passenger car.

A still further aspect of the invention relates to a method for operating an overtaking assistance system for a motor vehicle, in which an object to be overtaken is detected by means of a detection device and in which an electronic computing device of the overtaking assistance system is used as a function of at least one decision criterion and the object to be overtaken lateral distance of the motor vehicle to the object to be overtaken is determined during the overtaking process and as a function of a certain lateral distance, an overtaking signal is output on a display device of the overtaking assistance system.

It is provided that the lateral distance is determined and the overtaking signal is output by means of a taught-in model, for example a neural network, of the electronic computing device.

Advantageous configurations of the overtaking assistance system are to be regarded as advantageous configurations of the motor vehicle or the method. For this purpose, the overtaking assistance system and the motor vehicle have physical features which enable the method or an advantageous embodiment thereof to be carried out.

Further advantages, features and details of the invention emerge from the following description of preferred exemplary embodiments and from the drawings. The features and combinations of features mentioned above in the description and the features and combinations of features mentioned below in the description of the figures and / or shown alone in the figures can be used not only in the combination indicated in each case, but also in other combinations or on their own, without the scope of Leaving invention.

• 1 eine schematische Draufsicht auf eine Ausführungsform eines Kraftfahrzeugs mit einer Ausführungsform eines Überholassistenzsystems; und
• 2 ein schematisches Ablaufdiagramm für ein Verfahren.

Show:

• 1 a schematic plan view of an embodiment of a motor vehicle with an embodiment of an overtaking assistance system; and
• 2nd a schematic flow diagram for a method.

In the figures, identical or functionally identical elements are provided with the same reference symbols.

1 shows a schematic plan view of an embodiment of a motor vehicle 10th with an embodiment of an overtaking assistance system 12th . The car 10th is on a street 14 and especially on a right lane 16 the street 14 . In front of the motor vehicle 10th there is an object to be overhauled 18th , which is shown in the present embodiment in particular as a bicycle.

The overtaking assistance system 12th has a detection device 20th to capture the object to be overtaken 18th on. For example, the detection device 20th be designed as a lidar sensor and / or as a radar sensor and / or as an ultrasonic sensor and / or as a camera. Furthermore, the overtaking assistance system 12th an electronic computing device 22 on which depending on at least one decision criterion and the detected object to be overtaken 18th in an environment 32 of the motor vehicle 10th a lateral distance 24th of the motor vehicle 10th determined to the object to be overtaken during the overtaking process and depending on the determined lateral distance 24th an overtaking signal 26 on a display device 28 of the overtaking assistance system 12th issues.

It is provided that the electronic computing device 22 a trained model 30th , for example, has a neural network which is designed to determine the lateral distance 24th , especially when driving past, to determine and the overtaking signal 26 to spend.

In particular, the learned model 30th designed to be based on overhauls carried out by the motor vehicle in the past 10th and / or the lateral distance on the basis of a predetermined overtaking model as a decision criterion 24th to determine and the passing signal 26 to spend.

Furthermore, it is particularly provided that the electronic computing device 22 is designed to the overtaking signal 26 output when the certain lateral distance 24th one by the trained model 30th predetermined threshold distance exceeds. In the present exemplary embodiment, the opposite is shown in particular, that is to say that safe overtaking cannot be carried out because the lateral distance 24th is less than the threshold decency. It will be an overtaking signal 26 issued, which indicates that the lateral distance is below the threshold distance, so that a safe overtaking by the motor vehicle 10th is not feasible. In other words, the overtaking signal is in the present exemplary embodiment 26 for a warning signal. Alternatively, a signal can be output as an overtaking signal, which indicates that a safe overtaking is possible.

It is also advantageous if the electronic computing device 22 is designed to at least one the environment 32 characterizing parameters and / or at least one the motor vehicle 10th characterizing parameters and / or at least one object to be overtaken 18th characterizing parameters and / or at least one parameter characterizing a driver of the motor vehicle as a decision criterion by means of the learned model 30th to consider.

Furthermore, it can be provided that the electronic computing device 22 to is formed, the lateral distance 24th to determine continuously.

In particular, the 1 a system for overtaking distance prediction for the maintenance of the lateral safety distance, in particular for the determination of the lateral distance 24th . First of all, individual and / or general data about lateral distances occurring during the overtaking process 24th to the overtaken vehicle, in this embodiment the object 18th , for example via the radar measurement, in particular for bicycles, together with boundary conditions of the overtaking process, for example weather, temperature, number of lanes detected, lane width, oncoming traffic, route guidance, for example as information from the navigation system, visual conditions and extracted from optical vehicle sensors, type of vehicle being overtaken , Motorization of the overtaken vehicle, relative speed, distance to the vehicle to be overtaken and its position relative to the edge of the road are recorded. For example, actuation of the direction indicator, information from a driver assistance system about a lane change and other recognized vehicles and / or the information directly from, for example, optical sensor systems of the motor vehicle can be used to identify / mark an overtaking maneuver 10th be derived. General data is advantageously recorded during development and, if necessary, expanded with fleet data from customer vehicles, for example via a data sensor.

Based on this data, a predictive individual and / or predictive general learned model is created 30th trained using a machine learning method, for example using a neural network, for example by regression and classification. The trained model 30th learns the lateral distance depending on the boundary conditions 24th to predict. If an individual and / or general model with sufficient quality has been trained, this can be used while driving to guide a driver of the motor vehicle 10th to inform. If a vehicle in front, especially the object 18th , for example at low speed, in particular, for example, measured with the optical sensors, radar, lidar or the like, in other words with the detection device 20th , below the maximum speed permitted on this route, for example from speed sign detection and / or navigation data, information about the driving situation, number of lanes, e.g. oncoming traffic, traffic routing, route information, possibly no overtaking sign, possibly lane limitation, is evaluated to determine whether an overtaking process is possible and allowed.

The individual and / or general model continuously, that is to say in particular, predicts the expected lateral distance each time the boundary conditions change 24th to the object to be overhauled 18th . A warning is issued if a safe overtaking while maintaining the legally prescribed lateral distance 24th from the trained model 30th is not predicted as possible, that is, the predicted overtaking distance is smaller than the legally prescribed overtaking distance, an overtaking procedure recommendation may be given while maintaining the minimum distance.

This information can optionally be optically, for example, as a guideline and / or a security box around the object to be overtaken 18th virtually, for example using augmented reality, on a head-up display, on the windscreen of the motor vehicle 10th and / or physically via laser onto the road 16 be projected.

If necessary, the information about distances during the overtaking process can be evaluated to create a driver evaluation, in particular a so-called driver scoring, and to process it for insurance companies, for example. Furthermore, aggregated localized information about too small lateral distances enables 24th dangerous traffic routing / infrastructure and could be marked, for example, in the navigation system and / or made available to the relevant road construction office.

If necessary, acoustic / visual information can be sent to the driver of the motor vehicle 10th be given if the minimum distance was exceeded during the overtaking maneuver.

General models can be continuously improved by the information gathered in the field about distances during the overtaking process and the respective boundary conditions, for example when this data is transmitted to a data center.

In the present exemplary embodiment it can be seen that, for example, on a country road, which in the present example is through the road 14 is shown, a moving bicycle as an object 18th is recognized. The driver of the motor vehicle 10th is already using the learned model 30th recognized and the individual model is already sufficiently well trained to evaluate this situation. The model issues a warning about overhaul or the recommendation not to overtake the cyclist at the moment because the model has learned has that the driver in the current situation, for example driving towards a dome, vehicle engine with low torque, fog, low relative speed to the object 18th , the lateral distance 24th will not keep when overtaking. The driver is waiting, for example. After the dome, the individual model forecast reveals that the object is overtaking 18th while maintaining the lateral distance 24th is possible.

2nd shows a schematic flow diagram of a method according to the invention. In a first step S1 the process begins. In a second step S2 becomes in particular from internal sensors or from external sensors in the motor vehicle 10th an evaluation was carried out to determine whether an object 18th is ahead and, for example, there is no ban on overtaking and / or a traffic situation that makes overtaking impossible.

In a third step S3 it is checked whether a driver of the motor vehicle 10th is clearly identifiable. If this is not the case, the fourth step becomes S4 passed over. In the fourth step S4 becomes a general learned model 30th for example by means of a neural network. If this has already been learned sufficiently, then in a fifth step S5 a forecast of the overtaking distance, in particular the lateral distance 24th , given. If the overtaking distance is greater than the minimum necessary distance, then in the sixth step S6 issued information to the driver that overtaking is possible while maintaining the minimum necessary distance. Should be the lateral distance 24th be smaller than the minimum necessary distance, so in a seventh step S7 issued a warning to the driver that overtaking while maintaining the minimum necessary distance is not possible. In an eighth step S8 the process is stopped again. If, for example, it is determined by the general model that this is not yet sufficiently precise, the fourth step takes place S4 to the eighth step S8 passed over.

Should be in the third step S3 the driver must be identifiable, so in a ninth step S9 checks whether an individually trained model 30th has already been trained sufficiently precisely. If this is not the case, the eighth step becomes S8 passed over. If this is the case, the fifth step will be repeated S5 carried out and the overtaking distance predicted. If the overtaking distance is greater than the minimum necessary distance, the sixth step is also used here S6 passed over. If the overtaking distance is less than the minimum necessary distance, the seventh step is also used here S7 passed over.

Overall, the invention shows a system for overtaking distance prediction for maintaining the lateral distance 24th .

Reference list

10th

Motor vehicle

12th

Overtaking assistance system

14

road

16

roadway

18th

object to be overhauled

20th

Detection device

22

electronic computing device

24th

Lateral distance

26

Overtaking signal

28

Display device

30th

trained model

32

Surroundings

S1

first step

S2

second step

S3

Third step

S4

fourth step

S5

fifth step

S6

sixth step

S7

seventh step

S8

eighth step

S9

ninth step

QUOTES INCLUDE IN THE DESCRIPTION

This list of documents listed by the applicant has been generated automatically and is only included for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent literature cited

• WO 2017/047038 A1 [0002]

Claims (7)

Overtaking assistance system (12) for a motor vehicle (10), with a detection device (20) for detecting an object (18) to be overtaken in an environment (32) of the motor vehicle (10), with an electronic computing device (22), which depending on at least a decision criterion and the detected object (18) to be determined determines a lateral distance (24) of the motor vehicle (10) from the object to be overtaken (18) during the overtaking process and, depending on the determined lateral distance (24), an overtaking signal (26) on a display device (28) of the overtaking assistance system (12), characterized in that the electronic computing device (22) has a taught-in model (30) which is designed to determine the lateral distance (24) and to output the overtaking signal (26). Overtaking assistance system (12) Claim 1 , characterized in that the taught-in model (30) is designed to determine the lateral distance (24) and the overtaking signal as a decision criterion on the basis of overtaking processes carried out in the past by the motor vehicle (10) and / or on the basis of a predetermined overtaking model (26) output. Overtaking assistance system (12) Claim 1 or 2nd , characterized in that the electronic computing device (22) is designed to output the overtaking signal (26) when the determined lateral distance (24) exceeds a threshold value distance predetermined by the taught-in model (30). Overtaking assistance system (12) according to one of the preceding claims, characterized in that the electronic computing device (22) is designed to output the overtaking signal (26) when the specific lateral distance (24) falls below a threshold value distance predetermined by the taught-in model (30) . Overtaking assistance system (12) according to one of the preceding claims, characterized in that the electronic computing device (22) is designed to at least one parameter characterizing the environment (32) and / or at least one parameter characterizing the motor vehicle (10) and / or at least one to take into account the object (18) to be overtaken and / or at least one parameter characterizing a driver of the motor vehicle as a decision criterion by means of the learned model (30). Overtaking assistance system (12) according to one of the preceding claims, characterized in that the electronic computing device (22) is designed to continuously determine the lateral distance (24). Method for operating an overtaking assistance system (12) for a motor vehicle (10), in which an object (18) to be overtaken is detected by means of a detection device (20) and in which the overtaking assistance system (12) is dependent on using an electronic computing device (22) a lateral distance (24) of the motor vehicle (10) from the object to be overtaken (18) during the overtaking process is determined and at least one decision criterion and the detected object (18) to be overtaken, and an overtaking signal (26) as a function of the determined lateral distance (24) a display device (28) of the overtaking assistance system (12) is characterized in that the lateral distance (24) is determined and the overtaking signal (26) is output by means of a learned model (30) of the electronic computing device (22).

Images