DE102011001248A1 - Method for assisting driver of motor car, involves detecting motion information of object by radar measuring device so as to perform the object tracking - Google Patents

Abstract

The method involves providing a radar measuring device (8) for detecting the motion information of an object (2) so as to perform the object tracking (30). A camera (9) is used for detecting the vicinity of the motor vehicle so as to produce the detected object information (10). The object tracking is corrected using the detected object information. An independent claim is included for system for assisting motor car driver.

Description

The invention relates to a method for assisting a driver of a motor vehicle with a driver assistance system, which detects at least one radar measuring device with which movement information of an object that is used for an object tracking and at least one camera is detected with the object information in the environment of the motor vehicle , The invention further relates to a driver assistance system for assisting a driver who is operated according to such a method. The invention also relates to a motor vehicle with such a driver assistance system.

From the German patent application DE 10 2008 011 228 A1 a method for supporting a user of a vehicle is known in which vehicle state variables are detected or determined via sensors of the vehicle, and a camera of the vehicle detects a detection area of a road scene at least in front of the vehicle and outputs image signals. From the German patent application DE 10 2008 023 970 A1 is a method for detecting traffic-related information in a moving vehicle known, wherein image data of a visual sensor and map data of a navigation system for the detection of each pre-evaluated and the results of the pre-evaluation are merged and interpreted. From the German patent application DE 10 2008 028 736 A1 For example, a method for determining a speed gradient for a vehicle on a route is known. From the German patent application DE 10 2009 009 904 A1 For example, a method for identifying objects, in particular traffic signs, is known, wherein the objects are detected by means of a detection unit integrated in a vehicle and the detected objects are compared with stored patterns for identification.

The object of the invention is to provide a method according to the preamble of claim 1, by which the object tracking can be improved.

The object is in a method for assisting a driver of a motor vehicle with a driver assistance system that detects at least one radar measuring device with which movement information of an object that is used for object tracking and at least one camera is detected with the object information in the environment of the motor vehicle are achieved by using the object information captured by the camera to correct the object tracking. The object is, for example, a preceding vehicle. The object may also be the own motor vehicle. During object tracking, the movement or speed of the motor vehicle, in particular of the vehicle in front, is recorded over time and predicted with the aid of estimated values. Through the intelligent fusion of radar measurement and camera acquisition, the prediction or prediction, in particular the generation of new estimates, can be significantly improved during object tracking.

A preferred exemplary embodiment of the method is characterized in that the object information captured by the camera in the environment of the motor vehicle is compared with the movement information detected by the radar measurement device in order to predict the driving behavior of the driver and / or other road users. For example, in a traffic jam warning display, it can be assumed that both the driver and a preceding driver will decelerate, that is, reduce the speed.

A further preferred exemplary embodiment of the method is characterized in that traffic sign information acquired with the camera is used for object tracking. For object tracking, traffic sign information can be used with particular advantage, which brings about a clearly predictable change in the driving behavior of the driver of the motor vehicle and of other road users.

A further preferred exemplary embodiment of the method is characterized in that information acquired via the camera about a speed limit and / or a cancellation of a speed limit is used for the object tracking. With a speed limit, it can be assumed that the driver and other road users will slow down. Similarly, if the speed limit is removed, it can be assumed that the driver and other road users will accelerate.

A further preferred exemplary embodiment of the method is characterized in that light signal information recorded with the camera is used for the object tracking. The light signal information is preferably a light signal of a traffic light system, such as a traffic light. In the case of green, it can be assumed that the speed will not be changed significantly. With red, it can be assumed that braking will take place. For yellow further considerations are required.

A further preferred exemplary embodiment of the method is characterized in that light signal information acquired with the camera is used for object tracking. The light signal information is, for example, light signals from other road users, such as brake light signals or flashing light signals. The flashing light signals may be. Signals of a direction indicator or hazard warning lights act.

A further preferred embodiment of the method is characterized in that in a multi-hypothesis method, first of all different movement models are considered. The movement models are, for example, acceleration models or deceleration models.

A further preferred embodiment of the method is characterized in that it is decided in dependence on the detected with the Radarmesseinrichtung motion information which motion model is used for the object tracking. This can further improve the quality of object tracking.

The invention further relates to a driver assistance system for assisting a driver who is operated according to a method described above.

The invention further relates to a motor vehicle with a previously described driver assistance system.

Further advantages, features and details of the invention will become apparent from the following description in which, with reference to the drawings, various embodiments are described in detail. Show it:

1 a simplified representation of two motor vehicles, one of which is equipped with a driver assistance system, which is operated in accordance with a method according to the invention;

2 a Cartesian coordinate diagram in which the path of one of the motor vehicles is plotted over time and

3 a simplified representation of an object tracking.

In 1 are two vehicles 1 . 2 simplified shown in a side view. The car 1 is with a driver assistance system 5 equipped and drives behind the motor vehicle 2 her, which is therefore also referred to as a preceding vehicle.

The driver assistance system 5 includes an indicated only Radarmesseinrichtung 8th , With the radar measuring device 8th Motion information can be captured. The driver assistance system 5 further includes a camera also just hinted 9 , The camera 9 captures item information in the environment of the motor vehicle 1 , In the illustrated embodiment is with the camera 9 a traffic sign 10 detected. The traffic sign 10 concerns the. Suspension of a speed limit.

The driver assistance system 5 with the radar measuring device 8th and the camera 9 Among other things, it serves to recognize the position of other road users and their movement. Due to measured distances and relative speeds can then by the driver assistance system 5 a reaction will be triggered. The radar measuring device 8th measures the situation to objects, such as the preceding vehicle 2 , In addition, the Radarmesseinrichtung 8th measure the differential speed between two objects, for example, between the two vehicles 1 and 2 ,

An essential aspect of the signal processing of the acquired data is the tracking of object movement over time. Some motion parameters, such as relative acceleration, may be detected by the radar gauge 8th , which preferably comprises a plurality of radar sensors, are not measured. These parameters must therefore be estimated from the available measurements. Deviations due to measurement errors are amplified by subtraction. To avoid errors, state estimation methods are used for object tracking. Here, additional knowledge about the object movement is introduced, so that a convergence of the estimate can be achieved.

In the case of an inaccurate object model, an object loss may erroneously occur. A new object is then initialized even though the metric comes from an object that has already been tracked. This occurs when the readings move too far from the predicted position. As a consequence, the estimation result suffers, because after the reinitialization no recourse is made to the results obtained. Both stochastic causes - the measurement errors - as well as the erroneous description of the deterministic motion model come into question as a reason.

The loss of an object as a result of an abrupt change in acceleration represents a non-negligible factor for future assistance functions Problem dar. Especially in applications for accident prevention such situations with high dynamics are relevant. An object loss and reinitialization then mean the loss of valuable fractions of a second.

The underlying problem in defining a motion model is that there is no access to the inputs of the observed vehicle. Clearly speaking, although a change in the state of motion can be observed on the basis of the measured values, it can not be anticipated on the basis of the pedal operation carried out by the driver. The acceleration or deceleration represents a change in the relative acceleration assumed to be constant and is therefore not in conformity with the motion model. The expansion of the model by corresponding higher moments of motion is possible, but reduces the quality of the estimate.

A correction of the deterministic changes is therefore worthwhile in any case. When accelerating your own vehicle, this can be done using in-vehicle measures. Although there is a change in the actual relative acceleration, the changing proportion of the acceleration is known and a correction of the CA model (constant acceleration) can be made. However, not only does an acceleration of one's own vehicle, but also that of the vehicle in front lead to changes in the relative speed. The driving maneuvers carried out by the other road users would only be known in advance in the case of communication between the vehicles. Since a broad availability of such technology is not yet foreseeable, other methods must be used.

A new approach is therefore the use of camera data for the prediction of future driver behavior. Instead of just adding the radar measurement to additional object information, such as the object width, the camera should also improve motion detection. For this purpose, the recognition of traffic signs and light signals is used for object tracking. A recognition and interpretation of traffic signs is state of the art and is available in today's vehicles. As a rule, the driver is alerted to the currently valid speed limit via the vehicle display.

To improve the object tracking another context is exploited. Because in the event of a change in the speed limit, a change in driver behavior is to be expected. Typically, when a speed limit is lifted, the drivers will accelerate and, at the beginning of such, slow down the vehicle speed. The traffic sign recognition can therefore be used to anticipate an acceleration or deceleration of the other road users and to introduce a correction in the movement model. Analogous to the correction of an acceleration change of the own vehicle, a loss of object can thus be avoided while maintaining a simple CA motion model (constant acceleration).

It is also conceivable for assistance functions in inner-city scenarios. A light-emitting system changing from "green" to "yellow" results in the options for the immediately preceding vehicle to accelerate or decelerate. Both alternatives can be considered first with a multi-hypothesis method and followed by the actually measured radar sensor data before deciding on the more probable model. Ultimately, the vehicle is continuously tracked without object loss due to the sudden change in acceleration.

A third source of information is the taillights of the vehicles in front. Immediately upon actuation of the brake pedal, the brake lights indicate the following traffic on the pending speed change. While the driver of the vehicle perceives this information as a matter of course, it is not yet taken into account in object tracking by environmental sensors. Also in this case, the fusion of the radar data with the light signals detected and interpreted by the camera can improve motion tracking. This principle can be easily extended to the existing in more and more vehicles "adaptive brake light", which also visually announces the strength of the braking. Likewise, activating the hazard warning lights most likely announces a braking maneuver to a standstill.

In summary, object tracking improvement can be achieved through intelligent fusion of camera and radar data. In contrast to previous approaches, in which the measurements are merged at feature level, a change in driver behavior is predicted here.

In 2 is a Cartesian coordinate plot with an x-axis 21 and a y-axis 22 shown. On the x-axis 21 the time is applied in a suitable unit. On the y-axis 22 the way is applied in a suitable unit.

Through a straight, solid line 25 is indicated that the object is moving at a constant speed. As soon as that road sign 10 (please refer 1 ) from the camera 9 of the driver assistance system 5 is detected, it can be assumed that the driver of the tracked object, for example the motor vehicle 2 , will accelerate his vehicle.

According to an essential aspect of the invention, that of the camera 9 captures information about the traffic sign 10 used to, as in 2 through a dotted line 26 is hinted at anticipating driver behavior. The traffic sign recognition allows a prediction of the probable driver behavior. The prediction can be corrected by the additional acceleration. This can prevent unwanted object loss. A new measurement taking into account the prediction 28 is in 2 at the end of the corrected course of movement 26 indicated.

In 3 is the object tracking by a rectangle 30 indicated. By an arrow 31 indicates a previous value used in object tracking. By an arrow 32 a current measured value is indicated. By an arrow 33 is a movement model indicated, which is used in the object tracking. By an arrow 34 is indicated that in the inventive method, the anticipated driver behavior is used in the object tracking. By an arrow 35 is the new estimate 28 (please refer 2 ) indicated.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely 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.

Cited patent literature

• DE 102008011228 A1 [0002]
• DE 102008023970 A1 [0002]
• DE 102008028736 A1 [0002]
• DE 102009009904 A1 [0002]

Claims (10)

Method for assisting a driver of a motor vehicle ( 1 ) with a driver assistance system ( 5 ), the at least one Radarmesseinrichtung ( 8th ), with the movement information of an object ( 2 ) required for object tracking ( 30 ) and at least one camera ( 9 ), with the item information ( 10 ) in the environment of the motor vehicle ( 1 ), characterized in that the with the camera ( 9 ) recorded item information ( 10 ) are used to track the object ( 30 ) to correct. A method according to claim 1, characterized in that the with the camera ( 9 ) recorded item information ( 10 ) in the environment of the motor vehicle ( 1 ) with those of the Radarmesseinrichtung ( 8th ) are compared in order to predict the driving behavior of the driver and / or other road users. Method according to one of the preceding claims, characterized in that with the camera ( 9 ) detected traffic sign information for object tracking ( 30 ) be used. Method according to one of the preceding claims, characterized in that with the camera ( 9 ) information about a speed limit. and / or a cancellation of a speed limit for object tracking ( 30 ) be used. Method according to one of the preceding claims, characterized in that with the camera ( 9 ) detected light sign information for object tracking ( 30 ) be used. Method according to one of the preceding claims, characterized in that with the camera ( 9 ) detected light signal information for object tracking ( 30 ) be used. Method according to one of the preceding claims, characterized in that in a multi-hypothesis method, first of all different movement models are considered. A method according to claim 7, characterized in that depending on the with the Radarmesseinrichtung ( 8th ) is determined motion information, which movement model for object tracking ( 30 ) is used. A driver assistance system for assisting a driver who is operated according to a method according to one of the preceding claims. Motor vehicle with a driver assistance system ( 5 ) according to claim 9.

Images