Method and control unit for providing a proximity warning about a latent risk of collision between a vehicle and a following vehicle

Abstract

The present invention relates to a method for providing a proximity warning (100) about a latent risk of collision between a vehicle (102) and a following vehicle (104), a respective warning area (110) being defined around the following vehicle (104) and around the vehicle (102), the warning area (110) for at least one of the vehicles (102, 104) being defined to be larger than a collision warning area (112) of the respective vehicle (102, 104) in at least one spatial direction, the proximity warning (100) being provided when the warning areas (110) overlap.

Description

[0001] Description

[0002] title

[0003] Procedures and to provide a before a

[0004] Field of the invention

[0005] The invention relates to a method for providing an approach warning of a latent collision risk between a vehicle and a following vehicle, a corresponding control unit, and a corresponding computer program product.

[0006] State of the art

[0007] A vehicle may have a collision warning system. The collision warning system may, for example, warn a driver of the vehicle if an impending collision between another vehicle or object and the vehicle can be averted by a driver action.

[0008] DE 10 2017 219 892 A1 describes a method and a device for controlling the movement of a single-track motor vehicle.

[0009] Disclosure of the invention

[0010] Against this background, the approach presented here presents a method for providing an approach warning of a latent collision risk between a vehicle and a following vehicle, a corresponding control unit, and a corresponding computer program product according to the independent claims. Advantageous developments and improvements of the approach presented here emerge from the description and are described in the dependent claims. Advantages of the invention

[0011] If another vehicle violates a collision warning zone around a vehicle, there is a high risk that the vehicles will touch, resulting in a collision. If the other vehicle passes very close to the vehicle but does not violate the collision warning zone around the vehicle, no collision warning will be issued. However, the other vehicle may still come close enough to the vehicle that a driver of the vehicle will be startled and possibly make a rash steering movement, leading to a dangerous situation.

[0012] In the approach presented here, an enlarged warning area is defined around the vehicle in addition to the collision warning area. This means that the warning area is larger than the collision warning area in at least one direction. If the other vehicle enters the warning area but not the collision warning area, a proximity warning is provided to the driver.

[0013] Using the approach presented here, the driver of the vehicle can be informed in a timely manner of the unexpectedly close approach of another vehicle. If the driver does nothing in response to the proximity warning and continues driving as before, a collision will not occur. If the driver or the other vehicle reacts incorrectly and a collision risk arises, a collision warning can still be issued downstream, allowing the driver to correct their incorrect reaction or react correctly to the other vehicle's incorrect reaction.

[0014] A method is proposed for providing an approach warning of a latent collision risk between a vehicle and a following vehicle, wherein a warning zone is spanned around the following vehicle and around the vehicle, wherein the warning zone for at least one of the vehicles is spanned larger, at least in one spatial direction, than a collision warning zone of the respective vehicle, wherein the approach warning is provided when the warning zones overlap. Ideas for embodiments of the present invention can be considered, among other things, to be based on the ideas and findings described below.

[0015] In particular, a vehicle can be a single-track vehicle, such as a motorcycle. In addition to the single-track vehicle, there is usually space within a lane for another single-track vehicle. However, the approach presented here can also be used for a multi-track vehicle. If another vehicle unexpectedly appears next to the vehicle, the driver of the vehicle may be startled.

[0016] A latent collision risk may be less acute than an actual collision risk. A latent collision risk can exist when two vehicles approach each other closer than a definable minimum distance but will not touch. An actual collision risk can exist when the vehicles will touch, or will collide if neither vehicle swerves. The latent collision risk can become an actual collision risk if at least one of the drivers of the vehicles reacts incorrectly and, for example, drives toward the other vehicle.

[0017] A warning area can be a virtual area around a vehicle. The warning area can be projected into a plan view of the vehicle's surroundings. The warning area can extend around a position of the vehicle. The warning area can be larger than a collision warning area of ​​the vehicle. The collision warning area can be located entirely within the warning area. The collision warning area can essentially correspond to an outline of the vehicle. If the collision warning areas of the vehicles overlap, a collision warning can be provided. The collision warning area can also be slightly larger than the outline. At least in one direction, i.e., transverse to the vehicle and / or longitudinal to the vehicle, the warning area can be significantly larger than the collision warning area.

[0018] The warning area can be enlarged for one of the two vehicles or for both vehicles in at least one direction. If the warning area is enlarged for only one of the vehicles, this warning area can be larger than if the warning area is enlarged for both vehicles. Using kinematic information from the vehicles, future trajectories of the vehicles can be predicted. The warning areas can be spanned along the trajectories. To detect the latent collision risk, a future movement of the vehicles in the near future can be predicted. The future movement can be predicted from a current movement and / or a past movement from the recent past. For example, it can be assumed that the past movement will continue into the future with a certain probability.The further the movement is predicted, the greater the uncertainty of the forecast can become. Warning areas extending along the trajectories can be referred to as warning tubes.

[0019] Along the trajectories, future positions of the vehicles can be determined at predetermined time intervals. Warning zones can be defined around these positions. The positions of both vehicles can be determined at the same time interval. The time intervals between the time intervals can be equal. For example, the positions can be determined at a time interval between 0.025 seconds and 0.4 seconds. In particular, the positions can be determined at time intervals of 0.1 seconds.

[0020] The trajectories can be predicted for a predetermined time period into the future. The length of the predicted trajectories can be limited. For example, the trajectories can be predicted for time periods between one and five seconds. In particular, the trajectories can be predicted for three seconds. For shorter time periods, the inaccuracy of the prediction can be limited. For shorter time periods, the advance warning time for the approach warning can be short.

[0021] The proximity warning can be provided if the warning areas overlap by more than a predefined tolerance. The proximity warning can be provided if the warning areas overlap by more than a predefined overlap. This allows the possible approach of the following vehicle to be detected before the proximity warning is provided. If the warning areas overlap only briefly and / or by less than the tolerance, the proximity warning may not be issued. The tolerance can, for example, be between two and ten percent overlap. In particular, the tolerance can be five percent.

[0022] The proximity warning can be cancelled when the all-clear zones defined by an all-clear distance around the warning zones no longer overlap. An all-clear zone can be larger than the corresponding warning zone. A warning hysteresis can be achieved by using a smaller warning zone and a larger all-clear zone. Once the approach warning is issued, it remains active until the following vehicle comes significantly closer to the vehicle. This prevents the approach warning from flickering.

[0023] The warning zones can be spanned as rectangular frames around the vehicles. The warning zones can have a highly simplified shape. Since there is no acute, immediate danger of collision when the warning zones overlap without the collision warning zones overlapping, the warning zones can be simple rectangles. This simple shape can reduce the computational effort.

[0024] The warning area can be enlarged by a factor larger than the collision warning area. A magnification factor can expand the warning area proportionally to the collision warning area. For example, the magnification factor can be between 10 percent and 200 percent. The magnification factor can be influenced by the driver. If the driver wants to be warned early or frequently, the magnification factor can be set large. If the driver wants to be warned late or rarely, the magnification factor can be set small.

[0025] The method is preferably computer-implemented and can be implemented, for example, in software or hardware or in a mixed form of software and hardware, for example in a driver assistance system.

[0026] The approach presented here further creates a control unit, wherein the control unit is designed to carry out, control, or implement the steps of a variant of the method presented here in corresponding devices. The control unit can be an electrical device with at least one computing unit for processing signals or data, at least one memory unit for storing signals or data, and at least one interface and / or a communication interface for reading in or outputting data embedded in a communication protocol. The computing unit can be, for example, a signal processor, a so-called system ASIC, or a microcontroller for processing sensor signals and outputting data signals depending on the sensor signals. The memory unit can be, for example, a flash memory, an EPROM, or a magnetic storage unit.The interface can be configured as a sensor interface for reading sensor signals from a sensor and / or as an actuator interface for outputting data signals and / or control signals to an actuator. The communication interface can be configured to read or output data wirelessly and / or via a wired connection. The interfaces can also be software modules, which are present, for example, on a microcontroller alongside other software modules.

[0027] Also advantageous is a computer program product or computer program with program code that can be stored on a machine-readable carrier or storage medium such as a semiconductor memory, a hard disk memory or an optical memory and is used to carry out, implement and / or control the steps of the method according to one of the embodiments described above, in particular when the program product or program is executed on a computer or a device.

[0028] It should be noted that some of the possible features and advantages of the invention are described herein with reference to different embodiments. A person skilled in the art will recognize that the features of the control device and the method can be combined, adapted, or exchanged as appropriate to achieve further embodiments of the invention.

[0029] Short description of the drawings

[0030] Embodiments of the invention will be described below with reference to the accompanying drawings, wherein neither the drawings nor the description are to be interpreted as limiting the invention. Fig. 1 shows an illustration of providing a proximity warning according to one embodiment; and

[0031] Fig. 2 shows a flowchart of a method according to an embodiment.

[0032] The figures are merely schematic and not to scale. Like reference numerals denote like or equivalent features.

[0033] Embodiments of the invention

[0034] Fig. 1 shows an illustration of the provision of a proximity warning 100 according to one exemplary embodiment. A vehicle 102 is traveling in front of a following vehicle 104. The vehicle 102 is thus a leading vehicle 102. The vehicle 102 is traveling in one lane, and the following vehicle 104 is traveling in an adjacent lane. Here, the vehicle 102 is a motorcycle. The following vehicle 104 is a car.

[0035] The vehicle 102 has a sensor system 106 for monitoring a rear area 108 behind the vehicle 102. The sensor system 106 here comprises, for example, a rearward-facing radar sensor. The sensor system 106 detects objects in the rear area 108. The following vehicle 104 is also detected as an object. Kinematic information about the following vehicle 104 and the vehicle 102 itself is evaluated to provide the proximity warning 100.

[0036] To generate the distance warning 100, a warning area 110 is defined around a considered position of the vehicle 100. The warning area 110 is wider to the right and left of the vehicle 100 than a collision warning area 112 of the vehicle 100. A warning area 110 is also defined around a temporally corresponding position of the following vehicle 104. This warning area 110 is slightly larger in all spatial directions than the collision warning area 112 of the following vehicle 104. If the warning areas 110 overlap, the approach warning 100 is provided to a driver of the vehicle 102.

[0037] In one embodiment, to provide the approach warning 100, trajectories 114 of the vehicles 102, 104 are predicted for a predefined period of time into the future. The warning areas 110 are spanned along the predicted trajectories 114, and a check is made to determine whether the warning areas 110 will overlap within the time period. Current information and previously acquired information can be used for the prediction.

[0038] In one embodiment, a sequence of warning areas 110 is spanned around predefined positions 116 along the trajectories 114. The positions 116 are located in predefined journals along the trajectories 114. The proximity warning 100 is provided when the warning areas 110 of the same time step overlap.

[0039] In one embodiment, the trajectories 114 are predicted three seconds into the future and a position 116 for a warning area 110 is defined every tenth of a second.

[0040] In one embodiment, the proximity warning 100 is provided when the warning areas 110 overlap by more than a warning tolerance. This allows the proximity warning 100 to be provided with increased robustness, since the warning areas 110 have already overlapped for a certain period of time due to the warning tolerance before the proximity warning 100 is provided.

[0041] In one embodiment, the warning tolerance results in a warning hysteresis, since the proximity warning 100 is only canceled when the warning areas 110 no longer overlap. In this case, the warning area 110 can also be referred to as an all-clear area. The warning tolerance can be an absolute area tolerance or a percentage area tolerance. Likewise, the warning tolerance can be a temporal tolerance. In this case, the proximity warning 100 is provided when the warning areas 110 overlap for longer than a tolerance period.

[0042] In one embodiment, the warning areas 110 are rectangular boxes around the positions of the vehicles 102, 104. These boxes can be spanned with little computational effort and thus enable rapid provision of the approach warning 100. Fig. 2 shows a flow diagram of a method according to one embodiment. In a first step 200, information about objects detected by the radar is read in. In a second step 202, objects with a high collision relevance are identified from this. In a third step 204, a prediction is made as to whether these objects are critical collision objects, i.e., whether a collision time with the host vehicle is calculable and whether they meet criticality criteria. In a fourth step 206, a selection list is created from these objects, and in a fifth step 208, objects not relevant for a rear-end collision warning are filtered out.In a sixth step 210, the proximity warning is provided for the filtered objects.

[0043] To calculate collision times and criticality criteria, several sub-steps are executed in the third step 204. In a first sub-step 212, kinematic information of the objects with high collision relevance and of the host vehicle is read in. In a second sub-step 214, warning areas are drawn around the objects and the host vehicle. The warning areas have predefined dimensions. In a third sub-step 216, the warning area around the host vehicle and / or the warning area around at least one of the objects is enlarged using a scaling factor. In a fourth sub-step 218, movements of the warning areas are predicted in tenths of a second for three seconds into the future.

[0044] In the following, possible embodiments of the invention are summarized again or presented with slightly different wording.

[0045] An enhanced rear collision warning using artificial collision prediction by inflating an ego-vehicle bounding box is presented.

[0046] Currently, it is not possible to increase the probability of a rear-end collision warning. However, there are use cases where the feature user would like warnings for vehicles that are not currently predicted to be involved in a collision, but that still pose a high or latent collision risk. For example, scooters / bicycles that squeeze past the ego vehicle at high speed.

[0047] Therefore, an extension of the rear collision warning system is presented here that enables such warnings by enlarging / inflating the bounding box of the ego vehicle and / or the bounding box of the object, thus increasing the probability of a warning.

[0048] This warning may be desirable if the ego vehicle is in a lane and another object is passing quickly with close lateral clearance. In this case, there may be no time-to-collision (TTC), but there is a latent hazard / collision risk, e.g., if the driver or target vehicle changes its behavior or makes a driving error.

[0049] For example, the other vehicle is a two-wheeler and is passing the ego vehicle during a lane change, or the ego vehicle is between lanes or sharing lanes, or the ego vehicle is a two-wheeler and is traveling in a group formation.

[0050] With the approach presented here, a collision warning is issued via a human-machine interface even in situations with a low collision risk, for example, despite a significant overtaking distance.

[0051] During a method sequence according to one embodiment, radar objects are received in a radar perception layer. Objects with high collision relevance are identified from these in a radar situation analysis layer. To predict whether the objects are critical collision objects (those that have an actual time until collision with the ego vehicle and fulfill the conditions for a critical collision), a bounding box is drawn around the host vehicle (ego) and the respective object using hard-coded dimensions. An inflation factor is then applied to at least one of the bounding boxes. Using kinematic information of the host vehicle (ego) and the respective object, a prediction of the bounding box movements into the future is then made in 0.1-second increments for, for example, three seconds.A radar functional layer then receives a selection list of collision objects and filters out objects that are not relevant for the rear-end collision warning use case (e.g., not within the warning area). Vehicle warnings are then provided via a radar-vehicle interface (if possible within the vehicle). In the approach presented here, the bounding box of the ego vehicle and / or the other vehicle is artificially inflated, allowing greater flexibility for warning applications.

[0052] Finally, it should be noted that terms such as "comprising," "having," etc., do not exclude other elements or steps, and terms such as "a" or "an" do not exclude a plurality. Reference signs in the claims are not to be considered limiting.

Claims

Claims 1 . Method for providing an approach warning (100) of a latent collision risk between a vehicle (102) and a following vehicle (104), wherein a warning area (110) is spanned around the following vehicle (104) and around the vehicle (102), wherein the warning area (110) for at least one of the vehicles (102, 104) is spanned at least in one spatial direction larger than a collision warning area (112) of the respective vehicle (102, 104), wherein the approach warning (100) is provided when the warning areas (110) overlap.

2. Method according to claim 1, wherein future trajectories (114) of the vehicles (102, 104) are predicted using kinematic information of the vehicles (102, 104), wherein the warning areas (110) are spanned along the trajectories (114).

3. Method according to claim 2, wherein future positions (116) of the vehicles (102, 104) are determined along the trajectories (114) at predetermined time intervals, wherein the warning areas (110) are each spanned around the positions (116).

4. Method according to one of claims 2 to 3, wherein the trajectories (114) are predicted for a predetermined period of time into the future.

5. Method according to one of the preceding claims, wherein the proximity warning (100) is provided when the warning areas (110) overlap by more than a predefined tolerance.

6. Method according to claim 5, wherein the approach warning (100) is cancelled when all-clear areas spanned by an all-clear distance around the warning areas (110) no longer overlap.

7. Method according to one of the preceding claims, in which the warning areas (110) are spanned as rectangular frames around the vehicles (102, 104).

8. Method according to one of the preceding claims, in which the warning area (110) is spanned by a magnification factor larger than the collision warning area (112).

9. The method according to any one of the preceding claims, further comprising providing a collision warning of an acute collision risk when the collision warning areas (112) overlap.

10. Control device, wherein the control device is designed to carry out, implement and / or control the method according to one of the preceding claims in corresponding devices.

11. A computer program product configured to instruct a processor, upon execution of the computer program product, to execute, implement, and / or control the method according to any one of claims 1 to 9.

12. A machine-readable storage medium on which the computer program product according to claim 11 is stored.

Images