GB2617865A - Method of marking mirror objects in a cross-traffic scenario - Google Patents
Method of marking mirror objects in a cross-traffic scenario Download PDFInfo
- Publication number
- GB2617865A GB2617865A GB2206209.5A GB202206209A GB2617865A GB 2617865 A GB2617865 A GB 2617865A GB 202206209 A GB202206209 A GB 202206209A GB 2617865 A GB2617865 A GB 2617865A
- Authority
- GB
- United Kingdom
- Prior art keywords
- mirror
- ego vehicle
- marking
- detected object
- area
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 32
- 230000003213 activating effect Effects 0.000 claims abstract description 6
- 230000009471 action Effects 0.000 claims abstract description 5
- 230000006870 function Effects 0.000 claims description 36
- 230000004913 activation Effects 0.000 claims description 17
- 238000004590 computer program Methods 0.000 claims description 6
- 238000001994 activation Methods 0.000 description 13
- 102100035475 Blood vessel epicardial substance Human genes 0.000 description 6
- 101001094636 Homo sapiens Blood vessel epicardial substance Proteins 0.000 description 6
- 101000608194 Homo sapiens Pyrin domain-containing protein 1 Proteins 0.000 description 6
- 101000595404 Homo sapiens Ribonucleases P/MRP protein subunit POP1 Proteins 0.000 description 6
- 230000004927 fusion Effects 0.000 description 5
- 102100034112 Alkyldihydroxyacetonephosphate synthase, peroxisomal Human genes 0.000 description 3
- 101000799143 Homo sapiens Alkyldihydroxyacetonephosphate synthase, peroxisomal Proteins 0.000 description 3
- 101000942586 Homo sapiens CCR4-NOT transcription complex subunit 8 Proteins 0.000 description 3
- 101001094629 Homo sapiens Popeye domain-containing protein 2 Proteins 0.000 description 3
- 101000608230 Homo sapiens Pyrin domain-containing protein 2 Proteins 0.000 description 3
- 102100035482 Popeye domain-containing protein 2 Human genes 0.000 description 3
- 238000000848 angular dependent Auger electron spectroscopy Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 230000007613 environmental effect Effects 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 230000000903 blocking effect Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 229910017435 S2 In Inorganic materials 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
- G01S13/931—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of land vehicles
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/02—Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
- G01S13/06—Systems determining position data of a target
- G01S13/46—Indirect determination of position data
- G01S2013/462—Indirect determination of position data using multipath signals
- G01S2013/464—Indirect determination of position data using multipath signals using only the non-line-of-sight signal(s), e.g. to enable survey of scene 'behind' the target only the indirect signal is evaluated
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
- G01S13/931—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of land vehicles
- G01S2013/9317—Driving backwards
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
- G01S13/931—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of land vehicles
- G01S2013/9327—Sensor installation details
- G01S2013/93272—Sensor installation details in the back of the vehicles
Abstract
The invention refers to a computer-implemented method of marking mirror objects in a cross-traffic scenario, the cross-traffic scenario comprising an ego vehicle, a stationary object, and a real object crossing rearward direction of the ego vehicle. The method defining a mirror area, being an area obscured by the stationary object. Acquiring from the radar sensor the position of a detected object, crossing rearward direction of the ego vehicle, checking if one or more relevant corners P of the detected object are situated inside the mirror area, marking the detected object as a mirror object if the one or more relevant corners P are situated inside the mirror area, marking the detected object, or marking the detected object as the real object, if the one or more relevant corners P are situated outside the mirror area. Communicating the marked mirror object or, respectively, the marked real object to the warning function for taking action: not activating the warning function in case of the marked mirror object, or activating the warning function in case of the marked real object.
Description
Description
Method of marking mirror objects in a cross-traffic scenario The present invention relates to advanced driver assistance systems ADAS. In particular the invention relates to a method of mirror object filtering for cross-traffic scenario, when the cross-traffic scenario comprises an ego vehicle and a stationary object placed in the rear part, and laterally in respect to the ego vehicle.
A considerable reduction of vehicle crashes occurred since the introduction and progress of the advanced driver assistance systems ADAS because these systems have a major contribution in the avoidance of a significant part of the collisions by predicting them, and by taking actions.
Part of the crashes occur in cross-traffic scenarios, that is when an ego vehicle is moving rearward direction, a target vehicle is crossing rearward direction of the ego vehicle, and a stationary object, usually another vehicle, is placed in close vicinity of the ego vehicle in the rear part and laterally in respect to the ego vehicle, as shown in Fig.1. This typically happens when the ego vehicle wants to reverse out of a parking lot.
The advanced driver assistance systems ADAS of the ego vehicle comprise two or more radar sensors, such as for example Short-Range Radars SRRs, situated on the rear corners of the ego vehicle and a warning function, such as a Rear Cross 25 Traffic Alert/Brake RCTA/RCTB function.
When the ego vehicle is moving rearward direction, the radar sensors detect the presence of the target vehicle by using the radio waves. The warning function is activated based on an environmental prediction model to warn and/or to brake the ego vehicle if there is a risk of collision of the ego vehicle with the target vehicle.
Hereafter the real target vehicle shall be named "real object", whereas the mirrored target vehicle shall be named "mirror object". The person skilled in the art knows that the mirror object perfectly mirror the real object in respect to the relative position with respect to a longitudinal or a transverse axis of the ego vehicle, as represented in Fig. 1 and Fig.4, or can imperfectly mirror the real object, as represented in Fig. 2 and Fig. 3 in respect to the relative position with respect to the longitudinal, or respectively transverse axis of the ego vehicle, giving the impression that is positioned at a different distance from said axis than the real object.
The term "radar sensors" shall refer to the ego vehicle's radar sensors placed at the two rear corners of the ego vehicle. The wavelength of the radar sensors as well as 10 the number of radar sensors are the one usually used in the automotive industry and shall not limit the invention.
As shown in Fig. 1, in cases when the stationary object is placed in close vicinity to one of two radar sensors of the ego vehicle, the emitted radio waves of the respective radar sensor reflect and, for this reason, they may faulty detect the real object, detecting a mirror object crossing in the opposite direction to the real object.
The fault in detecting the real object will cause a wrong warning and/or braking of the ego vehicle by the warning function, which is the first disadvantage of prior art.
Some methods of prior art include using additional hardware components in order to facilitate a fusion of the information sent by the radar sensors in the environmental prediction model having the role to double-check the existence of the mirror object. However, the fusion has two disadvantages: it has limitations, and it can be computationally-intensive. The limitations occur when there are two or more stationary objects blocking two or more radar sensors, because the fusion cannot ascertain which of the objects is the real object and which is the mirror object. In such cases, when the first object is detected by one of the two radar sensors, this is considered to be the real object, prediction that has a 50% probability of being wrong.
The technical problem to be solved is to mark mirror objects in the cross-traffic scenario when one or more rear radar sensors are obstructed by stationary objects such that to reduce the number of false positive activations of the warning function, eliminating the need of fusion of the two radar sensors.
In order to overcome the disadvantages of prior art, in a first aspect of the invention it is presented a computer-implemented method of marking mirror objects in a cross-traffic scenario, the cross-traffic scenario comprising an ego vehicle, a stationary object, and a real object crossing rearward direction of the ego vehicle, the ego vehicle equipped with radar sensors, placed on the respective rear corners of the ego vehicle, and with a warning function for warning and/or braking the ego vehicle if there is a risk of collision of the ego vehicle with the real object.
The method comprises the following steps carried out at each sensor cycle when the ego vehicle is reversing out, and the stationary object is placed at a distance under a vicinity threshold in respect to one of the radar sensors: Si defining a mirror area as a triangle defined by a rear ego corner closest to the stationary object, a lateral direction point, in respect to the rear ego corner and a backward direction point in respect to the rear ego corner, S2 acquiring from the radar sensor the position of a detected object, the detected object crossing rearward direction of the ego vehicle, S3 checking if one or more relevant corners of the detected object are situated inside the mirror area, S4 marking the detected object as a mirror object if the one or more relevant corners P are situated inside the mirror area, marking the detected object, S5 communicating the marked mirror object or, respectively, the marked real object to the warning function for taking action: not activating the warning function in case of the marked mirror object, or activating the warning function in case of the marked real object.
In a second aspect of the invention, it is presented a hardware processing unit comprising one or more processors, at least one non-volatile memory, and a non-transitory computer-readable storage medium, the hardware processing unit being configured to perform operations of the computer-implemented method of marking mirror objects in a cross-traffic scenario.
In a third aspect of the invention, it is presented a non-transitory computer-readable storage medium encoded with a computer program, the computer program comprising instructions executable by one or more processors of the collision point processing unit which, upon such execution by the hardware processing unit, causes the one or more processors to perform operations of the computer-implemented method of marking mirror objects in a cross-traffic scenario of any preferred embodiment.
Further advantageous embodiments are the subject matter of the dependent claims.
The main advantages of using the invention are as follows: - the method of the invention is reducing the number of false positive activations of the warning function, -the method of the invention has computational efficiency by eliminating the need of fusion of the two radar sensors, - the method of the invention is robust allowing to easily fit into the majority of the embedded systems without requiring high memory or high processing power, which makes it highly affordable, -the method of the invention is applicable to manned vehicles as well as for autonomous vehicles.
Figures Further special features and advantages of the present invention can be taken from the following description of an advantageous embodiment by way of the accompanying drawings: Fig. 1 illustrates the cross-traffic scenario and the disadvantages of prior art, Fig. 2 illustrates the definition of the mirror area according to the method of the invention, Fig. 3 illustrates a first preferred embodiment, Fig. 4 illustrates a second preferred embodiment, Fig. 5 illustrates the second preferred embodiment. Detailed description The computer-implemented method of marking mirror objects in a cross-traffic scenario of the first aspect of the invention has 5 steps.
With reference to Fig. 2, the cross-traffic scenario comprises an ego vehicle, a stationary object placed in the rear part, and laterally in respect to the ego vehicle, and a real object crossing rearward direction of the ego vehicle.
The ego vehicle is equipped with: radar sensors, placed on the respective rear corner of the ego vehicle, a warning function for warning and/or braking the ego vehicle if there is a risk of collision of the ego vehicle with the real object.
The computer-implemented method of the invention is carried out only when the ego vehicle is reversing out and the stationary object is placed at a distance under a vicinity threshold in respect to one of the radar sensors. The computer-implemented method of the invention can be carried out simultaneously for two stationary objects, placed in both rear lateral parts of the ego vehicle.
The computer-implemented method is carried out at each radar sensor cycle, where the setting of the radar sensor cycle is included in the configuration of the environmental prediction model, and it is not limiting the invention.
Step Si With reference to Fig. 2, in step 1 a mirror area is defined behind the stationary object coinciding with the rear lateral part in respect to the ego vehicle. For the ease of understanding of the invention, the mirror area is represented only on one side of the ego vehicle. It shall be understood that another mirror area can be defined on the opposite side, or two mirror areas can be defined, one on each respective rear lateral part in respect to the ego vehicle.
The mirror area is defined as a triangle, the sides of which are dynamically computed at each radar sensor cycle.
According to Fig. 2, the area of the mirror angle triangle is defined by the three points PO, P1, P2 where PO is the rear corner of the ego vehicle closest to the stationary object. The following are defined, with reference to Fig. 2: PO = rear ego corner, that is the same as the radar sensor position P1 = lateral direction point, in respect to the rear ego corner P2 = backward direction point, in respect to the rear ego corner R = closest rear corner of the stationary object F = closest front corner of the stationary object Ci (0<=i<=3) = mirrored object's four corners The lateral direction point, P1 and, respectively, the backward direction point P2 are 15 defined by the range of radar sensor, which is the radar sensor's maximum range at which an object is correctly detected and classified.
The line equations for the sides of the triangle starting from the rear ego corner PO have the following general formula y=mx+n, where m is slope and n is y-intercept The line equation defined by the rear ego corner PO and the closest front corner of the stationary object F is used to compute the lateral direction point P1 as following: = mi*x + ni [1] The line equation defined by the rear ego corner PO and the closest rear corner of the stationary object F is used to compute the backward direction point P2 as following: = mzlex + nz [2] where mi and mz are the respective slopes of the sides of the triangle POP1 and, respectively POP2 in respect to the Cartesian axis x and ni and nz are the y-intercepts.
Using equations 1 and 2, it is possible to compute the lateral direction point P1 and, respectively, the backward direction point, P2: P1 = point on y1 such that POP1 = range of the radar sensor [3] P2 = point on y2 such that POP2 = range of the radar sensor [4] Assuming that the range of the radar sensor is the same in all directions, from the Wel and the 4th equation, it can be deduced that POP1 = POP2, thus the triangle POP1 P2 is isosceles. If the range of the radar sensor is not the same in all directions, the triangle POP1 P2 is not isosceles. The teaching of the invention is the same irrespective of whether the triangle POP1 P2 is isosceles or not.
Step S2 In step 2, with reference to Fig. 2, the position of a detected object crossing rearward direction of the ego vehicle is acquired from the radar sensor. For the ease of understanding of the invention, in the description it shall be exemplified the acquisition from one radar sensor. This acquisition is carried out at each sensor cycle.
Step S3 In step 3, with reference to Fig. 2, it is checked whether the detected object is inside or outside the mirror area.
Due to the representation of the mirror area as a triangle, in order to check if the detected object is situated inside the mirror area, one computationally efficient solution is using barycentric coordinates.
In geometry, a barycentric coordinate system is a coordinate system in which the location of a point is specified by reference to a simplex, in this case a triangle.
Barycentric coordinates can be used to express the position of any point located on a triangle with three scalars, which are noted with alpha, beta, gamma. The sum of alpha, beta, gamma is 1.
In order for the detected object be classified as mirror object, one or more relevant corners P must be situated inside the mirror area. Usually, one or more relevant corners P of the detected object are the closest to the ego vehicle corner whose radar sensor acquired the position of the detected object.
For computational efficiency, in a non-limiting example illustrated in Fig. 2, it is sufficient to check if one relevant corner P of the detected object is situated inside the mirror area. The position of the relevant corner P is computed using barycentric coordinates according to the following equation: Position of the relevant corner P = alpha*P1 + beta*P2 + gamma*P3 [5] Thus, the relevant corner P is inside the triangle POP1P2, if the following condition is fulfilled: 0 <= alpha + beta + gamma < =1 [6] If the one or more relevant corners P are inside the mirror area, then the detected object is inside the mirror area.
If the one or more relevant corners P are outside the mirror area, then the detected object is outside the mirror area.
Step S4 In step 4, if the one or more relevant corners P are situated inside the mirror area, the detected object is marked as a mirror object, as schematically illustrated in Fig. 2. Otherwise, the detected object is marked as the real object.
Step 35 In step 5, the marked mirror object or, respectively, the marked real object is communicated to the warning function for taking action.
In case of mirror objects, the warning function is not activated. In case of real objects, the warning function is activated. The mechanisms of activation of the warning function are adapted to the situations when the ego vehicle is manned or is autonomous, said mechanism of activation being outside the scope of the invention.
The accuracy of the computer-implemented method of the invention can be further improved in order to address some situations arising from real life.
The improvements are described hereafter with reference to a first and a second preferred embodiment.
The first preferred embodiment of the invention, with reference to Fig. 3, addresses the situations when the detected object is placed in a critical area of the rear part of the ego vehicle.
As illustrated schematically in Fig. 3, in the previous sensor cycle the relevant corner P was outside the mirror area, thus the detected object was marked as mirror object and hence the warning function was not activated. In some rare cases, the marked mirror object might actually be the real object, thus the collision risk is also real.
In order to improve the accuracy of the marking of the mirror objects, and reduce the possibility of wrong marking of a real object as mirror object, in the first preferred embodiment, steps 1 and 4 of the method further comprise specific sub-steps.
Step Si further comprises the following sub-step of defining the critical area placed at the rear of the ego vehicle standing in the way of reversing out, perpendicular to 25 the direction of reversing out, having a predetermined width w, for example w= 2m.
Step S4 further comprises the following sub-steps carried out after the sub-step of marking the detected object: In the first additional sub-step of step S4 it is checked if the detected object is placed within the critical area.
In case the detected object is placed inside the critical area, it is triggered a supplementary marking of the detected object based on a pre-determined activation threshold.
In case the detected object is placed outside the critical area, the general situation described above applies.
The triggering of the supplementary marking of the detected object based on a pre-determined activation threshold involves specific sub-steps: - adding a binary cycle-based history vector having a length of k sensor cycles and creating a binary cycle-based history, adding to the binary cycle-based history at each of the k sensor cycles: - if the one or more relevant corners P are situated inside the mirror area, then add 1 to the binary cycle-based history, - otherwise add 0 to the binary cycle-based history, comparing at the end of the k sensor cycles the binary cycle-based history with the activation threshold.
marking the detected object as the real object if the binary cycle-based history is below the activation threshold, Or marking the detected object as the mirror object if the binary cycle-based history is above or equal the activation threshold.
For example, the length k is of 4 sensor cycles and the activation threshold is 5.
-If the detected object is inside the mirror area, then 1 was added to the binary cycle-based history, - If the detected object is outside the mirror area, then 0 was added to the binary cycle-based history. In the binary cycle-based history of 4 sensor cycles, in 3 previous cycles the detected object was inside the mirror area, and in 1 current cycle the detected object is outside the mirror area.
The binary cycle-based history is: 0111 and the sum is 7: (2^0 + 2^1 + 2^2 = 7) Since 7 is larger than the activation threshold = 5, the detected object is marked as the mirror object at the end of step 4 and the warning function is not activated in step 5 S5 of the method of the example of this preferred embodiment.
In the second preferred embodiment, the principle of improving the accuracy of the activation of the warning function lies in comparing the mirror area with a field of view FOV of the radar sensor. The field of view FOV the radar sensor is known from
the technical specifications.
Fig. 4 shows the cross-traffic scenario of the second preferred embodiment.
In step S3 of the method from the second preferred embodiment, the following sub-step is comprised: an overlap area is defined by overlapping the mirror area with the field of view FOV of the radar sensor, an overlap ratio is defined between the overlap area and the field of view FOV, the overlap ratio is compared with an overlap threshold.
In step 55 of the method from the second preferred embodiment, the following sub-step is comprised: if the overlap ratio is larger than or equal to the overlap threshold, a request is sent to the warning function to activate an additional warning function.
As it can be seen from Fig. 5, the difference between the field of view FOV and the overlap area is an available field of view FOV.
The overlap ratio of Fig. 4 is more than 95%. Implication of the high overlap ratio is that, if the radar sensor's field of view FOV is almost totally obstructed, the radar sensor may provide reliable information and additional warning is needed.
As the ego vehicle reverses out, the overlap area has a lesser overlap ratio, as seen in Fig. 5, thus the available field of view FOV is larger as compared with Fig.4.
The same Fig. 5 shows for comparison the opposite side of the ego vehicle in respect to the stationary object where there is no stationary object blocking the field of view FOV of the radar sensor on the opposite side-in the image the left side, is 100% not obstructed, whereas the field of view FOV of the radar sensor on the stationary object's side-in the image the right side, is obstructed at around 60%.
In our non-limiting example, the overlap threshold is 70%.
In case the overlap ratio is larger than or equal to the overlap threshold-as it is for example in Fig. 4, step S5 of the second preferred embodiment further comprises the sub-step of sending a request to the warning function to activate the additional warning function. In case the overlap ratio is smaller than the overlap threshold-as it is for example in Fig. 5, step S5 of the second preferred embodiment does not further comprises the sub-step of sending the request to the warning function to activate the additional warning function.
If the ego vehicle is manned, the additional warning function can be a visual alert or an audio message to the driver. Additional warning function alerts the driver that the higher the overlap ratio, the smaller the available field of view FOV and the more cautious he must be. If using audio messages, the higher the overlap ratio is, the higher the tone of the message must be, whereas if using visual messages, the higher the overlap ratio is, the brighter the visual alert must be.
For autonomous vehicles, the additional warning function can be for example slowing down the vehicle. The higher the overlap ratio, the quicker the braking function must be activated.
It is possible to combine the first preferred embodiment with the second preferred embodiment as they address different types of improvement: the first preferred embodiment improves the accuracy of the marking of the mirror objects, whereas the second preferred embodiment improves the accuracy of the activation of the warning function. Each of the preferred embodiments taken alone as well as the combination of the two preferred embodiments have the advantage of flexibility to adapt to various situations.
In a second aspect of the invention, it is presented a hardware processing unit comprising one or more processors, at least one non-volatile memory and a non-transitory computer-readable storage medium, the hardware processing unit being configured to perform operations of the computer-implemented method of marking mirror objects in a cross-traffic scenario.
Non limiting examples of hardware processing unit are controllers or electronic control units. The hardware processing unit can carry out other functions that are outside the scope of the invention.
In a third aspect of the invention, it is presented a non-transitory computer-readable storage medium encoded with a computer program, the computer program comprising instructions executable by one or more processors of the collision point processing unit which, upon such execution by the hardware processing unit, causes the one or more processors to perform operations of the computer-implemented method of marking mirror objects in a cross-traffic scenario of any preferred embodiment.
While certain embodiments of the present invention have been described in detail, those familiar with the art to which this invention relates will recognize various alternative designs and embodiments for practicing the invention as defined by the following claims.
List of reference signs -Ego vehicle 111 -Real object -Mirror object LL -Stationary object -Mirror area PO -rear ego corner, that is the same as the radar sensor position P1 -lateral direction point in respect to the rear ego corner P2 -backward direction point in respect to the rear ego corner R -closest rear corner of the stationary object F -closest front corner of the stationary object Ci (0 i.s 3) mirrored object's four corners P -one or more relevant corners Critical area w -width of the critical area
FOV -field of view
-Available FOV -Invalid/Blocked FOV A-FOV Area
Claims (2)
- Claims 1. Computer-implemented method of marking mirror objects in a cross-traffic scenario, the cross-traffic scenario comprising an ego vehicle, a stationary object, and a real object crossing rearward direction of the ego vehicle, the ego vehicle equipped with radar sensors, placed on the respective rear corners of the ego vehicle, and with a warning function for warning and/or braking the ego vehicle if there is a risk of collision of the ego vehicle with the real object, characterized in that the method comprises the following steps carried out at each sensor cycle when the ego vehicle is reversing out and the stationary object is placed at a distance under a vicinity threshold in respect to one of the radar sensors: Si defining a mirror area as a triangle POP1P2 defined by a rear ego corner PO closest to the stationary object, a lateral direction point, P1 in respect to the rear ego corner PO and a backward direction point P2 in respect to the rear ego corner PO, S2 acquiring from the radar sensor the position of a detected object, the detected object crossing rearward direction of the ego vehicle, S3 checking if one or more relevant corners P of the detected object are situated inside the mirror area, S4 marking the detected object as a mirror object if the one or more relevant corners P are situated inside the mirror area, marking the detected object, S5 communicating the marked mirror object or, respectively, the marked real object to the warning function for taking action: not activating the warning function in case of the marked mirror object or activating the warning function in case of the marked real object.
- 2. The method of claim 1, characterized in that Si further comprises the following sub-step: defining a critical area placed at the rear of the ego vehicle standing in the way of reversing out, perpendicular to the direction of reversing out, having a predetermined width w, S4 further comprises the following sub-steps: checking if the detected object is placed within the critical area, triggering a supplementary marking of the detected object based on a pre-determined activation threshold if the detected object is placed inside the critical area: adding a binary cycle-based history vector having a length of k sensor cycles and creating a binary cycle-based history, adding to the binary cycle-based history at each of the k sensor cycles: if the one or more relevant corners P are situated inside the mirror area, then add 1 to the binary cycle-based history, otherwise add 0 to the binary cycle-based history, comparing at the end of the k sensor cycles the binary cycle-based history with the activation threshold, marking the detected object as the real object if the binary cycle-based history is below the activation threshold or marking the detected object as the mirror object if the binary cycle-based history is above or equal the activation threshold.a The method of claim 1, or 2, characterized in that S3 further comprises the following sub-step: defining an overlap area by overlapping the mirror area with a field of view FOV of the radar sensor, defining an overlap ratio between the overlap area and the field of view FOV, and comparing the overlap ratio with an overlap threshold, S5 further comprises the following sub-step: if the overlap ratio is larger than or equal to the overlap threshold, sending a request to the warning function to activate an additional warning function.4. A hardware processing unit characterized in that it comprises one or more processors, at least one non-volatile memory and a non-transitory computer-readable storage medium, the hardware processing unit being configured to perform operations of the computer-implemented method of marking mirror objects in a cross-traffic scenario of any of the claims 1 to 3.5. A non-transitory computer-readable storage medium encoded with a computer program, characterized in that it the computer program comprises instructions executable by one or more processors of the hardware processing unit which, upon such execution by the hardware processing unit, causes the one or more processors to perform operations of the computer-implemented method of marking mirror objects in a cross-traffic scenario of any of the claims 1 to 3.
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP22465530 | 2022-04-21 |
Publications (2)
Publication Number | Publication Date |
---|---|
GB202206209D0 GB202206209D0 (en) | 2022-06-15 |
GB2617865A true GB2617865A (en) | 2023-10-25 |
Family
ID=81850794
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
GB2206209.5A Pending GB2617865A (en) | 2022-04-21 | 2022-04-28 | Method of marking mirror objects in a cross-traffic scenario |
Country Status (1)
Country | Link |
---|---|
GB (1) | GB2617865A (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140225722A1 (en) * | 2011-09-12 | 2014-08-14 | Toyota Jidosha Kabushiki Kaisha | Alarm device |
US20150353078A1 (en) * | 2014-06-04 | 2015-12-10 | Toyota Jidosha Kabushiki Kaisha | Driving assistance apparatus |
JP2016148547A (en) * | 2015-02-10 | 2016-08-18 | トヨタ自動車株式会社 | Detection device |
EP3299841A1 (en) * | 2016-09-22 | 2018-03-28 | Continental Automotive GmbH | Method and apparatus for analyzing reflection signals, driver assistance system, and vehicle |
US20210354717A1 (en) * | 2019-02-01 | 2021-11-18 | Denso Corporation | Warning apparatus |
-
2022
- 2022-04-28 GB GB2206209.5A patent/GB2617865A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140225722A1 (en) * | 2011-09-12 | 2014-08-14 | Toyota Jidosha Kabushiki Kaisha | Alarm device |
US20150353078A1 (en) * | 2014-06-04 | 2015-12-10 | Toyota Jidosha Kabushiki Kaisha | Driving assistance apparatus |
JP2016148547A (en) * | 2015-02-10 | 2016-08-18 | トヨタ自動車株式会社 | Detection device |
EP3299841A1 (en) * | 2016-09-22 | 2018-03-28 | Continental Automotive GmbH | Method and apparatus for analyzing reflection signals, driver assistance system, and vehicle |
US20210354717A1 (en) * | 2019-02-01 | 2021-11-18 | Denso Corporation | Warning apparatus |
Also Published As
Publication number | Publication date |
---|---|
GB202206209D0 (en) | 2022-06-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP3070492B1 (en) | Radar object detection system | |
US9889858B2 (en) | Confidence estimation for predictive driver assistance systems based on plausibility rules | |
US10933883B2 (en) | Driving control apparatus and method for vehicle | |
US20160288799A1 (en) | Sensor abnormality detection device | |
EP3070493B1 (en) | Vehicle radar system with image reflection detection | |
WO2010064282A1 (en) | Pre-crash safety system | |
US20210146921A1 (en) | Driving assistance device | |
US20190033443A1 (en) | Monitoring device and monitoring method | |
CN111341148A (en) | Control system and control method for a motor vehicle for processing multiple reflection signals | |
US11891081B2 (en) | Warning apparatus | |
WO2019142825A1 (en) | Vehicle radar system | |
JP6095197B2 (en) | Vehicle object detection device | |
US10144418B2 (en) | Control apparatus for vehicle | |
US10343696B2 (en) | Travel control device of vehicle | |
GB2617865A (en) | Method of marking mirror objects in a cross-traffic scenario | |
US20230182722A1 (en) | Collision avoidance method and apparatus | |
CN115520131A (en) | Alighting assistance device, alighting assistance method, and non-transitory storage medium | |
CN110733498B (en) | Control device and method for forward collision avoidance of vehicle | |
US20240135823A1 (en) | Vehicle object detection system and method for detecting a target object in a detection area located behind and lateral of a subject vehicle | |
CN115214637A (en) | Backing-up brake assisting method, assisting controller, driving assisting system and automobile | |
EP4358062A1 (en) | Vehicle object detection system and method for detecting a target object in a detection area located behind a subject vehicle | |
EP4357811A1 (en) | Vehicle object detection system and method for detecting a target object in a detection area located behind a subject vehicle | |
US20240131987A1 (en) | Vehicle object detection system and method for detecting a target object in a detection area located behind a subject vehicle | |
US20240132099A1 (en) | Vehicle object detection system and method for detecting a target object in a detection area located behind a subject vehicle | |
JP2024060587A (en) | Vehicle object detection system and method for detecting target objects in detection areas located behind and to the sides of a vehicle - Patents.com |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
COOA | Change in applicant's name or ownership of the application |
Owner name: CONTINENTAL AUTONOMOUS MOBILITY GERMANY GMBH Free format text: FORMER OWNER: CONTINENTAL AUTOMOTIVE GMBH |