Method and system for monitoring vehicle surroundings - Patents.com

By locally adapting monitoring range boundaries based on measurement deviations, the method addresses measurement uncertainties in vehicle object detection systems, enhancing detection reliability and reducing false positives.

JP2024545191AInactive Publication Date: 2024-12-05ZF FRIEDRICHSHAFEN AG
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
JP2024534717
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Priority Date
2021-12-22
Filing Date
2022-11-23
Publication Date
2024-12-05
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Existing object detection systems in vehicles suffer from measurement uncertainties that lead to unreliable decision-making due to varying measurement deviations in the monitoring range, resulting in false positive detections.

Method used

The method involves locally adapting the monitoring range boundaries based on locally determined measurement deviations, such as angular and distance measurement errors, to ensure reliable object detection by expanding or reducing the range boundaries as needed, thereby minimizing false positives.

Benefits of technology

This approach enhances the reliability of object detection by ensuring that detected objects are accurately within the monitoring range, reducing false positives and improving the certainty of object assignment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024545191000001_ABST
    Figure 2024545191000001_ABST
Patent Text Reader

Abstract

A method and system for detecting objects in the vicinity of a vehicle is provided. [Solution] A method for detecting objects around the vehicle 100 by a surrounding detection sensor 10 arranged on the vehicle 100, comprising a step of setting a monitoring range 102 in the detection range of the surrounding detection sensor 10 to monitor for objects remaining in the monitoring range 102 in order to monitor the surroundings of the vehicle 100 by the surrounding detection sensor 10, the method further comprising a step of locally identifying a measurement deviation 14 for the measurement value of the surrounding detection sensor in the set monitoring range 102, and a step of adapting the range boundary 104 of the set monitoring range 102 to the locally identified measurement deviation 14 in order to monitor the surroundings of the vehicle 100 by the surrounding detection sensor 10 for objects remaining in the adapted monitoring range.
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to a method and system for detecting objects in the surroundings of a vehicle. [Background technology]

[0002] It is known from the prior art to detect objects in the surroundings of a vehicle by means of a surrounding detection sensor arranged in the vehicle. Such surrounding detection can form the basis for a driver assistance system or an automated operation of the vehicle. However, object detection by the surrounding detection sensor can involve measurement technology uncertainties that can make decisions based on object detection for operating the vehicle unreliable.

[0003] In order to avoid possible collisions of a vehicle with objects, it is known from DE 10 2005 13 2005 A1 to determine possible intersection paths of the objects with the vehicle based on the objects detected in the vehicle's surroundings. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] International Publication No. 2018 / 045055 Summary of the Invention [Problem to be solved by the invention]

[0005] In one aspect, the present invention relates to a method for detecting objects in the surroundings of a vehicle by means of a surroundings detection sensor disposed in the vehicle. [Means for solving the problem]

[0006] The vehicle may be essentially any vehicle, for example a non-rail bound vehicle or a rail bound vehicle. The vehicle may be a traffic participant. According to one embodiment, the vehicle is a commercial vehicle, for example a truck. In a commercial vehicle, monitoring the surroundings of the commercial vehicle for objects that remain around the commercial vehicle may be a pre-set premise for operating the commercial vehicle.

[0007] The surroundings of the vehicle may comprise at least one spatial extent around the vehicle. The spatial extent may be a two-dimensional spatial extent around the vehicle, e.g. a horizontal or vertical plane, or a three-dimensional spatial extent. The object may be a static object or a dynamic object around the vehicle. The object may be a traffic participant that remains statically or dynamically around the vehicle. The object may be, for example, another vehicle, a pedestrian or a street installation.

[0008] The surroundings detection sensor may comprise at least one detection sensor. The surroundings detection sensor or the detection sensor may comprise at least one of a sensor for measuring distance and a sensor for detecting images. The sensor for measuring distance may be formed as a scanning sensor. The sensor for measuring distance may be, for example, a radar measuring device, a laser scanner or an ultrasonic measuring device. The sensor for detecting images may be formed as a camera.

[0009] The method includes as a step setting a monitoring range in a detection range of the perimeter detection sensor. The step of setting the monitoring range is performed for monitoring the surroundings of the vehicle by the perimeter detection sensor for objects remaining in the monitoring range. The method may also include monitoring the surroundings of the vehicle by the perimeter detection sensor for objects remaining in the monitoring range. In a further step of the method, it is possible to determine whether an object detected by the perimeter detection sensor is in the monitoring range.

[0010] The detection range may be a range in the surroundings of the vehicle in which the surroundings detection sensor can detect an object. The detection range may therefore be a measurement range in which the surroundings detection sensor can generate measurements relating to the object. The detection range may be defined by at least one angle measurement range of the surroundings detection sensor. Alternatively or additionally, the detection range may be defined by at least one distance measurement range of the surroundings detection sensor. The monitoring range may be formed in the detection range as a subrange of the detection range. The monitoring range may therefore be set relative to the vehicle in the detection range. According to one embodiment, the monitoring range is formed laterally to the vehicle in the driving direction of the vehicle.

[0011] The method comprises, as a further step, determining a measurement deviation locally for at least one measured value of the surroundings detection sensor in the set monitoring range. The measurement deviation can be determined locally in a range boundary of the set monitoring range for the at least one measured value of the surroundings detection sensor. Alternatively or additionally, the measurement deviation can be determined locally in a subrange of the set monitoring range for the at least one measured value of the surroundings detection sensor. The measurement deviation can represent or be based on a measurement uncertainty for the measured value of the surroundings detection sensor.

[0012] The present invention is based on the realization that the measurement deviations can vary significantly in the set monitoring range for the measurement values ​​of the surroundings detection sensor. Thus, the measurement deviations can differ significantly from one another in different subranges of the monitoring range depending on the corresponding measurement values. The locally determined measurement deviations can depend on the relative situation of the object detected by the surroundings detection sensor with respect to said surroundings detection sensor. The relative situation of the detected object can comprise the relative position of the object with respect to the surroundings detection sensor. Thus, the measurement deviation can be a position deviation of the detected position of the object with respect to the target position of the object. The relative situation of the detected object can comprise the relative orientation of the object with respect to the surroundings detection sensor. The locally determined measurement deviations can depend on at least one of the angle measurement and the distance measurement of the surroundings detection sensor with respect to the objects remaining around the vehicle.

[0013] The method comprises as a further step locally adapting a range boundary of the set monitoring range to the locally determined measurement deviation. The locally adapting step is performed for monitoring the surroundings of the vehicle by means of a surroundings detection sensor for objects remaining in the adapted monitoring range. The method can comprise as a further step monitoring the surroundings of the vehicle by means of a surroundings detection sensor for objects remaining in the adapted monitoring range.

[0014] The local adaptation of the range boundaries can include a local expansion of the range boundaries based on the locally identified measurement deviations. Thus, the local adaptation of the range boundaries can include a local expansion of the range boundaries by the locally identified measurement deviations. Alternatively or in addition to the local expansion, the local adaptation of the range boundaries can also include a local narrowing of the range boundaries based on the locally identified measurement deviations.

[0015] The present invention is based on the understanding that for a reliable object detection in the monitoring range in terms of measurement technology, it is sufficient to locally adapt the spatial extension of the set monitoring range to the spatially varying measurement deviation of the surrounding detection sensor. Therefore, in order to ensure a reliable object detection in the monitoring range, the monitoring range can be locally extended based on the locally determined measurement deviation. Thus, the locally adapted monitoring range can be a locally extended monitoring range compared to the set monitoring range.

[0016] According to the invention, it is possible to locally adapt the monitoring range of the perimeter detection sensor to the spatially varying measurement deviation of the perimeter detection sensor. The adapted monitoring range can have a safety range which can be arranged around a range boundary of the set monitoring range. The set monitoring range and the safety range can form the adapted monitoring range. A local extension of the safety range can correspond to the local measurement deviation. The safety range can thus extend along the range boundary of the set monitoring range depending on the locally determined measurement deviation. The safety range can thus form a locally varying safety belt around the set monitoring range.

[0017] An advantageous effect of the present invention may be that it is possible to minimize false positive object detections by the perimeter detection sensor, i.e. objects that are actually outside the monitoring range but are detected and assigned to the monitoring range. The advantageous effect may be based on the fact that the monitoring range is only locally extended in such a way that a local measurement deviation or local measurement certainty of the perimeter detection device is necessary in order to also assign objects that are detected with a certain certainty to the monitoring range.

[0018] According to an embodiment of the method, in the step of locally determining the measurement deviation, a measurement deviation can be locally determined for the measurement value based on an angle measurement error of the surroundings detection sensor. The measurement deviation can be a corresponding distance-dependent measurement deviation. The measurement value can be an angle measurement relative to an object or a localization of the object, which can have an angle measurement error. The angle measurement error can be locally determined depending on the measurement distance relative to the object. The measurement deviation can thus be a measurement uncertainty in the angle measurement or localization by the surroundings detection sensor. The measurement deviation can be a distance-dependent measurement uncertainty. The angle measurement or measurement value can relate to a measured angle relative to the object or a measured position of the object in the sensor coordinate system of the surroundings detection sensor. The measurement deviation or angle measurement error can comprise a random angle measurement error, for example an angle measurement error following a normal distribution. Alternatively or additionally, the measurement deviation or angle measurement error can comprise a systematic angle measurement error, for example due to an incorrect calibration of the surroundings detection sensor.

[0019] According to another embodiment of the method, in the step of locally determining the measurement deviation, a measurement deviation can be locally determined for the measurement value based on a distance measurement error of the surroundings detection sensor. The measurement deviation can be a measurement deviation that depends on the corresponding distance. The measurement value can be a distance measurement or a localization of the object, which can have a distance measurement error. The distance measurement error can be locally determined depending on the measurement distance of the object. The measurement deviation can thus be a measurement uncertainty in the distance measurement or localization by the surroundings detection sensor. The measurement deviation can be a measurement uncertainty that depends on the distance. The distance measurement or the measurement value can relate to a measured distance to the object or a measured position of the object in the sensor coordinate system of the surroundings detection sensor. The measurement deviation or the distance measurement error can be a random distance measurement error, for example a distance measurement error that follows a normal distribution. Alternatively or additionally, the measurement deviation or the distance measurement error can be a systematic distance measurement error, for example due to an incorrect calibration of the surroundings detection sensor.

[0020] According to another embodiment of the method, in the step of locally determining the measurement deviation, the measurement deviation can be locally determined based on predefined technical data of the perimeter detection sensor. The data can be operational data of the sensor. The predefined technical data can comprise manufacturer-specific data on the measurement deviation of the perimeter detection sensor. The predefined technical data can comprise the measurement deviation of the perimeter detection sensor depending on the distance to the object measured by the perimeter detection sensor, the angle to the object measured by the perimeter detection sensor or the position to the object measured by the perimeter detection sensor.

[0021] According to another embodiment of the method, in the step of locally determining the measurement deviation, the measurement deviation can be locally determined based on an empirical test of the surroundings detection sensor for detecting objects in the surroundings of the vehicle. The empirical test can comprise comparing a measured distance to the test object, an angle measured to the test object or a measured position of the test object in the monitoring range with a corresponding target quantity. The empirical test can thus comprise comparing a measured position of the test object with a target position. Based on the comparison, which can be a target-to-actual comparison, it is possible to derive the locally determined measurement deviation. In performing the empirical test, the test object can be at a range boundary of the set monitoring range. The measurement deviation can thus be locally determined at the range boundary.

[0022] According to another embodiment of the method, in the step of locally determining the measurement deviation, the measurement deviation for the measurement values ​​of the surrounding detection sensor can be locally determined along a range boundary of the set monitoring range. The measurement deviation can be locally determined along the range boundary based on an angle measurement error. Alternatively or additionally, the measurement deviation can be locally determined along the range boundary based on a distance measurement error. Furthermore, alternatively or additionally, the measurement deviation can be locally determined along the range boundary based on an error in localization. The measurement deviation can be locally determined along the range boundary of the set monitoring range based on predetermined technical data. Alternatively or additionally, the measurement deviation can be locally determined along the range boundary of the set monitoring range based on empirical tests.

[0023] According to the above-mentioned embodiment, in the step of locally adapting the range boundary of the set monitoring range, the adapted monitoring range may have a polygon. The polygon can be adapted to a predefined measurement deviation for at least one measured value. The polygon can draw (form) a safety zone around the set monitoring range, in which objects detected by the surrounding detection sensor are assigned to the set monitoring range based on the locally determined measurement deviation. The spacing of the polygon relative to the set monitoring range or the spacing of the range boundary of the adapted monitoring range relative to the range boundary of the set monitoring range may vary locally. The local variation of the spacing may be based on the locally determined measurement deviations independently of each other.

[0024] According to another embodiment of the method, in the step of locally adapting the range boundary of the set monitoring range, the range boundary can be locally extended on the basis of a locally determined measurement deviation. The range boundary can be locally extended on the basis of a locally determined angle measurement error. Alternatively or additionally, the range boundary can be locally extended on the basis of a locally determined distance measurement error. The range boundary can be extended by only a part of the measurement deviation, for example by half the measurement deviation.

[0025] According to another embodiment of the method, the method may comprise, as another step, locally setting a measurement error range for at least one measurement value of the surrounding detection sensor based on the locally determined measurement deviation. The locally set measurement error range can be drawn based on a locally determined angle measurement error. Alternatively or additionally, the locally set measurement error range can be drawn locally based on a locally determined distance measurement error. For example, the measurement error range can be a measurement error ellipse drawn based on the angle measurement error and the distance measurement error. The measurement error ellipse can be a confidence interval of the corresponding measurement uncertainty.

[0026] According to the above-mentioned embodiment, in the step of locally adapting the range boundary of the set monitoring range, the range boundary can be locally extended based on the locally set measurement error range, and the set measurement error range can be extended by a partial range of the locally set measurement error range that goes outside the set monitoring range.

[0027] In another aspect, the present invention relates to a system for detecting objects in the surroundings of a vehicle, the system comprising a surroundings detection sensor disposable in the vehicle, the surroundings detection sensor being as described in the previous aspect, and a computing unit configured to set a monitoring range in a detection range of the surroundings detection sensor for monitoring the surroundings of the vehicle for objects remaining in the monitoring range.

[0028] The computing unit is configured to locally determine a measurement deviation for at least one measured value of the surroundings detection sensor in the set monitoring range, and further configured to adapt a range boundary of the set monitoring range to the locally determined measurement deviation for monitoring the surroundings of the vehicle for objects remaining in the adapted monitoring range.

[0029] In another aspect, the present invention relates to a vehicle equipped with a surroundings detection sensor. In addition, the vehicle is equipped with a system according to the above aspect. The vehicle, the surroundings detection sensor and the system can be formed as described in the above aspect.

[0030] Embodiments and features of one aspect of the invention may form corresponding embodiments and features of another aspect of the invention. The system may comprise suitable system components that may be configured to perform the method steps described in the method. [Brief description of the drawings]

[0031] [Figure 1] 1 illustrates a vehicle and system according to embodiments of the present invention; [Diagram 2] FIG. 1 shows a diagram illustrating the localization of measurement deviations to explain the present invention. [Diagram 3] FIG. 13 shows another local determination of the measurement deviation for another illustration of the present invention. [Figure 4] FIG. 13 illustrates locally adapting a monitoring range according to an illustrative embodiment of the present invention. [Diagram 5] FIG. 11 illustrates another locally adapted monitoring range according to another illustrative embodiment of the present invention. [Figure 6] FIG. 2 shows a flow chart of method steps for detecting objects in the surroundings of a vehicle, according to one embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0032] In Fig. 1 a vehicle 100 is shown. The vehicle 100 comprises a system 200 comprising an environment detection sensor 10 and a computing unit 202. The computing unit is adapted to carry out the method steps illustrated in Fig. 6.

[0033] In the embodiment shown in FIG. 1, the surroundings detection sensor 10 comprises two detection sensors 11. The detection sensors 11 each have a detection range 12, which form a common detection range 12 of the surroundings detection sensor 10. An object 2 remaining in the surroundings 4 of the vehicle 100 is detected by the surroundings detection sensor 10 in the detection range 12. The object 2 then remains in the detection range 12. In the surroundings 4 and in the detection range 12, a monitoring range 102 is set relative to the vehicle 100. In the embodiment shown in FIG. 1, the monitoring range 102 is set at a distance from the longitudinal axis of the vehicle 100 or to the side of the vehicle 100. The monitoring range 102 has a range boundary 104, within which the object 2 can be detected by the surroundings detection sensor 10. The monitoring range 102 is monitored by the surroundings detection sensor 10 for the object 2 remaining in the monitoring range.

[0034] In Fig. 2, a measurement deviation 14 during the detection of an object 2 by the perimeter detection sensor 10 is locally determined on the basis of empirical tests 18. In the empirical tests 18, a test object 3 moves along a range boundary 104. In the embodiment shown in Fig. 2, the test object 3 moves along at least one range boundary 104 extending in the main extension direction of the monitoring range 102. In this case, the test object 3 moves relative to the vehicle 100, which is stationary during the empirical tests 18. The perimeter detection sensor 10 detects an object trajectory 19 of the test object 3, which locally differs from the extension of the range boundary 104. The local measurement deviation 14 is locally determined on the basis of the deviation of the detected object trajectory 19 from the range boundary 104. The locally determined measurement deviation 14 varies along the range boundary 104 at least depending on the measurement distance of the perimeter detection sensor 10 during the detection of the test object 3.

[0035] In the embodiment shown in Fig. 3, the test object 3 moves along each range boundary 104 extending in a sub-extension direction of the monitoring range 102. The locally determined measurement deviation 14 varies along the range boundary 104 with increasing distance when the test object 3 is detected by the surrounding detection sensor 10. In Figs. 2 and 3, the measurement deviation 14 describes (forms) a measurement error range 20 along the range boundary 104, which measurement error range includes the locally determined measurement deviation 14. The width of the measurement error range 20 varies locally. The measurement error range 20 is a polygonal range, the measurement error range 20 having a boundary extending non-parallel to the range boundary 104 of the monitoring range 102.

[0036] 4 shows a monitoring range 103 adapted to the locally determined measurement deviation 14. The adapted monitoring range 103 is expanded by the locally determined measurement deviation 14. The locally determined measurement deviation 14 describes a polygon 106 forming a boundary range of the adapted monitoring range 103. Objects 2 remaining within the range boundary of the adapted monitoring range 103 are assigned to a monitoring range 102 based on the locally determined measurement deviation 14. Objects 2 not detected within the adapted monitoring range 103 are not assigned to the monitoring range 102.

[0037] 5 shows a locally determined measurement deviation 14 according to a further embodiment and a range boundary 104 of a set monitoring range 102 that is locally adapted to the locally determined measurement deviation 14. The measurement deviation 14 is determined or read out on the basis of technical data 16. An angle measurement error 22 and a distance measurement error 24 are determined locally along the range boundary 104 of the monitoring range 102 on the basis of the technical data 16. The local determination of the corresponding errors takes place at the bend of a polygon 106. The angle measurement error 22 and the distance measurement error 24 describe an error ellipse along the range boundary 104 as a measurement error range 20. The measurement error range 20 again describes a polygon 106 that forms the range boundary of the locally adapted monitoring range 103.

[0038] In Fig. 6 a flow chart of method steps S1 to S3 of a method for detecting objects in the surroundings 4 of the vehicle 100 by means of a surroundings detection sensor 10 arranged on the vehicle 100 is shown diagrammatically. The method is executed by a system 200. In a first step S1 a monitoring range 102 is set in the detection range 12 of the surroundings sensor 10. In a further step S2 a measurement deviation 14 of the surroundings detection sensor 10 is locally determined in the set monitoring range 102. In a further step S3 a range boundary 104 of the set monitoring range 102 is adapted to the locally determined measurement deviation 14. [Explanation of symbols]

[0039] 2 objects 3 Test Object 4. Surroundings 10 Surrounding detection sensor 11 Detection sensor 12 Detection range 14 Measurement deviation 16 Technical data 18 Empirical Tests 19 Detected object trajectories 20 Measurement error range 21 Error Ellipse 22 Angle measurement error 24 Distance measurement error 100 vehicles 102 Monitoring range 103 Adapted monitoring range 104 Range Boundary 106 Polygon 200 Systems 202 Calculation Unit S1 Monitoring range setting S2 Identifying measurement deviations S3 Range Boundary Fitting

Claims

1. A method for detecting an object (2) in a surrounding (4) of a vehicle (100) by a surrounding detection sensor (10) arranged in the vehicle (100), the method comprising a step (S1) of setting a monitoring range (102) for monitoring an object (2) remaining in the monitoring range (102) in a detection range (12) of the surrounding detection sensor (10) in order to monitor the surroundings (4) of the vehicle (100) by the surrounding detection sensor (10), the method comprising: The method comprises a step (S2) of locally determining a measurement deviation (14) for at least one measured value of the surroundings detection sensor (10) in the set monitoring range (102), and a step (S3) of locally adapting a range boundary (104) of the set monitoring range (102) to the measurement deviation (14) locally determined for monitoring the surroundings (4) of the vehicle (100) by the surroundings detection sensor (10) for an object (2) remaining in an adapted monitoring range (103).

2. 2. The method according to claim 1, characterized in that in the step (S2) of locally determining the measurement deviation (14), the measurement deviation (14) is determined locally for the measured value based on an angle measurement error (22) of the surrounding detection sensor (10).

3. 3. The method according to claim 1 or 2, characterized in that in the step (S2) of locally determining the measurement deviation (14), the measurement deviation (14) is determined locally for the measured value based on a distance measurement error (24) of the surrounding detection sensor (10).

4. 4. The method according to claim 1, wherein in the step (S2) of locally determining the measurement deviation (14), the measurement deviation (14) is locally determined based on predefined technical data (16) of the surrounding detection sensor (10).

5. 5. The method according to claim 1, wherein in the step (S2) of locally determining the measurement deviation (14), the measurement deviation (14) is locally determined based on empirical testing (18) of the surroundings detection sensor (10) detecting objects (2) in the surroundings (4) of the vehicle (100).

6. 6. The method according to claim 1, characterized in that in the step (S2) of locally determining the measurement deviation (14), the measurement deviation (14) for at least one measured value of the surrounding detection sensor (10) is determined locally along the range boundary (104) of the set monitoring range (102), and in the step (S3) of locally adapting the range boundary (104) of the set monitoring range (102), the adapted monitoring range (103) has a polygon (106) that is adapted to the determined measurement deviation (14) for the at least one measured value.

7. 7. The method according to claim 1, wherein in the step (S3) of locally adapting the range boundary (104) of the set monitoring range (102), the range boundary (104) is locally expanded based on the locally determined measurement deviation (14).

8. 8. The method according to claim 1, further comprising a further step of locally setting the measurement error range (20) for the at least one measured value of the surrounding detection sensor (10) based on the locally determined measurement deviation (14), characterized in that in the step (S3) of locally adapting the range boundary (104) of the set monitoring range (102), the range boundary (104) is locally expanded based on the locally set measurement error range (20).

9. A system (200) for detecting an object (2) in a surrounding (4) of a vehicle (100), comprising a surrounding detection sensor (10) that can be arranged on the vehicle (100) and a computing unit (202) configured to set a monitoring range (102) in a detection range (12) of the surrounding detection sensor (10) in order to monitor the surrounding (4) of the vehicle (100) for an object (2) remaining in the surrounding (4) of the vehicle (100), The system (200) is characterized in that the computing unit (202) is configured to locally determine a measurement deviation (14) for at least one measured value of the surroundings detection sensor (10) in the set monitoring range (102) and to locally adapt a range boundary (104) of the set monitoring range (102) to the locally determined measurement deviation (14) for monitoring the surroundings (4) of the vehicle (100) for an object (2) remaining in an adapted monitoring range (103).

10. In a vehicle (100) equipped with a surrounding detection sensor (10), A vehicle (100), characterized in that the vehicle is equipped with a system (200) according to claim 9.

Citation Information

Patent Citations

  • Moving object detection device for vehicle

    JP2013024673A

  • Object position detector and object position detection method

    JP2015087211A

  • Object detection device

    JP2017111054A

  • Target detection apparatus

    JP2017211249A

  • Improved detection of a target object utilizing automotive radar

    WO2018045055A1