Radar blindness and obstruction detector, radar detector system, driving assistance system, and method for detecting radar blindness and obstruction

By combining information from multiple sensors and external data sources, the probability indicators of radar sensors being blind or blocked are calculated, which solves the problem of high false alarm rates in the prior art and improves the reliability and accuracy of radar sensor detection.

CN119087364APending Publication Date: 2024-12-06ROBERT BOSCH GMBH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410722029.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2023-06-05
Filing Date
2024-06-05
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

The prior art has a high false alarm rate when detecting blindness and obstruction of radar sensors, especially under specific environmental conditions, which cannot effectively improve the reliability of detection.

Method used

By combining information from other sensors and external data sources, calculate probability indicators of radar sensor blindness or blockage, adjust filter parameters and thresholds to dynamically evaluate and reduce the risk of false alarms.

Benefits of technology

It improves the reliability of radar sensor blindness and obstructed detection, reduces false alarm rates, and enhances detection accuracy under different environmental conditions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119087364A_ABST
    Figure CN119087364A_ABST
Patent Text Reader

Abstract

A radar sensor is detected to be blinded or blocked. For this purpose, it is proposed to refer to sensor data from one or more further sensors and / or data from an external data source. Based on these additional data, a probability or likelihood that the radar sensor is blinded or blocked is calculated, and a final determination of blinded or blocked radar sensor is performed by evaluating blocked detection of the radar sensor itself in conjunction with the calculated probability. In this way, false detection of radar sensor blindness or blocking can be reduced, and the reliability of radar sensor blindness or blocking detection can be improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to a radar blindness and obstruction detector and a method for detecting radar blindness and obstruction. The invention also relates to a radar detector system and a driver assistance system having such a radar detector system. Background Art

[0002] Modern vehicles can use advanced driver assistance systems (ADAS) for partially or fully automated driving. To this end, sensor signals from multiple sensors are used in order to scan the vehicle's environment and identify surrounding objects or obstacles. For example, radar sensors can be used to detect obstacles in the vehicle's path.

[0003] In order to evaluate the reliability of the radar sensor output, it is necessary to identify faults of the radar sensor such as blindness or obstruction. For example, obstruction can be detected by monitoring the characteristics of the radar signal.

[0004] German patent application DE 10 2021 202 299 A2 describes a method for detecting sensor blindness of a radar sensor, wherein a radar spectrum is determined based on a received radar signal and a portion of the radar spectrum is provided as an input variable to a trained machine learning module. Summary of the invention

[0005] The present invention provides a radar blindness and obstruction detector, a method for detecting radar blindness and obstruction, a radar detector system and a driver assistance system.

[0006] According to a first aspect, a radar blindness and obstruction detector is provided. The radar blindness and obstruction detector comprises a radar sensor and a processing device. The radar sensor is configured to monitor a characteristic of a radar signal. The radar sensor is further configured to provide radar sensor obstruction information based on the characteristic of the monitored radar signal. The processing device is configured to receive sensor data from at least one further sensor. Additionally or alternatively, the processing device may receive data from an external data source. The processing device is further configured to calculate a probability indicator of sensor obstruction of the radar sensor. The probability indicator is calculated based on the sensor data received from at least one further sensor and / or data received from the external data source. The processing device is further configured to determine blindness and obstruction of the radar sensor. Blindness and obstruction are determined by processing the sensor obstruction information from the radar sensor and associating the calculated probability indicator of sensor obstruction.

[0007] According to another aspect, a radar detector system is provided. The radar detector system comprises a radar blindness and obstruction detector according to the first aspect. In the radar detector system, the radar sensor is further configured to transmit a radar signal and receive a radar echo based on the transmitted radar signal. The radar sensor can also identify an object based on the received radar echo. In particular, the radar sensor can determine the distance, spatial position and / or velocity of one or more objects in the field of view of the radar sensor.

[0008] According to a further aspect, a driver assistance system is provided. The driver assistance system may include a radar detector system according to the second aspect. Specifically, the driver assistance system is configured to process radar data from the radar detector. In particular, the driver assistance system may process the radar data based on the determined blindness and obstruction of the radar sensor. For example, the driver assistance system may only consider the output of the radar detector if blindness or obstruction is not detected.

[0009] According to yet another aspect, a method for detecting radar blindness and obstruction is provided. The method comprises the steps of monitoring a characteristic of a returned radar signal by a radar sensor and providing radar sensor obstruction information based on the characteristic of the monitored radar signal. The method further comprises receiving sensor data from at least one further sensor. Additionally or alternatively, data from an external data source may be received. Furthermore, the method further comprises calculating a probability indicator of sensor obstruction of the radar sensor. The probability indicator is calculated based on data received from at least one further sensor and / or data from an external data source. Furthermore, the method further comprises determining blindness and obstruction of the radar sensor. Blindness and obstruction are determined by processing the sensor obstruction information from the radar sensor and in combination with the calculated sensor obstruction probability indicator.

[0010] When using sensor data, in particular for object detection by radar sensors, it is very important to evaluate the reliability of the sensor data provided. If the radar sensor is obstructed, obstacles in the vehicle's environment cannot be correctly identified. For this purpose, conventional radar sensors can perform monitoring of the radar sensor by evaluating the characteristics of the returned radar signal. The monitored characteristics of the radar signal can be, for example, absorption or distortion of the radar signal. For this purpose, the returned radar signal can be evaluated relative to a corrected radar signal. However, depending on the environmental conditions, such monitoring can lead to false detections of obstructions. For example, when driving for a long time through scenes with no objects around, obstructions may be detected even if the radar sensor is working properly due to the lack of returned radar signals.

[0011] The present invention therefore aims to use additional information in order to improve the reliability of the blindness or obstruction detection of the radar sensor. In particular, the present invention does not only perform a simple combination of blindness and obstruction detections by the radar sensor or simple output signals of further sensors, but the present invention also introduces likelihood or probability information for improving the reliability of the detected blindness and obstruction of the radar sensor. In other words, data from one or more additional sensors or data from further external data sources can be used in order to calculate the likelihood that the radar sensor is blind or obstructed. This likelihood is used to evaluate the blindness or obstruction information provided by the radar sensor.

[0012] Therefore, in the case where the probability or possibility of blindness or obstruction is determined to be relatively high based on data from another sensor or external data source, the blindness and obstruction information from the radar sensor is rated as relatively high. On the other hand, if the probability or possibility of blindness or obstruction based on data from another sensor or external data source is low, the blindness and obstruction information from the radar sensor is rated as relatively low. In this way, false alarms of blindness or obstruction of the radar sensor due to specific environmental conditions can be reduced, and the reliability of blindness and obstruction detection can be improved.

[0013] According to one embodiment, the processing device is configured to be able to adjust one or more parameters for determining blindness and obstruction of the radar sensor. In particular, these parameters may relate to filter parameters and / or threshold parameters for determining blindness or obstruction. The parameters may be adjusted based on a calculated probability indicator of sensor obstruction. In this way, the determination of whether the radar sensor is obstructed may be easily adjusted based on the determined probability or likelihood of the radar sensor being blind or obstructed. In other words, in the case where the probability of determining that the radar sensor is blind or obstructed is high, the degree of blindness or obstruction of the radar sensor determined based on the information provided by the radar sensor itself is set to relatively high, wherein, if the calculated probability or likelihood of blindness and obstruction is low, the condition for determining the blindness and obstruction of the radar sensor based on the information provided by the radar sensor is set to low.

[0014] According to one embodiment, the processing device is configured to be able to receive sensor data from an optical sensor, an ultrasonic sensor, a temperature sensor and / or a humidity or rain sensor. The optical sensor can be, for example, a camera or a camera system. The ultrasonic sensor can relate to, for example, a sensor for measuring the distance of an object (e.g. a sensor for a parking distance assistance system). In addition, weather conditions can be determined based on, for example, temperature, humidity or detected rain. Based on the sensor data provided by the corresponding sensors, weather conditions can be determined. Additionally or alternatively, further conditions of the environment such as road conditions or surfaces around the vehicle can be determined in order to assess the blindness and obstruction reliability of the radar sensor. However, it should be understood that any other appropriate sensor can also be used to assess the blindness and obstruction reliability of the radar sensor.

[0015] According to one embodiment, the processing device is configured to be able to receive external data from a cloud server. However, it is also possible to receive external data from any other data source. The external data may relate to, for example, weather data, map data, in particular road map data or any other suitable type of data for assessing the reliability of radar sensor obstruction.

[0016] According to one embodiment, the processing device is configured to detect weather conditions, in particular weather conditions in the environment of the radar sensor. Therefore, the processing device can calculate the probability indicator based on the detected weather conditions. For example, in rainy or snowy conditions, the probability that the radar sensor is blinded or blocked may increase. On the other hand, in sunny or similar good weather conditions, the probability that the radar sensor is blinded or blocked will decrease.

[0017] According to one embodiment, the processing device is configured to detect driving surfaces in the environment of the radar sensor. Accordingly, the processing device can calculate a probability indicator based on the detected driving surfaces. For example, driving through mud or slurry may cause the sensor to become dirty and thus result in a high probability of blinding or obstruction of the radar sensor. On the other hand, driving through an area without any additional objects, such as through a desert or similar area, may result in a situation where the radar sensor will not receive radar echoes for a long time even if blinding or obstruction does not occur. In this scenario, the probability of blindness or obstruction can be set to low so that blindness or obstruction information of the radar sensor based on absorption or distortion evaluation will not result in false alarms of blindness or obstruction.

[0018] According to another embodiment, the processing device is configured to be able to dynamically increase and / or decrease the probability indicator of sensor obstruction. The probability can be increased or decreased based on sensor data received from at least one additional sensor and / or data from an external data source. For example, based on data provided from another sensor or an external data source, the probability or likelihood of blindness or obstruction of a radar sensor can be increased or decreased over time. In this way, sudden changes or effects within a short period of time can be easily ignored.

[0019] According to another embodiment, the processing device is configured to be able to dynamically increase and / or decrease a threshold or timer for detecting sensor obstruction. The threshold or timer can be increased or decreased based on sensor data received from at least one additional sensor and / or data from an external data source. For example, based on data provided from another sensor or an external data source, a threshold or timer for blindness or obstruction detection of a radar sensor can be increased or decreased over time. In this way, sudden changes or effects within a short period of time can be easily ignored. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In the following, possible embodiments of different aspects of the present invention are described in more detail with reference to the accompanying drawings.

[0021] Figure 1 A schematic diagram of a vehicle having a driver assistance system including a radar detector system having a radar blindness and obstruction detector is illustrated according to an embodiment.

[0022] Figure 2 Radar blindness and obstruction detectors according to embodiments; and

[0023] Figure 3 A flow chart illustrating a method for radar blindness and obstruction detection according to an embodiment. DETAILED DESCRIPTION

[0024] Figure 1 A schematic block diagram illustrating a vehicle with a driver assistance system 2, such as an advanced driver assistance system (ADAS), according to an embodiment is shown. The driver assistance system may be used to partially or fully automatically drive the vehicle or at least to assist the driver during driving the vehicle.

[0025] The driver assistance system 2 may include a radar detector system 1 for detecting objects O in the vehicle environment. The radar detector system 1 may generate and transmit radar signals and receive radar echoes of the transmitted radar signals. Based on the received radar echoes, the radar detector system 1 may detect one or more objects O. In particular, the radar detector system 1 may provide information about, for example, the position, distance and / or speed of one or more detected objects O.

[0026] In addition, the driving assistance system 2 of the vehicle may include one or more additional sensors 20. The additional sensors 20 may be, for example, optical sensors, such as cameras, LiDAR sensors or similar sensors. In addition, the additional sensors 20 may include ultrasonic sensors. Such ultrasonic sensors can emit ultrasonic signals and receive echoes of the emitted ultrasonic waves to detect obstacles in the field of view of the sensor 20 and determine the distance between the sensor 20 and the obstacles. Additionally or alternatively, the additional sensors 20 may include any other type of appropriate sensors, such as temperature sensors, humidity sensors, rain sensors, etc. The additional sensors 20 may also include, for example, a global navigation satellite system (GNSS), such as GPS, Galileo or similar systems, to determine the spatial position of the vehicle. Based on such a determined position, the system can refer to map data to obtain further information, such as the specifications of the driving surface in the road or vehicle environment.

[0027] Additionally or alternatively, the driving assistance system 1 may include a communication interface 30. The communication interface 30 may establish a communication link to an external data source 100. The external data source 100 may be, for example, a cloud server or a similar device. In this way, the external data source 100 may provide additional information, such as traffic information, further information about available roads or driving surfaces in the vehicle environment. Additionally or alternatively, the external data source 100 may provide, for example, information about weather conditions or any other suitable data that the driving assistance system may use.

[0028] As described above, the radar detector system 1 can perform object detection by transmitting radar signals and receiving radar echoes. Accordingly, objects in the field of view of the radar detector system 1 are identified by evaluating the radar echoes. However, if the transmission of the radar signal and / or the reception of the radar echo is disturbed, a correct detection of the object cannot be performed. For example, the transmission of the radar signal or the reception of the radar echo may be disturbed or blocked due to contaminants such as dirt on the antenna. This may occur, for example, when driving through mud or slush areas. In addition, snow or ice may be located on the surface of the radar antenna. However, there may also be other reasons that affect the transmission or reception of the radar signal.

[0029] If proper object detection cannot be performed, the radar detector system 1 may be considered blind or obstructed.

[0030] Conventional radar detection systems can perform detection of a blind or obstructed system by monitoring the characteristics of the returning radar signal. However, there may be some situations where the radar detector system 1 is unable to receive the radar signal even though it is operating normally. For example, if the vehicle is traveling in an area without any additional objects, such as a desert or similar area, the radar detector system 1 will not receive the radar return. In this case, the radar detector system 1 may believe that the radar detector is in a blind or obstructed condition even though the system is operating normally. Therefore, the driver assistance system 1 will receive a notification that the radar detector system is obstructed even though the system is operating normally.

[0031] In order to improve the reliability of blindness or obstruction detection of the radar detector system 1, the present invention performs enhanced blindness and obstruction detection by further taking additional information into account. In particular, a method is provided for determining the likelihood or probability of blindness of a radar sensor, and wherein blindness detection information of the radar sensor is further processed and evaluated based on the likelihood or probability.

[0032] Figure 2 A radar detector system 1 with a blinding and obstruction detector 10 according to one embodiment is shown. The radar sensor 11 may perform monitoring of returned radar signals. Based on the monitoring, the radar sensor 11 may provide sensor obstruction information. The sensor obstruction information may indicate whether the radar detector system 1 is likely to be obstructed. For example, if no radar return is received within a predetermined time period, the sensor obstruction information may indicate that the radar detector system 1 is obstructed. However, any further suitable method for evaluating the radar detector system 1 in order to identify obstruction of the radar detector system 1 is also feasible.

[0033] The radar blindness and obstruction detector 10 may further include a processing device 12. The processing device 12 may receive sensor obstruction information from the radar sensor 11. In addition, the processing device 12 may also receive sensor data from at least one further sensor 20. Additionally or alternatively, the processing device may also receive data from an external data source. For example, the processing device 12 may receive data from an external data source 100 via the communication interface 30.

[0034] The processing device 12 may calculate a probability indicating a likelihood of obstruction of the radar sensor 11. To this end, the processing device may take into account sensor data from at least one further sensor 20. Additionally or alternatively, the processing device may take into account data from an external data source 100.

[0035] For example, the processing device 12 may analyze optical data, such as camera images, in order to assess the vehicle's environment. In particular, the processing device 12 may assess the vehicle's driving surface. If environmental information, such as the driving surface, leads to the conclusion that the driving surface is likely to be almost flat without any objects or at least with only a very small number of radar-relevant objects. This information may lead to the conclusion that the radar detector system 1 may not receive significant radar returns. However, since additional information, such as information about the driving surface, may be used to infer that the radar sensor may not be obstructed or blinded, the processing device 12 may determine that the probability that the radar detector system 1 is blinded or obstructed is very low.

[0036] Furthermore, if the additional data from the further sensor 20 is used to identify a very dirty environment with a lot of mud or slurry, such an environment may result in a lot of dirt in front of the radar antenna. Accordingly, the probability of the radar detector system 1 being blinded or blocked may be high.

[0037] In addition to this, the probability of a radar sensor 11 being blinded or blocked can also be derived from weather conditions. For example, heavy rain or snow can also cause a radar sensor 11 to be blinded or blocked. Conversely, good and cloudless weather conditions (e.g. dry and sunny weather) may generally result in a very low probability of a radar sensor 11 being blinded or blocked. Weather conditions can also be derived from optical data such as camera images or temperature sensors or rain sensors.

[0038] In addition to this, information about driving surfaces in the vehicle environment, weather information or similar information may also be received from the external data source 100. Such additional data from the external data source 100 may also be considered when determining the probability or likelihood that the radar sensor 11 is blinded or obstructed.

[0039] The likelihood or probability that the radar sensor 11 is blinded or obstructed may not only be calculated based on information related to a single point in time. Moreover, the probability that the radar sensor 11 is blinded or obstructed may also be adjusted dynamically. For example, based on newly received data from further sensors 20 and / or additional data from external data sources 100, a predetermined probability or probability may be increased or decreased. In this way, a single event in the acquired data may not immediately lead to a rapid change in the determined probability or probability that the radar sensor 11 is blinded or obstructed.

[0040] After determining the probability or likelihood that the radar sensor 11 is blinded or obstructed, the processing device 12 may consider the obstruction information from the radar sensor 11 in conjunction with the determined probability that the radar sensor 11 is blinded or obstructed. To this end, a threshold for determining whether the information provided by the radar sensor 11 will lead to a decision of whether the radar sensor 11 is blinded or obstructed may be adjusted according to the determined probability that the radar sensor 11 is blinded or obstructed. Alternatively, a weighting factor for evaluating the obstruction information from the radar sensor may be adjusted according to the calculated probability that the radar sensor 11 is blinded or obstructed. However, any further suitable scheme for adjusting the evaluation of the obstruction information from the radar sensor 10, such as modification of filter parameters, thresholds, timers, etc., is also feasible.

[0041] Based on the evaluation of the obstruction information from the radar sensor 11, in combination with the probability of blindness or obstruction calculated based on the additional information from the further sensor 20 and / or the additional data source 100, the processing device 12 may perform a decision whether the radar sensor 11 may be considered blind and obstructed. This decision may be used by the driver assistance system 2. For example, the driver assistance system 2 may consider information from the radar detector system 1 only if the radar detector system 1 is not considered blind and obstructed. Otherwise, the information from the radar detector system 1 may be skipped. However, any other suitable method for considering information from the radar detector system 1 based on the detection of a blind and obstructed radar sensor 11 is also feasible.

[0042] Figure 3 A schematic flow chart of a method for blindness and obstruction detection of radar sensor 11 according to one exemplary embodiment is shown.

[0043] Radar sensor 11 may monitor the characteristics of the returning radar signal in step S1 a and provide obstruction information of radar sensor 11 based on the monitoring result in step S1 b .

[0044] Furthermore, in step S2, sensor data may be received from one or more further sensors 20. Additionally or alternatively, further data from an external data source 100 may be received.

[0045] In step S3 , a probability or likelihood that radar sensor 11 is blocked or blinded may be calculated based on data from one or more further sensors 20 and / or external data source 100 .

[0046] Finally, in step S4 , the blindness and obstruction of radar sensor 11 is determined by taking into account the obstruction information from radar sensor 11 in accordance with the determined probability or likelihood of blindness and obstruction of radar sensor 11 .

[0047] In summary, the present invention relates to the detection of a radar sensor being blinded or obstructed. To this end, it is proposed to refer to sensor data from one or more further sensors and / or data from external data sources. Based on these additional data, a probability or likelihood that the radar sensor is blinded or obstructed is calculated, and a final decision on the radar sensor being blinded or obstructed is performed by evaluating the obstructed detection of the radar sensor itself in conjunction with the calculated likelihood that the radar sensor is blinded or obstructed. In this way, false detections of the radar sensor being blinded or obstructed can be reduced and the reliability of the detection of the radar sensor being blinded or obstructed can be improved.

Claims

1. A radar blindness and obstruction detector (10), comprising: A radar sensor (11) configured to provide sensor obstruction information by monitoring characteristics of a radar signal; A processing device (12) is configured to receive sensor data from at least one further sensor (20) and / or data from an external data source (100), to calculate a probability indicator of sensor obstruction of the radar sensor (11) based on the sensor data received from the at least one further sensor (20) and / or the data from the external data source (100), and to determine blindness and obstruction of the radar sensor (11) by processing sensor obstruction information from the radar sensor (11) using the calculated probability indicator of sensor obstruction.

2. The radar blindness and obstruction detector (10) according to claim 1, wherein: The processing device (12) is configured to adjust one or more filter or threshold parameters used to determine blindness and obstruction of the radar sensor (11) based on the calculated probability indicator of sensor obstruction.

3. The radar blindness and obstruction detector (10) according to claim 1 or 2, wherein: The processing device (12) is configured to receive sensor data from an optical sensor, an ultrasonic sensor, a temperature sensor and / or a humidity or rain sensor.

4. The radar blindness and obstruction detector (10) according to any one of claims 1 to 3, wherein: The processing device (12) is configured to be able to receive data from an external cloud server.

5. The radar blindness and obstruction detector (10) according to any one of claims 1 to 4, wherein: The processing device (12) is configured to detect weather conditions in an environment of the radar blindness and obstruction detector (10) and to calculate the probability indicator based on the detected weather conditions.

6. The radar blindness and obstruction detector (10) according to any one of claims 1 to 5, wherein: The processing device (12) is configured to detect driving surfaces in the environment of the radar blindness and obstruction detector (10) and to calculate the probability indicator based on the detected driving surfaces.

7. The radar blindness and obstruction detector (10) according to any one of claims 1 to 6, wherein: The processing device (12) is configured to dynamically increase and / or decrease the sensor obstruction probability indicator based on sensor data received from at least one other sensor and / or data from an external data source (100).

8. The radar blindness and obstruction detector (10) according to any one of claims 1 to 7, wherein: The processing device (12) is configured to dynamically increase and / or decrease a threshold or timer for detecting sensor obstruction.

9. A radar detector system (1) comprising a radar blindness and obstruction detector (10) according to any one of claims 1 to 8.

10. A driver assistance system (2) comprising a radar detector system (1) according to claim 9, in, The driver assistance system (2) is configured to process radar data from the radar detector system (1) based on the determined blindness and obstruction of the radar sensor (11).

11. A radar blinding and obstruction detection method comprising: (S1a) monitoring a characteristic of a radar signal through a radar sensor, and (S1b) providing sensor obstruction information based on the characteristic of the radar signal; (S2) receiving sensor data from at least one further sensor (20) and / or receiving data from an external data source (100); (S3) calculating a probability indicator of sensor obstruction of the radar sensor (11) based on sensor data received from at least one further sensor (20) and / or data from an external data source (100); and (S4) The blindness and obstruction of the radar sensor (11) are determined by processing the sensor obstruction information from the radar sensor (11) using the calculated probability index of sensor obstruction.