Method for determining perceptual performance, intelligent driving method and related device

Abstract

The invention relates to a method for determining perceptual performance, an intelligent driving method and a related device. The method includes obtaining first environmental data, wherein the first environmental data indicates environmental characteristics of a region of interest of a sensor at a first moment. The method also includes acquiring an environmental performance relationship set corresponding to the sensor, wherein the environmental performance relationship set includes at least one environmental performance relationship indicating a correspondence between the environmental data and the perceptual performance information. In addition, the method further comprises the step of determining the perceptual performance of the sensor in the region of interest at the first moment based on the first environment data and the environment performance relation set. In this way, the perceptual performance of the sensor can be directly determined using the environmental data and the set of environmental performance relationships without relying on a truth-valued system for obtaining the truth of the road traffic. Therefore, according to the embodiment of the invention, the efficiency of determining the sensing performance of the sensor can be improved, and real-time detection of the sensing performance of the sensor is realized.

Description

Technical Field

[0001] The present disclosure relates to the field of sensor technology, and more particularly, to a method for determining perception performance, an intelligent driving method, and related devices. Background Art

[0002] Sensors play an important role in modern life. For example, traffic lights, signboards and other roadside equipment can be equipped with sensors such as cameras to collect road perception information, which can be used to monitor road conditions in real time or to adjust traffic control strategies. With the development of intelligent vehicles, road information collected by sensors installed on vehicles or roadside equipment is also used for intelligent driving of vehicles.

[0003] When using sensor perception information, it is often necessary to ensure that it is accurate enough. For example, for intelligent driving of cars, if the sensor's perception performance is poor and the collected perception information is inaccurate, it may cause the intelligent driving system to produce incorrect intelligent driving strategies, thereby affecting driving safety. In order to ensure that the perception information obtained by the sensor can accurately reflect the actual state of the road, the sensor's perception performance needs to be tested. Summary of the invention

[0004] The embodiments of the present disclosure propose a method for determining perception performance, an intelligent driving method, and related devices. In the embodiments of the present disclosure, environmental data that can indicate the environmental characteristics of the area of ​​interest of the sensor can be obtained. In order to determine the perception performance of the sensor, the embodiments of the present disclosure can obtain an environmental performance relationship set corresponding to the sensor, and the environmental performance relationship set includes at least one environmental performance relationship indicating the correspondence between environmental data and perception performance information. Then, the embodiments of the present disclosure can determine the perception performance information of the sensor in its area of ​​interest based on the environmental data and the environmental performance relationship set. In this way, after obtaining the environmental data of the area of ​​interest of the sensor, the perception performance of the sensor can be directly determined using the environmental data and the environmental performance relationship set, without having to rely on a truth system for detecting the actual situation of road traffic, thereby significantly improving the efficiency of determining the sensor's perception performance, and realizing all-weather detection of the sensor's perception performance, which is more practical.

[0005] In a first aspect of the present disclosure, a method for determining perceptual performance is provided. The method includes acquiring first environmental data, wherein the first environmental data indicates environmental characteristics of an area of ​​interest of a sensor at a first moment. The method also includes acquiring an environmental performance relationship set corresponding to the sensor, wherein the environmental performance relationship set includes at least one environmental performance relationship indicating a corresponding relationship between environmental data and perceptual performance information. In addition, the method also includes determining perceptual performance information of the sensor in the area of ​​interest at the first moment based on the first environmental data and the environmental performance relationship set.

[0006] In a second aspect of the present disclosure, a method for generating an environmental performance relationship is provided. The method includes acquiring environmental data, wherein the environmental data indicates environmental characteristics of an area of ​​interest of a sensor. The method also includes determining sensor perception performance information in the area of ​​interest. In addition, the method also includes storing the environmental data and the perception performance information in association as an environmental performance relationship in an environmental performance relationship set.

[0007] In a third aspect of the present disclosure, a device for determining perceptual performance is provided. The device includes a data acquisition module configured to acquire first environmental data, wherein the first environmental data indicates environmental characteristics of an area of ​​interest of the sensor at a first moment. The device also includes a relationship acquisition module configured to acquire an environmental performance relationship set corresponding to the sensor, wherein the environmental performance relationship set includes at least one environmental performance relationship indicating a corresponding relationship between environmental data and perceptual performance information. In addition, the device also includes a performance determination module configured to determine perceptual performance information of the sensor in the area of ​​interest at a first moment based on the first environmental data and the environmental performance relationship set.

[0008] In a fourth aspect of the present disclosure, a device for generating an environment-performance relationship is provided. The device includes an environment data acquisition module configured to acquire environment data, wherein the environment data indicates an environment characteristic of an area of ​​interest of a sensor. The device also includes a performance determination module configured to determine the sensor's perceived performance information in the area of ​​interest. In addition, the device also includes a storage module configured to store the environment data and the perceived performance information in association as an environment-performance relationship in an environment-performance relationship set.

[0009] In a fifth aspect of the present disclosure, an intelligent driving method is provided, the method comprising obtaining perceptual performance information of a sensor, wherein the perceptual performance information is determined according to the method provided in the first aspect of the present disclosure. The method further comprises determining the availability of at least one intelligent driving function associated with the sensor based on the perceptual performance information.

[0010] In a sixth aspect of the present disclosure, an electronic device is provided. The electronic device includes one or more processors; and a storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method provided according to the first aspect, the second aspect, or the fifth aspect of the present disclosure.

[0011] In a seventh aspect of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer-executable instructions, wherein the computer-executable instructions are executed by a processor to implement the method provided according to the first aspect, the second aspect, or the fifth aspect of the present disclosure.

[0012] It should be understood that the contents described in the summary of the invention are not intended to limit the key or important features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. In the accompanying drawings, the same or similar reference numerals represent the same or similar elements, wherein:

[0014] Figure 1 A schematic diagram illustrating an example environment in which various embodiments of the present disclosure may be implemented;

[0015] Figure 2A A schematic diagram illustrating yet another example environment in which various embodiments of the present disclosure may be implemented;

[0016] Figure 2B A schematic diagram showing a region of interest of a sensor according to some embodiments of the present disclosure;

[0017] Figure 3 A schematic diagram showing a method for determining perceived performance according to some embodiments of the present disclosure is shown;

[0018] Figure 4 A schematic diagram showing division of a region of interest of a sensor according to some embodiments of the present disclosure is shown;

[0019] Figure 5A A schematic diagram showing a region of interest of a sensor according to some embodiments of the present disclosure;

[0020] Figure 5B Another schematic diagram of dividing the region of interest of a sensor according to some embodiments of the present disclosure is shown;

[0021] Figure 6Another schematic diagram of a method for determining perceived performance according to some embodiments of the present disclosure is shown;

[0022] Figure 7 A schematic diagram showing a method for generating an environment-performance relationship according to some embodiments of the present disclosure;

[0023] Figure 8 Another schematic diagram showing a method for generating an environment-performance relationship according to some embodiments of the present disclosure;

[0024] Fig. 9 A schematic diagram of an intelligent driving method according to some embodiments of the present disclosure is shown;

[0025] Fig.10 A schematic block diagram of an apparatus for determining perceived performance according to some embodiments of the present disclosure is shown;

[0026] Fig.11 A schematic block diagram showing an apparatus for generating an environment-performance relationship according to some embodiments of the present disclosure; and

[0027] Fig.12 A schematic block diagram of a device that can implement multiple embodiments of the present disclosure is shown. DETAILED DESCRIPTION

[0028] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0029] In the description of the embodiments of the present disclosure, the term "including" and similar terms should be understood as open inclusion, that is, "including but not limited to". The term "based on" should be understood as "based at least in part on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The terms "first", "second", etc. may refer to different or the same objects. Other explicit and implicit definitions may also be included below.

[0030] In the embodiments of the present disclosure, sensors include but are not limited to radars, cameras, etc., which can sense road conditions and identify objects such as vehicles, pedestrians and obstacles on the road. The sensor can be located in a vehicle, or in a roadside device, edge device or other device. The sensor's perception of road conditions may include but is not limited to the position of people on the road, the position of the vehicle, the driving angle of the vehicle, the speed of the vehicle, the position of obstacles or the size of obstacles. In the embodiments of the present disclosure, the data used to carry the perception information collected by the sensor is referred to as identification data. In the embodiments of the present disclosure, the vehicle may include but is not limited to cars, trucks, electric vehicles, motorcycles, buses, ships, airplanes, drones, recreational vehicles, construction equipment, intelligent robots, etc., and the embodiments of the present disclosure are not particularly limited. For ease of explanation, in the embodiments of the present disclosure, the method provided by the present disclosure is described by taking a car as an example, and it should be understood that this cannot be a limitation on the embodiments of the present disclosure.

[0031] In the embodiments of the present disclosure, intelligent driving refers to the function that the vehicle assists the driver in controlling the vehicle through the configured sensors, controllers, actuators, communication modules and other devices, or even completely replaces the driver to achieve unmanned driving. Intelligent driving can include assisted driving and automatic driving. Among them, assisted driving can include, for example, adaptive cruise control, lane keeping, automatic emergency braking, forward collision warning, lane departure warning, attention detection, automatic parking, etc. Among them, automatic driving can include, for example, partial automatic driving, conditional automatic driving, fully automatic driving, etc.

[0032] As mentioned above, sensors play an important role in modern life. For example, taking the scenario of intelligent vehicle driving as an example, Figure 1 Schematic diagram of an example environment in which various embodiments of the present disclosure may be implemented. Figure 1 As shown, in the environment 100, the roadside equipment 101 (in Figure 1 The infrastructure (in which a traffic light is configured) is equipped with a sensor 102, and the environmental state of the road 103 can be sensed by the sensor 102. The roadside device 101 can communicate with the vehicle 104 on the road 103, so that the vehicle 104 can receive the target recognition data from the roadside device and realize the intelligent driving function. For example, there are vehicles 105 and 106 in front of the vehicle 104, and the roadside device 101 can identify the position and speed of the vehicle 105 and the vehicle 106 on the road 103 through the sensor 102, and send the identified position and speed to the vehicle 104 in real time, so that the vehicle 104 can adjust the intelligent driving strategy based on this.

[0033] The environment 100 may also include a detection device (in Figure 1(not shown), the detection device can detect the perception performance of the sensor of the roadside equipment, for example, whether the recognition result of the sensor 102 to the vehicle 103 is accurate. In the embodiments of the present disclosure, the detection device can be an independent device, such as an embedded device, a computer, a computer system or a server, etc. The detection device can also be a chip or component in the device, and the detection device can also be a software module. In some embodiments, the detection device can be configured in the roadside equipment 101, can communicate with the vehicle 104, and send the detection result to the roadside equipment vehicle 104. In some embodiments, the detection device can also be configured in the vehicle 104, can obtain the target recognition data from the sensor 102 to detect the perception performance of the sensor 102. In some embodiments, the detection device can also be configured in the cloud, can obtain the target recognition data of the sensor 102 and detect the perception performance of the sensor 102, and the detection device can send the perception performance of the sensor 102 to the vehicle 104.

[0034] It should be understood that Figure 1 This is only an example and not a limitation of the present disclosure. For example, the vehicle 104 in the figure may also be configured with a sensor. The vehicle 104 may identify the position and speed of the vehicle 105 and the vehicle 106 on the road 103 through the sensor configured by itself, and adjust the intelligent driving strategy based on this. The detection device may also detect the perception performance of the sensor in the vehicle 104.

[0035] The detection device can detect the sensor's perception performance through a ground truth (GT) system. The GT system is a system that obtains the real situation of the road for the purpose of detection. The GT system can detect the real state of the road and provide real identification data. By comparing the target identification data from the sensor configured on the roadside equipment or vehicle with the real identification data of the GT system, the accuracy of the sensor's perception of the environment can be determined, thereby determining the sensor's perception performance. The GT system can be implemented in the form of a GT car or a GT device on the roadside. However, the GT car has a limited time to obtain the real road conditions, and can only provide real identification data corresponding to the sensor in a fixed time period, so it is impossible to detect the sensor's perception performance at any time. If the GT device is deployed on the roadside, it will increase additional costs.

[0036] The inventors of the present disclosure have found that the sensor's perception performance (e.g., visibility) within its region of interest (ROI) is related to environmental features within the region of interest, including but not limited to weather type, visible distance, lighting conditions, etc. Exemplarily, rainy and snowy weather or high-brightness reflective areas within the region of interest will reduce the sensor's perception performance. Therefore, the sensor's perception performance can be determined by the environmental features of the sensor's region of interest. On this basis, an embodiment of the present disclosure proposes a method for determining perception performance. In an embodiment of the present disclosure, a detection device acquires environmental data and an environmental performance relationship set corresponding to the sensor, wherein the environmental data indicates the environmental features of the sensor's region of interest, and the environmental performance relationship set includes at least one environmental performance relationship indicating the corresponding relationship between the environmental data and the perception performance information. The detection device determines the sensor's perception performance information in its region of interest based on the environmental data and the environmental performance relationship set.

[0037] In this way, after obtaining the environmental data of the area of ​​interest of the sensor, the sensor's perception performance can be directly determined based on the environmental performance relationship set using the environmental data without relying on the GT system. Therefore, the cost of setting up the GT system can be reduced and the efficiency of detecting the sensor's perception performance can be significantly improved. In addition, all-weather detection of the sensor's perception performance can be achieved, which is more practical. The publicly provided method is described below in conjunction with the accompanying drawings.

[0038] Figure 2A 1 is a schematic diagram of an example environment 200A in which various embodiments of the present disclosure may be implemented. Figure 2A As shown, in environment 200A, there is a camera 201 (an example of a sensor) configured on a traffic light at an intersection, and a vehicle 203. The dashed area 202 in the intersection is an area of ​​interest in the field of view of the camera 201, and the camera 201 can sense objects such as people, vehicles or obstacles in the area of ​​interest 202. Exemplarily, Figure 2B is an example of the field of view 200B of the camera 201. The car 203 is located in the area of ​​interest 202, and the camera 201 can identify the position and speed of the car 203. When the environmental characteristics of the area of ​​interest 202 of the camera 201 change, the perception performance of the camera 201 will be affected, for example, the recognition accuracy of the position or speed of the car 203 will change.

[0039] In environment 200A, a detection device located in a roadside device, a vehicle, an edge device or a cloud can obtain environmental data indicating environmental features within an area of ​​interest 202 of camera 201. The environmental data can be obtained by analyzing and identifying raw data collected by camera 201 by the environmental detection device, or collected by the environmental detection device through an independently set sensor. The detection device can obtain an environmental performance relationship set corresponding to camera 201. The environmental performance set corresponding to camera 201 can include multiple environmental performance relationships, each of which can indicate perceptual performance information corresponding to a type of environmental data. These environmental performance relationships can be pre-set, or collected by detecting the perceptual performance of camera 201 when the environmental features are different. The detection device can determine the perceptual performance information corresponding to the environmental data based on the environmental data and the environmental performance relationship set, and then determine the perceptual performance of camera 201 in the area of ​​interest 202.

[0040] It should be understood that Figure 2A and Figure 2B The environment shown is only an example given for the convenience of explanation and cannot be a limitation of the present disclosure. The method provided in the embodiments of the present disclosure can also be applied to other types of sensors, and the sensors can be configured on different roads or on various vehicles. The examples of regions of interest in the drawings of the embodiments of the present disclosure cannot be a limitation of the present disclosure. In some embodiments of the present disclosure, the region of interest of the sensor can be a part of the perception range of the sensor that is preset, such as a predefined box area in the field of view of a camera. In some embodiments, the region of interest can also be the entire perception range of the sensor, for example, it can be the entire area of ​​the field of view of a camera.

[0041] Figure 3 1 shows a flow chart of a method 300 for determining perceived performance according to some embodiments of the present disclosure. The method 300 may be performed by the aforementioned detection device, which may be implemented in software and / or hardware. Figure 3 As shown, method 300 may include blocks 302 to 306 .

[0042] In box 302, the detection device obtains first environmental data, wherein the first environmental data indicates environmental characteristics of the area of ​​interest of the sensor at a first moment. In some examples, the first environmental data may indicate that the area of ​​interest of the sensor is in a light environment, and indicate the average illumination of the area of ​​interest. In other examples, the first environmental data may indicate that the area of ​​interest of the sensor is in a foggy environment, and indicate the average visibility of the area of ​​interest. When the detection device needs to detect the perception performance of the sensor at the first moment, the detection device obtains the first environmental data. In some embodiments, the first environmental data may be data that can indicate environmental characteristics and is obtained by processing the raw data collected by the sensor at the first moment. Exemplarily, the detection device may obtain and process the raw data collected by the sensor, such as Figure 2A The detection device can perform feature recognition on the raw data of the image captured by the camera 201 in the environment 200A shown at the first moment, thereby determining the environmental features in the area of ​​interest and obtaining the first environmental data. In some embodiments, the first environmental data is collected by other environmental detection devices at the first moment. In some embodiments, the first environmental data can be obtained by the detection device through the Internet.

[0043] In box 304, the detection device obtains an environmental performance relationship set corresponding to the sensor, wherein the environmental performance relationship set includes at least one environmental performance relationship indicating the corresponding relationship between environmental data and perceptual performance information. In an embodiment of the present disclosure, the perceptual performance information may include an evaluation result of the perceptual performance of the sensor, including but not limited to the recognition accuracy of the sensor (for example, the detection accuracy of the position, the detection accuracy of the speed, etc.) or the recall rate of the sensor (also known as the recall rate, which indicates the detection coverage of the sensor for all targets to be detected). In some embodiments, the perceptual performance information may also include perceptual performance levels divided based on the recognition accuracy or recall rate of the sensor. The environmental performance relationship set may include multiple pre-established environmental performance relationships, each of which may indicate perceptual performance information corresponding to a type of environmental data. These environmental performance relationships may be pre-stored in the memory of the detection device, or may be collected by detecting the perceptual performance of the sensor when the environmental characteristics are different. The method for obtaining the environmental performance relationship may also refer to Figure 7 The environment-performance relationship set may be in the form of a list, and for example, may be as shown in Table 1.

[0044] Table 1: Example of an environment-performance relationship set

[0045] Environmental data Perceived performance information 1 Foggy environment, visibility 1-5 meters Level 1 2 Foggy environment, visibility 5-10 meters Level 2 3 Foggy environment, visibility 10-50 meters Level 3 4 Lighting environment, illumination 100-500 lux Level 3 5 Lighting environment, illumination 500-1000 lux Level 2 … … …

[0046] It should be understood that Table 1 is only an example of the present disclosure and cannot be a limitation of the present disclosure. The environmental performance relationship set can also be presented in other forms, the environmental data can also be presented in other forms, and the perception performance information can also be presented in other forms, such as in the form of specific numerical values ​​of recognition accuracy or recall rate.

[0047] In box 306, the detection device determines the perceptual performance information of the sensor in the area of ​​interest at the first moment based on the first environmental data and the environmental performance relationship set. By matching the first environmental data with the environmental performance relationship in the environmental performance relationship set, the detection device can determine the perceptual performance information of the sensor in the area of ​​interest. According to method 300 of an embodiment of the present disclosure, by acquiring the environmental data of the area of ​​interest of the sensor and the environmental performance relationship set, the perceptual performance of the sensor can be directly determined without relying on the GT system, which can reduce costs and increase practicality. Since the environmental performance relationship set of the sensor can be stored locally and available at any time, it is possible to achieve instant detection of the sensor's perceptual performance, which is more flexible.

[0048] It should be understood that although Figure 3 302 is shown before block 304, but it is not intended to limit the order of the operations performed at blocks 302 and 304. On the contrary, the operations performed at blocks 302 and 304 can be performed in reverse order or simultaneously.

[0049] In some embodiments, the detection device may divide the area of ​​interest of the sensor into multiple areas, and determine the perception performance information for each of the multiple areas. Figure 2A Taking the camera 201 in FIG. 1 as an example, the detection device may divide the region of interest of the camera 201 into three regions, for example, Figure 4 shown. Figure 4 is an example of the field of view of camera 201. Figure 4 In the example, the region of interest 401 of the camera 201 is divided into sub-region of interest 402, sub-region of interest 403 and sub-region of interest 404 by the detection device. The sensor's strategy for dividing the region of interest may be predefined. For example, the strategy may be based on the straight-line distance from the sensor, where the region less than 5 meters is the sub-region of interest 402, the region greater than 5 meters and less than 20 meters is the sub-region of interest 403, and the region greater than 20 meters is the sub-region of interest 404. It should be understood that Figure 4 The region division shown is only an example of the present disclosure and cannot be a limitation of the present disclosure. It should also be understood that the detection device can obtain data from the sensor to determine the region of interest of the sensor and perform region division based on this.

[0050] Determining the sensor's perception performance information by region can more finely divide the sensor's perception performance. For example, in a foggy environment, in the sub-region of interest 404, the sensor's visibility is low, and in the sub-region of interest 402, the sensor's visibility is high. The detection device can determine that the perception performance level of the sub-region of interest 404 is low, the recognition accuracy of the object position is low, and the error is large based on the perception performance information of the sub-region of interest 404; the detection device can determine that the perception performance level of the sub-region of interest 402 is high, the recognition accuracy of the object position is high, and the error is small based on the perception performance information of the sub-region of interest 402. When the vehicle performs intelligent driving based on the target recognition data of the sensor of the roadside equipment, the target recognition data in the sub-region of interest 402 can be adopted, and the target recognition data in the sub-region of interest 404 can be discarded. In this way, the situation where the target recognition data of the entire sensor is discarded due to the poor local environmental characteristics of the sensor's region of interest can be avoided. Finely dividing the sensor's perception performance can make the application of the sensor more flexible.

[0051] In some embodiments, the first environmental data may include multiple sub-data, each of which corresponds to environmental features at different locations in the area of ​​interest of the sensor. Figure 5A In the first environmental data, the region of interest 501 of the sensor may be presented in the form of a grid (or a grid), and the first environmental data may include the location of each grid region and the environmental score value corresponding to each grid region. The environmental score value may be generated based on the raw data collected by the sensor, for example, by identifying the raw data, determining the illumination of each part in the region of interest, and generating the illumination value corresponding to the location of each part as the environmental score value. After the detection device divides the sensor into regions, based on the multiple sub-data included in the first environmental data, the average environmental score value corresponding to each divided region may be determined, and then the perception performance information of each divided region may be determined based on the environmental performance relationship set.

[0052] In some embodiments, the detection device divides the region of interest of the sensor based on the acquired environmental data. For example, a plurality of region division strategies may be stored in a memory configured in the detection device. Based on the acquired first environmental data, the detection device may determine the corresponding region division strategy, and divide the region of interest on this basis. Exemplarily, when the first environmental data indicates that the region of interest of the sensor is in a foggy environment, the detection device may determine the region division strategy to be based on the straight-line distance to the sensor. When the first environmental data indicates that the region of interest of the sensor is in a light environment, the detection device may determine the region division strategy to be based on the illumination. The first environmental data may include multiple sub-data indicating environmental score values ​​of different parts of the region of interest. After determining the region division strategy, the detection device may divide the region of interest into multiple regions based on the environmental score values ​​of different parts of the region of interest indicated by the first environmental data. Exemplarily, referring to Figure 5B Based on the multiple sub-data in the first environment data, the detection device can divide the region of interest 501 of the sensor into sub-region of interest 501, sub-region of interest 503 and sub-region of interest 504. For example, the division strategy of the region of interest is based on the illumination of the region. Sub-regions of interest 502 and 503 correspond to sub-data in the first environment data with an illumination greater than or equal to 1000 lux, and sub-region of interest 504 corresponds to sub-data in the first environment data with an illumination less than 1000 lux. It should be understood that the boundaries of the divided regions may be irregular.

[0053] In some embodiments, the detection device may determine the working boundary state of the sensor based on the sensor's perception performance information in the region of interest. The working boundary state indicates that the sensor's perception performance in the region of interest is within or outside the working boundary. The working boundary is used to indicate the conditions that need to be met to enable a specific function (such as automatic parking, automatic driving, etc.) that requires the use of the sensor's identification data. When the perception performance of the sensor's region of interest is within the working boundary, it means that the perception performance of the sensor's region of interest can meet the requirements for enabling the specific function. When the perception performance of the sensor's region of interest is outside the working boundary, it means that the perception performance of the sensor's region of interest cannot meet the requirements for enabling the specific function. In some embodiments, the working boundary may correspond to the vehicle's operational design domain (ODD). ODD is a condition for enabling a specific automatic driving function predefined in the vehicle. The enabling of the vehicle's automatic driving function requires that the sensor's perception performance meet the ODD conditions.

[0054] The detection device may obtain a perception performance threshold for determining the working boundary state based on the perception performance information. For example, the perception performance information includes a perception performance level, and the perception performance threshold may be a perception performance level threshold. The detection device may determine the working boundary state accordingly. For example, when the perception performance level of the sensor in the region of interest is greater than or equal to the perception performance level threshold, the working boundary state of the sensor in the region of interest may be determined to be within the working boundary. When the perception performance level of the sensor in the region of interest is less than the perception performance level threshold, the working boundary state of the sensor in the region of interest may be determined to be outside the working boundary. In some embodiments, a sensor may correspond to multiple types of perception performance, such as the recognition accuracy of the position of an object and the detection accuracy of the type of object. The detection device may determine the perception performance information corresponding to different types of perception performance of a sensor, and determine the working boundary state on this basis.

[0055] In some embodiments, different intelligent driving functions of a vehicle may have different working boundaries, that is, the conditions that need to be met for the vehicle to activate different intelligent driving functions may be different. Exemplarily, the working boundary of lane keeping can include the sensor's recognition accuracy of the lane line type and position, while the working boundary of forward collision warning can include the sensor's recognition accuracy of the obstacle position. The detection device can obtain the working boundaries corresponding to different intelligent driving functions, and obtain the perception performance thresholds corresponding to the different working boundaries, so as to determine the working boundary states corresponding to the sensor for different intelligent driving functions. In some embodiments, for the same perception performance of the same sensor, different intelligent driving functions may correspond to different perception performance thresholds, that is, for the same sensor, it may be within the working boundary of one intelligent driving function and outside the working boundary of another intelligent driving function.

[0056] In some embodiments, the detection device may divide the region of interest of the sensor into a plurality of sub-regions of interest, and after determining the sensing performance information of each sub-region of interest, determine the working boundary state of each sub-region of interest. That is, for a sensor, different sub-regions of interest may have different working boundary states.

[0057] Next, combine Figure 6 , further describing the method provided by the embodiment of the present disclosure from the perspective of module interaction. Figure 6 FIG. 6 is a flowchart of a method 600 for determining perceived performance provided by an embodiment of the present disclosure. Figure 6The detection device 601 includes a detection device 601 and an environmental data determination module 602. The detection device 601 can determine the perception performance of the sensor at the first moment. The sensor can be, for example, the sensor 102 in Figure 100 or the camera 201 in Figure 200A. In some embodiments, the detection device 601 can be an ODD detection module configured in the vehicle, which is used to detect whether the operating conditions of the vehicle meet the conditions for starting the automatic driving function. The environmental data determination module 602 can obtain the raw data from the sensor at the first moment, and identify and process the data to generate the first environmental data. The detection device 601 obtains the first environmental data corresponding to the sensor from the environmental data determination module 602, and the first environmental data can indicate the environmental characteristics of the sensor in the area of ​​interest at the first moment. In some embodiments, the environmental data determination module 602 can be a device independent of the detection device 601. In some embodiments, the environmental data determination module 602 can be configured within the detection device 601.

[0058] The detection device 601 may include a strategy determination module 611, a region division module 612, a performance determination module 613, and a boundary state determination module 614. The strategy determination module 611 may determine the region division strategy based on the first environment data obtained from the environment data determination module 602, for example, obtaining the region division strategy matching the first environment data from a predefined database. In some embodiments, the strategy determination module 611 may generate the region division strategy based on the first environment data through a predefined algorithm. The region division module 612 may obtain the region division strategy from the strategy determination module 611, and divide the region of interest of the sensor into a plurality of sub-regions of interest based on the region division strategy and the first environment data obtained from the environment data determination module 602. The performance determination module 613 determines the perceptual performance information of each sub-region of interest based on the plurality of sub-regions of interest divided by the region division module 612 and the first environment data. Among them, each sub-area of ​​interest can respectively correspond to one or more sub-data in the first environmental data. When determining the perceived performance information, the performance determination module 613 can obtain the environmental performance relationship set 615, and based on the matching of the sub-data with the environmental performance relationship in the environmental performance relationship set 615, determine the perceived performance information of the sensor in the sub-area of ​​interest at the first moment.

[0059] The boundary state determination module 614 can obtain the boundary state determination strategy 616, which includes the perception performance threshold. The boundary state determination module 614 can generate the boundary state determination result of the sensor based on the boundary state determination strategy 616 and the perception performance information determined by the performance determination module 613. The boundary state determination result can indicate the working boundary state of the sensor in the sub-region of interest at the first moment. For example, the boundary state determination result 617 can correspond to the sub-region of interest within the working boundary, and the boundary state determination result 618 can correspond to the sub-region of interest outside the working boundary. The region of interest of a sensor can be divided into multiple sub-regions of interest, and each sub-region of interest can have a different working boundary state. Therefore, the detection device can generate the boundary state determination result 617 and the boundary state determination result 618 at the same time.

[0060] It should be understood that Figure 6 The process shown is only an example of an embodiment of the present disclosure and is not a limitation of the present disclosure. For example, in some embodiments, the detection device 601 does not include a strategy determination module 611 and an area division module 612. That is, the performance determination module 613 in the detection device 601 can directly determine the sensor's perception performance information in the area of ​​interest based on the environmental data and the environmental performance relationship set 615.

[0061] Combined with the above Figures 2A to 6 The method for determining the perception performance provided by the embodiment of the present disclosure is described. In the method for determining the perception performance provided by the embodiment of the present disclosure, the detection device needs to determine the perception performance of the sensor based on the environment performance relationship set. Figure 7 and Figure 8 The method for generating an environmental performance relationship provided in an embodiment of the present disclosure is described. Through this method, an environmental performance relationship in an environmental set can be generated for use by a detection device when determining the perception performance of a sensor.

[0062] Figure 7 A flow chart of a method 700 for generating an environment-performance relationship according to some embodiments of the present disclosure is shown. Method 700 may be executed by an environment-performance relationship generating device. In some embodiments, the environment-performance relationship generating device may be configured in the same device as the aforementioned detection device. In some embodiments, the environment-performance relationship generating device may be the aforementioned detection device. In some embodiments, the environment-performance relationship generating device may be configured in the same roadside device or vehicle as the sensor. The environment-performance relationship generating device may be implemented in software and / or hardware. For example, Figure 7 As shown, method 700 may include blocks 702 to 706 .

[0063] In block 702, the environment-performance relationship generating device obtains environment data, wherein the environment data indicates the environment characteristics of the region of interest of the sensor, for example, it may indicate that the region of interest of the sensor is in a lighting environment and indicates the average illumination of the region of interest. In some embodiments, the environment data may come from a sensor to be detected, in some embodiments, the environment data may come from an independently set device for detecting the environment, and in some embodiments, the environment data may be obtained by the environment-performance relationship generating device from a network. The method in block 702 may also be performed with reference to the aforementioned block 302.

[0064] In block 704, the environment-performance relationship generating device determines the sensor's perceived performance information in the region of interest. In some embodiments, the perceived performance information may include the sensor's recognition accuracy, such as the position detection accuracy, the speed detection accuracy, or the type recognition accuracy, etc. In some embodiments, the perceived performance information may include the sensor's recall rate. The environment-performance relationship generating device may determine the sensor's perceived performance information in the region of interest based on the sensor's perceived information of the object in the region of interest.

[0065] In some embodiments, the environment-performance relationship generating device may obtain target recognition data of the sensor in the region of interest and real recognition data of the GT system in the region of interest, and determine the sensor's perception performance information in the region of interest based on the target recognition data and the real recognition data. The target recognition data from the sensor reflects the sensor's perception of road traffic, and the real recognition data from the GT system reflects the real situation of road traffic. By comparing the target recognition data with the real recognition data, the sensor's perception performance in the region of interest can be determined.

[0066] In some embodiments, the perception performance information may include the recognition accuracy of the sensor. Based on the target recognition data and the real recognition data, the recognition accuracy of the sensor in the area of ​​interest can be determined. In some embodiments, the recognition accuracy can be the detection accuracy of the position, the target recognition data indicates the position of the object recognized by the sensor, the real recognition data indicates the position of the object recognized by the GT system, and the absolute value of the difference between the position indicated by the target recognition data and the position indicated by the real recognition data is the detection accuracy of the sensor for the position. In some embodiments, the recognition accuracy can be the detection accuracy of the speed, the target recognition data indicates the speed of the object recognized by the sensor, the real recognition data indicates the speed of the object recognized by the GT system, and the absolute value of the difference between the speed indicated by the target recognition data and the speed indicated by the real recognition data is the detection accuracy of the sensor for the speed. It should be understood that the description of the recognition accuracy in the present disclosure is only an example, and the present disclosure includes but is not limited to this.

[0067] In some embodiments, the perceptual performance information includes a perceptual performance level. The environmental performance relationship generation device can determine the recognition accuracy or recall rate of the sensor in the area of ​​interest based on the target recognition data and the real recognition data, and obtain the corresponding level determination strategy, and then determine the perceptual performance level of the sensor in the area of ​​interest based on the recognition accuracy or recall rate, and the level determination strategy. Taking recognition accuracy as an example, the level determination strategy can be the correspondence between recognition accuracy and perceptual performance level. The correspondence between recognition accuracy and perceptual performance level can be predefined, or obtained by the environmental performance relationship generation device from a standard website, for example, it can be from an international standard, a national standard or an industry standard. Exemplarily, the correspondence between recognition accuracy and perceptual performance level can be as shown in Table 2.

[0068] Table 2: Correspondence between recognition accuracy and perceptual performance level

[0069]

[0070] Table 2 takes the recognition accuracy as the detection accuracy of the position of different objects as an example to illustrate a possible correspondence between recognition accuracy and perception performance level. As shown in Table 2, for different perception targets, there may be different correspondences between recognition accuracy and perception performance level. It should be understood that Table 2 is only an example and cannot be a limitation to the present disclosure. The correspondence between recognition accuracy and perception performance level may also be other correspondences. In some embodiments, recognition accuracy may include but is not limited to detection accuracy of speed, detection accuracy of size, recognition accuracy of type, or detection accuracy of heading angle. It should be understood that the description of perception performance level and perception performance information in the present disclosure is only an example, not a limitation to the present disclosure. For example, perception performance information may also include the recall rate of the sensor.

[0071] In box 706, the environmental performance relationship generation device stores the environmental data and the perceived performance information in association with each other as an environmental performance relationship in the environmental performance relationship set. Through the methods in boxes 704 to 708, the environmental performance relationship generation device determines the perceived performance information of the sensor in the area of ​​interest. In the aforementioned box 702, the environmental performance relationship generation device obtains the environmental data. The environmental performance relationship generation device can store the environmental data in association with the perceived performance information, thereby generating an environmental performance relationship in the environmental performance relationship set. By repeatedly executing method 700 more than once, multiple environmental performance relationships can be generated, thereby constructing a corresponding relationship between various environmental features and perceived performance information. Exemplarily, the generated environmental performance relationship can be as shown in the aforementioned Table 1.

[0072] It should be understood that although Figure 7The order of box 702 and box 706 is shown in FIG, but it is not intended to limit the order of the operations performed at box 702 and box 706. On the contrary, the operations performed at box 702 and box 706 can be performed in a reversed order or simultaneously. Based on the above method, the environment-performance relationship generation device can determine the sensor's perceived performance information based on the real recognition data of the GT system and generate an environment-performance relationship. By executing method 700 in multiple environments or at multiple times, the environment-performance relationship generation device can generate multiple environment-performance relationships, thereby forming an environment-performance relationship set.

[0073] In some embodiments, the environmental data acquired at one time may include multiple sub-data, and each sub-data may correspond to the environmental features of different parts of the sensor's area of ​​interest. The acquired target recognition data and real recognition data may also include recognition results for objects in different parts of the area of ​​interest. The environmental performance relationship generation device may divide the sensor's area of ​​interest into multiple sub-areas of interest based on the environmental data, and determine the perceptual performance information of the sub-areas of interest based on the target recognition data and the real recognition data, thereby generating multiple environmental performance correspondences based on the multiple sub-data in the environmental data and the perceptual performance information of the multiple sub-areas of interest. In this way, at the same time, the environmental data is acquired once, and the environmental performance relationship generation device can generate multiple environmental performance correspondences, which can improve the efficiency of generating environmental correspondences.

[0074] In some embodiments, the environment-performance relationship generating device may also determine the working boundary state of the sensor in the area of ​​interest based on the determined perception performance information and the perception performance threshold. In some embodiments, the environment-performance relationship generated by the environment-performance relationship generating device includes the working boundary state corresponding to the environmental data. In this way, when the detection device determines the perception performance of the sensor in the area of ​​interest based on the environment-performance relationship through method 300 or method 600, the working boundary state can be directly determined based on the environment-performance relationship, without comparing the perception performance information with the threshold after determining it, thereby improving efficiency.

[0075] Figure 8 The method for generating an environment-performance relationship provided by the embodiment of the present disclosure is shown from the perspective of module interaction. Figure 8 In the method 800 shown, the environment-performance relationship generating device 801 can obtain environment data from the environment data determination module 802 , obtain target recognition data from the sensor 803 , and obtain real recognition data from the GT system 804 , thereby determining the environment-performance relationship of the sensor 803 .

[0076] The environment-performance relationship generating device 801 may include a strategy determining module 811, a region dividing module 812, a performance determining module 813, a relationship generating module 814, and a boundary state determining module 815. Figure 6 According to the method performed by the strategy determination module 611 and the area division module 612 in the environment-performance relationship generation device 801, the strategy determination module 811 and the area division module 812 can divide the area of ​​interest of the sensor 803 into multiple sub-areas of interest. The performance determination module 813 can determine the perception performance level of the sensor 803 in multiple sub-areas of interest based on the target recognition data obtained from the sensor 803, the real recognition data obtained from the GT system 804, and the level determination strategy 816. Based on the perception performance level of the sub-area of ​​interest determined by the performance determination module 813 and the sub-data corresponding to the sub-area of ​​interest in the environmental data obtained from the environmental data determination module 802, the relationship generation module 814 can generate an environment-performance relationship and store it in the memory 817. The memory 817 can be a memory configured by the environment-performance relationship generation device 801, or an independently configured memory, or a cloud memory, or a memory of the aforementioned detection device. The boundary state determination module 815 can obtain the boundary state determination strategy 818 and the environmental performance relationship generated by the relationship generation module 814, and determine the working boundary state of each sub-area of ​​interest of the sensor 803, and generate boundary state determination results 819 and boundary state determination results 820. The specific method can refer to the method executed by the boundary state determination module 614 in the aforementioned detection device 601.

[0077] It should be understood that Figure 8 The process shown is only an example of an embodiment of the present disclosure, and is not intended to limit the present disclosure. In some embodiments, the relationship generation module 814 may obtain the working boundary state of each sub-region of interest from the boundary state determination module 815, and generate an environment-performance relationship including the working boundary state. In some embodiments, the boundary state determination module 815 may generate an environment-performance relationship indicating the corresponding relationship between the environment data and the working boundary state based on the sub-data corresponding to the sub-region of interest in the environment data and the working boundary state of the sub-region of interest, and store it in the memory 817.

[0078] In the embodiments of the present disclosure, when there is real identification data of the GT system, the environment-performance relationship generation device can execute method 700 or method 800 to generate an environment-performance relationship. When there is no real identification data of the GT system, the detection device can execute method 300 or method 600 to detect the sensor's perception performance in the area of ​​interest based on the generated environment-performance relationship. In addition, while the environment-performance relationship generation device executes method 700 or method 800, the detection device can also detect the sensor's perception performance in the area of ​​interest through method 300 or method 600 based on the existing environment-performance relationship. In this way, the detection device can realize real-time detection of the sensor's perception performance. Compared with the detection method that relies on the GT system, the method provided by the embodiments of the present disclosure is more flexible.

[0079] The above content describes the method for determining the perception performance and the method for generating the environment-performance relationship provided in the embodiments of the present disclosure. The embodiments of the present disclosure also provide an intelligent driving method, which can use the perception performance information of the sensor determined in the aforementioned method to determine the availability of the intelligent driving function of the vehicle. Fig. 9 , the intelligent driving method 900 provided in an embodiment of the present disclosure is described. Fig. 9 The method shown may be executed by a controller, which may be configured in the cloud or in the vehicle.

[0080] refer to Fig. 9 , method 900 may include box 902 and box 904. In box 902, the controller obtains the perception performance information of the sensor, for example, the perception performance information may be obtained from a detection device configured in a roadside device, a vehicle, an edge device or a cloud, and the perception performance information may be determined by the detection device through the aforementioned method 300 or method 600, and may include an evaluation result of the perception performance of the sensor, for example, may include the recognition accuracy of the position or speed.

[0081] In box 904, the controller determines the availability of at least one intelligent driving function associated with the sensor based on the perception performance information. The vehicle can be configured with a variety of intelligent driving functions, such as lane keeping function, automatic parking function, etc. These intelligent driving functions can be associated with the same or different sensors. For example, the implementation of the lane keeping function needs to rely on the cameras configured in front of the vehicle and on the left and right sides of the vehicle. In addition to relying on the cameras in front of the vehicle and on the left and right sides of the vehicle, the implementation of the automatic parking function also needs to rely on the camera configured at the rear of the vehicle and the radar of the vehicle. The controller can determine whether the intelligent driving function associated with the sensor is available based on the perception performance information. Exemplarily, it can be determined whether the automatic parking function of the vehicle is available based on the perception performance information of the camera configured at the rear of the vehicle. In some embodiments, after the vehicle receives the start instruction of the intelligent driving function, the controller can obtain the perception performance information of the sensor associated with the intelligent driving function, and determine whether the intelligent driving function is available on this basis. In some embodiments, the controller can issue a fault indication when it is determined that the intelligent driving function is not available to indicate that the intelligent driving function is not available. In some embodiments, the controller can turn on the intelligent driving function when it is determined that the intelligent driving function is available.

[0082] In some embodiments, the controller may determine the working boundary state of the sensor in the area of ​​interest based on the sensor's perception performance information. The controller may turn on, keep running, not turn on or exit the vehicle's intelligent driving function based on the working boundary state. The method for the controller to determine the working boundary state based on the perception performance information may be performed with reference to the aforementioned method 300, which will not be repeated here. In some embodiments, a sensor may correspond to a variety of intelligent driving functions, and different intelligent driving functions may rely on different perception performances of the same sensor, that is, a sensor may have multiple types of perception performance information. The controller may determine the working boundary states corresponding to the sensor and the various intelligent driving functions based on these perception performance information, and control the intelligent driving function to be turned on, kept running, not turned on or exited on this basis. In some embodiments, when the perception performance of all sensors associated with an intelligent driving function is within the working boundary of the intelligent driving function, the controller may control the intelligent driving function to be turned on or kept running. In some embodiments, when receiving an instruction to turn on the intelligent driving function, if the controller determines that the perception performance of one or more sensors associated with the intelligent driving function is outside the working boundary, the controller may control the vehicle not to turn on the intelligent driving function. In some embodiments, when the intelligent driving function of the vehicle is turned on, if the controller determines that the perception performance of one or more sensors associated with the intelligent driving function is outside the working boundary, the controller can control the vehicle to exit the intelligent driving function.

[0083] Through method 900, the controller can determine whether the intelligent driving function that needs to use the sensor's perception information is available based on the sensor's perception performance information. The controller can also control the intelligent driving function to start, keep running, not start, or exit based on the sensor's perception performance. For example, if the sensor's perception performance is insufficient, the controller controls the vehicle to exit or not start the intelligent driving function. In this way, it can ensure that the vehicle's intelligent driving function is performed on the basis of reliable perception information, and can ensure the safety of the vehicle's intelligent driving function.

[0084] It should be noted that the methods provided by the present disclosure are described in the above embodiments with a detection device, an environment-performance relationship generating device, and a controller, respectively. This is merely an example given for the sake of convenience of explanation, and is not a limitation of the present disclosure. For example, in some embodiments, the detection device and the environment-performance relationship generating device are the same device, that is, the detection device can generate an environment-performance relationship through method 700 or method 800, and then the sensor's perception performance in the area of ​​interest can be determined through method 300 or method 600. In some embodiments, the detection device can be the same device as the controller, that is, the controller can determine the sensor's perception performance in the area of ​​interest through method 300 or method 600, and then the intelligent driving function of the vehicle can be controlled through method 900.

[0085] Fig.10 FIG. 1 is a schematic block diagram of an apparatus 1000 for determining perceived performance according to some embodiments of the present disclosure. The apparatus 1000 may correspond to the detection apparatus in the aforementioned method embodiment, for example. Fig.10 As shown, the device 1000 may include a data acquisition module 1010, configured to acquire first environmental data, wherein the first environmental data indicates environmental characteristics of the sensor's area of ​​interest at a first moment. The device 1000 may also include a relationship acquisition module 1020, configured to acquire an environmental performance relationship set corresponding to the sensor, wherein the environmental performance relationship set includes at least one environmental performance relationship indicating a corresponding relationship between environmental data and perceived performance information. In addition, the device 1000 may also include a performance determination module 1030, configured to determine the perceived performance information of the sensor in the area of ​​interest at a first moment based on the first environmental data and the environmental performance relationship set. It can be understood that by using the device 1000 of the present disclosure, at least one of the many advantages that can be achieved by the method or process described above can be achieved.

[0086] In some embodiments, the apparatus 1000 further comprises a first region division module configured to determine a first sub-region of interest in the region of interest based on the first environmental data, wherein the first sub-region of interest corresponds to the first sub-data in the first environmental data. And wherein the performance determination module 1030 comprises a first sub-region performance determination unit configured to determine, at a first moment, the sensor's perceived performance information in the first sub-region of interest based on the first sub-data and the environmental performance relationship set.

[0087] In some embodiments, the first area division module includes: a first environment type determination unit, configured to determine a first environment type corresponding to the first environment data; a first strategy determination unit, configured to determine a first area division strategy based on the first environment type; and a first area division unit, configured to determine a first sub-area of ​​interest based on the first area division strategy and the first environment data.

[0088] In some embodiments, the device 1000 also includes: a threshold acquisition module, configured to acquire a first threshold; and a boundary state determination module, configured to determine the working boundary state of the sensor in the first sub-region of interest at the first moment based on the perception performance information of the first sub-region of interest and the first threshold.

[0089] In some embodiments, the device 1000 also includes: a second data acquisition module, configured to acquire second environmental data, wherein the second environmental data indicates a second environmental feature of the area of ​​interest of the sensor at a second moment, wherein the second moment is before the first moment; an identification data acquisition module, configured to acquire target identification data of the sensor in the area of ​​interest; a real data acquisition module, configured to acquire real identification data of a truth system in the area of ​​interest, wherein the truth system is used to acquire real conditions of road traffic; a second performance determination module, configured to determine second perceptual performance information of the sensor in the area of ​​interest at the second moment based on the target identification data and the real identification data; and a storage module, configured to store the second environmental data and the second perceptual performance information in association with each other as an environmental performance relationship in an environmental performance relationship set.

[0090] In some embodiments, the device 1000 also includes a second area division module, which is configured to determine a second sub-area of ​​interest in the area of ​​interest based on the second environmental data, wherein the second sub-area of ​​interest corresponds to the second sub-data in the second environmental data; wherein the second performance determination module includes a second sub-area performance determination unit, which is configured to determine the perceived performance information of the sensor in the second sub-area of ​​interest at the second moment based on the target recognition data and the real recognition data; and wherein the storage module includes a second storage unit, which is configured to store the second sub-data and the perceived performance information of the sensor in the second sub-area of ​​interest at the second moment in association as an environmental performance relationship in the environmental performance relationship set.

[0091] In some embodiments, the second area division module includes: a second environment type determination unit, configured to determine the second environment type corresponding to the second environment data; a second strategy determination unit, configured to determine the second area division strategy based on the second environment type; and a second area division unit, configured to determine the second sub-area of ​​interest based on the second area division strategy and the second environment data.

[0092] In some embodiments, wherein the second perceptual performance information includes a perceptual performance level, the second performance determination module further includes: a strategy acquisition unit configured to acquire a level determination strategy; an accuracy determination unit configured to determine the recognition accuracy of the sensor in the area of ​​interest based on the target recognition data and the real recognition data; and a level determination unit configured to determine the perceptual performance level of the sensor in the area of ​​interest based on the recognition accuracy and the level determination strategy.

[0093] Fig.11 FIG. 1 is a schematic block diagram of an apparatus 1100 for generating an environment-performance relationship according to some embodiments of the present disclosure. The apparatus 1100 may correspond to the environment-performance relationship generating apparatus in the aforementioned method embodiment, for example. Fig.11 As shown, the device 1100 may include an environmental data acquisition module 1110, which is configured to acquire environmental data, wherein the environmental data indicates environmental characteristics of the area of ​​interest of the sensor. The device 1100 may also include a performance determination module 1120, which is configured to determine the sensor's perceived performance information in the area of ​​interest. In addition, the device 1100 may also include a storage module 1130, which is configured to store the environmental data and the perceived performance information in association as an environmental performance relationship in an environmental performance relationship set.

[0094] In some embodiments, the performance determination module 1120 includes: an identification data acquisition unit, configured to acquire target identification data of the sensor in the area of ​​interest; a real data acquisition unit, configured to acquire real identification data of the truth system in the area of ​​interest, wherein the truth system is used to acquire the real situation of road traffic; and a performance determination unit, configured to determine the perception performance information based on the target identification data and the real identification data.

[0095] In some embodiments, the device 1100 also includes a region division module, which is configured to determine a sub-region of interest in the region of interest based on the environmental data, wherein the sub-region of interest corresponds to the sub-data in the environmental data; wherein the performance determination unit includes a sub-region performance determination unit, which is configured to determine the perceived performance information of the sensor in the sub-region of interest based on the target recognition data and the real recognition data; and wherein the storage module 1130 includes a sub-data storage unit, which is configured to store the sub-data and the perceived performance information of the sensor in the sub-region of interest in association with each other as an environmental performance relationship in the environmental performance relationship set.

[0096] Fig.12 A schematic block diagram of an example device 1200 that can be used to implement an embodiment of the present disclosure is shown. As shown, the device 1200 includes a computing unit 1201, which can perform various appropriate actions and processes according to computer program instructions stored in a read-only memory (ROM) 1202 or loaded from a storage unit 1208 into a random access memory (RAM) 1203. In the RAM 1203, various programs and data required for the operation of the device 1200 can also be stored. The computing unit 1201, the ROM 1202, and the RAM 1203 are connected to each other via a bus 1204. An input / output (I / O) interface 1205 is also connected to the bus 1204.

[0097] A number of components in the device 1200 are connected to the I / O interface 1205, including: an input unit 1206, such as a keyboard, a mouse, etc.; an output unit 1207, such as various types of displays, speakers, etc.; a storage unit 1208, such as a disk, an optical disk, etc.; and a communication unit 1209, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 1209 allows the device 1200 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0098] The computing unit 1201 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 1201 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 1201 performs the various methods and processes described above, such as method 300, method 600, method 700, method 800, or method 900. For example, in some embodiments, method 300, method 600, method 700, method 800, or method 900 may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 1208. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 1200 via ROM 1202 and / or communication unit 1209. When the computer program is loaded into RAM 1203 and executed by computing unit 1201, one or more steps of method 300, method 600, method 700, method 800 or method 900 described above may be executed. Alternatively, in other embodiments, computing unit 1201 may be configured to execute method 300, method 600, method 700, method 800 or method 900 in any other appropriate manner (e.g., by means of firmware).

[0099] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chip systems (SOCs), load programmable logic devices (CPLDs), and the like.

[0100] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0101] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In addition, although each operation is depicted in a specific order, this should be understood as requiring such operations to be performed in the specific order shown or in a sequential order, or requiring that all illustrated operations should be performed to obtain the desired result. In certain environments, multitasking and parallel processing may be advantageous. Similarly, although several specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Certain features described in the context of separate embodiments can also be implemented in a single implementation in combination. On the contrary, the various features described in the context of a single implementation can also be implemented in multiple implementations individually or in any suitable sub-combination.

[0102] Although the subject matter has been described in language specific to structural features and / or methodological logical actions, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely example forms of implementing the claims.

Claims

1. A method for determining perceived performance, include: Acquire first environmental data, wherein the first environmental data indicates environmental characteristics of an area of ​​interest of the sensor at a first moment; Acquire an environment-performance relationship set corresponding to the sensor, wherein the environment-performance relationship set includes at least one environment-performance relationship indicating a corresponding relationship between environment data and perception performance information; as well as Based on the first environmental data and the environmental performance relationship set, perception performance information of the sensor in the area of ​​interest at the first moment is determined.

2. The method according to claim 1, further comprising: include: Based on the first environment data, determining a first sub-region of interest in the region of interest, wherein the first sub-region of interest corresponds to first sub-data in the first environment data; as well as Determining the sensing performance information of the sensor in the area of ​​interest at the first moment includes: Based on the first sub-data and the environmental performance relationship set, perception performance information of the sensor in the first sub-area of ​​interest at the first moment is determined.

3. The method according to claim 2, wherein a first sub-region of interest in the region of interest is determined include: determining a first environment type corresponding to the first environment data; Based on the first environment type, determining a first area division strategy; as well as Based on the first area division strategy and the first environmental data, the first sub-region of interest is determined.

4. The method according to claim 2, further comprising: include: Obtaining a first threshold value; as well as Based on the sensing performance information of the first sub-region of interest and the first threshold, a working boundary state of the sensor in the first sub-region of interest at the first moment is determined.

5. The method according to any one of claims 1 to 4, wherein before acquiring the first environmental data, the method further include: Acquire second environmental data, wherein the second environmental data indicates a second environmental feature of the area of ​​interest of the sensor at a second moment, wherein the second moment is before the first moment; Acquire target recognition data of the sensor in the region of interest; Acquire real recognition data of a truth value system in the region of interest, wherein the truth value system is used to acquire the real situation of road traffic; Determining second perception performance information of the sensor in the region of interest at the second moment based on the target recognition data and the real recognition data; as well as The second environmental data and the second perceived performance information are stored in association as an environmental performance relationship in the environmental performance relationship set.

6. The method according to claim 5, further comprising: include: determining, based on the second environment data, a second sub-region of interest in the region of interest, wherein the second sub-region of interest corresponds to second sub-data in the second environment data; Wherein determining the second perception performance information of the sensor in the area of ​​interest at the second moment comprises: Determining, based on the target recognition data and the real recognition data, sensing performance information of the sensor in the second sub-region of interest at the second moment; and The storing of the second environmental data and the second perceived performance information as an environmental performance relationship in the environmental performance relationship set in association with each other includes: The second sub-data and the sensing performance information of the sensor in the second sub-region of interest at the second moment are associated with each other and stored as an environmental performance relationship in the environmental performance relationship set.

7. The method according to claim 6, wherein a second sub-region of interest in the region of interest is determined include: Determining a second environment type corresponding to the second environment data; Based on the second environment type, determining a second area division strategy; as well as Based on the second area division strategy and the second environment data, the second sub-region of interest is determined.

8. The method according to claim 5, wherein the second sensory performance information comprises a sensory performance level, wherein the second sensory performance information of the sensor in the region of interest at the second moment is determined include: Get the level determination strategy; Determining, based on the target recognition data and the real recognition data, a recognition accuracy of the sensor in the region of interest; as well as The sensing performance level of the sensor in the region of interest is determined based on the recognition accuracy and the level determination strategy.

9. A method for generating an environment-performance relationship, include: acquiring environmental data, wherein the environmental data indicates environmental characteristics of an area of ​​interest of the sensor; determining sensing performance information of the sensor at the region of interest; as well as The environmental data and the perceived performance information are stored in association as an environmental performance relationship in an environmental performance relationship set.

10. The method according to claim 9, wherein the sensing performance information of the sensor at the area of ​​interest is determined include: Acquire target recognition data of the sensor in the region of interest; Acquire real recognition data of a truth value system in the region of interest, wherein the truth value system is used to acquire the real situation of road traffic; as well as Based on the target recognition data and the real recognition data, the perception performance information is determined.

11. The method according to claim 10, further comprising: include: Based on the environmental data, determining a sub-region of interest in the region of interest, wherein the sub-region of interest corresponds to sub-data in the environmental data; Wherein determining the perceived performance information comprises: Determining sensing performance information of the sensor in the sub-area of ​​interest based on the target recognition data and the real recognition data; and The storing of the environmental data and the perceived performance information as an environmental performance relationship in an environmental performance relationship set in association with each other includes: The sub-data and the sensing performance information of the sensor in the sub-area of ​​interest are stored in association with each other as an environmental performance relationship in the environmental performance relationship set.

12. A device for determining perceived performance, include: A data acquisition module is configured to acquire first environmental data, wherein the first environmental data indicates environmental characteristics of an area of ​​interest of the sensor at a first moment; a relationship acquisition module configured to acquire an environment-performance relationship set corresponding to the sensor, wherein the environment-performance relationship set includes at least one environment-performance relationship indicating a corresponding relationship between environmental data and perception performance information; as well as The performance determination module is configured to determine the perception performance information of the sensor in the area of ​​interest at the first moment based on the first environmental data and the environmental-performance relationship set.

13. An apparatus for generating an environment-performance relationship, include: an environmental data acquisition module, configured to acquire environmental data, wherein the environmental data indicates environmental characteristics of an area of ​​interest of the sensor; A performance determination module configured to determine sensing performance information of the sensor at the region of interest; as well as The storage module is configured to store the environmental data and the perceived performance information in association with each other as an environmental performance relationship in an environmental performance relationship set.

14. An electronic device, include: processor; as well as A memory coupled to the processor, the memory having instructions stored therein, wherein when the instructions are executed by the processor, the electronic device executes the method according to any one of claims 1 to 8 or claims 9 to 11.

15. A computer-readable storage medium having computer-executable instructions stored thereon, wherein the computer-executable instructions are executed by a processor to implement the method according to any one of claims 1 to 8 or claims 9 to 11.

16. An intelligent driving method, include: Acquire sensory performance information of the sensor, wherein the sensory performance information is determined according to the method of any one of claims 1-8; as well as Based on the perceived performance information, availability of at least one intelligent driving function associated with the sensor is determined.

17. The intelligent driving method according to claim 16, further comprising: include: Determine, according to the perception performance information, a working boundary state of the sensor in the region of interest, wherein the working boundary state is used to indicate whether the perception performance of the sensor in the region of interest is within a working boundary or outside a working boundary; as well as According to the working boundary state, the at least one intelligent driving function is started, kept running, not started or exited.

Images