Obstacle detection method, device, equipment and medium for autonomous driving vehicle

Abstract

The present application discloses an obstacle detection method, device, equipment and medium for an autonomous driving vehicle, wherein the method includes: obtaining first obstacle information collected by the V2X device of the autonomous driving vehicle, and obtaining second obstacle information collected by the autonomous driving vehicle's own sensor; determining whether the collection time of the first obstacle information and the second obstacle information is time synchronized; if the collection time is time synchronized, matching the attributes of the first obstacle information and the second obstacle information, otherwise, if the collection time is not time synchronized, obtaining the attributes of the third obstacle information synchronized with the first obstacle information, and matching the attributes of the third obstacle information with the attributes of the second obstacle information; obtaining the fusion result of the first obstacle information and the second obstacle information according to the matching result, and identifying the actual obstacles around the autonomous driving vehicle based on the fusion result. Thus, the problem of limited detection distance, blind spots in the visual angle, and susceptibility to external environmental interference caused by only detecting by the vehicle's own sensors is solved, which greatly improves the vehicle's perception range and perception accuracy, and improves the safety of the autonomous driving vehicle.

Description

Technical Field

[0001] The present application relates to the field of vehicle technology, and in particular to an obstacle detection method, device, equipment and medium for an autonomous driving vehicle. Background Art

[0002] Autonomous driving can be divided into single-vehicle and networked types in terms of construction methods. Single-vehicle type is mainly realized through vehicle-side sensors, while networked type uses V2X (vehicle to X, vehicle wireless communication technology) to achieve multi-terminal integration. Among related technologies, autonomous driving mainly relies on the integration of its own sensors.

[0003] However, since autonomous vehicles are not isolated individuals, focusing only on the vehicle and the surrounding road environment can no longer meet the higher requirements of safe, efficient, and energy-saving driving. In addition, there is the problem of sensors failing or misjudging in certain situations, causing traffic accidents, which needs to be solved urgently.

[0004] Application Contents

[0005] The present application provides an obstacle detection method, device, equipment and medium for an autonomous driving vehicle to solve the problems of limited detection distance, blind spots in viewing angle, and susceptibility to external environmental interference caused by detection only through the vehicle's own sensors, greatly improving the vehicle's perception range and perception accuracy, and improving the safety of autonomous driving vehicles.

[0006] The first aspect of the present application provides an obstacle detection method for an autonomous driving vehicle, comprising the following steps:

[0007] Obtaining first obstacle information collected by a V2X device of the autonomous driving vehicle, and obtaining second obstacle information collected by a sensor of the autonomous driving vehicle itself;

[0008] Determining whether the collection time of the first obstacle information and the second obstacle information is synchronized;

[0009] If the acquisition time is time-synchronized, matching the attributes of the first obstacle information and the second obstacle information; otherwise, if the acquisition time is not time-synchronized, obtaining the attributes of the third obstacle information that is time-synchronized from the first obstacle information, and matching the attributes of the third obstacle information and the attributes of the second obstacle information; and

[0010] A fusion result of the first obstacle information and the second obstacle information is obtained according to the matching result, and actual obstacles around the autonomous driving vehicle are identified based on the fusion result.

[0011] Optionally, the attributes of the third obstacle information time-synchronized by the first obstacle information include:

[0012] Cleaning the first obstacle information, filtering out obstacle information that meets an abnormal condition, and obtaining cleaned obstacle information;

[0013] Interpolate or extrapolate the cleaned obstacle information to obtain attributes of the time-synchronized third obstacle information.

[0014] Optionally, the matching of attributes of the first obstacle information and the second obstacle information includes:

[0015] generating a corresponding feature code according to the obstacle feature of each obstacle in the first obstacle information and the second obstacle information;

[0016] It is determined whether the first obstacle information and the second obstacle information match according to the corresponding feature code.

[0017] Optionally, obtaining a fusion result of the first obstacle information and the second obstacle information according to the matching result includes:

[0018] When the match is complete, taking the first obstacle information or the second obstacle information as the fusion result;

[0019] When the obstacle types do not match, the first obstacle information is corrected according to the confidence of the first obstacle information, and the fusion result is obtained by combining the corrected first obstacle information and the second obstacle information;

[0020] When the spatial coordinates do not match and the confidence level is less than a preset threshold, the first obstacle information is discarded, and the fusion result is obtained from the second obstacle information;

[0021] When the spatial coordinates do not match, and the confidence level is greater than or equal to the preset threshold, re-matching is performed according to the spatial and temporal characteristics of the obstacle to obtain a new matching result, and the fusion result is obtained based on the new matching result.

[0022] Optionally, re-matching according to the temporal and spatial characteristics of the obstacle to obtain a new matching result includes:

[0023] Calculating the similarity of obstacle historical trajectory information of the first obstacle information and the second obstacle information;

[0024] If the similarity is greater than a preset value, the first obstacle information is discarded, and the fusion result is obtained from the second obstacle information;

[0025] If the similarity is less than or equal to the preset value, the first obstacle information and the second obstacle information are combined to obtain the fusion result.

[0026] A second aspect of the present application provides an obstacle detection device for an autonomous driving vehicle, comprising:

[0027] An acquisition module, used to acquire first obstacle information collected by a V2X device of the autonomous driving vehicle, and to acquire second obstacle information collected by a sensor of the autonomous driving vehicle itself;

[0028] A judging module, used to judge whether the collection time of the first obstacle information and the second obstacle information is synchronized;

[0029] a matching module, configured to match the attributes of the first obstacle information and the second obstacle information if the acquisition time is time-synchronized, otherwise, obtain the attributes of third obstacle information that is time-synchronized from the first obstacle information if the acquisition time is not time-synchronized, and match the attributes of the third obstacle information with the attributes of the second obstacle information; and

[0030] An identification module is used to obtain a fusion result of the first obstacle information and the second obstacle information according to the matching result, and to identify actual obstacles around the autonomous driving vehicle based on the fusion result.

[0031] Optionally, the attributes of the third obstacle information time-synchronized by the first obstacle information include:

[0032] Cleaning the first obstacle information, filtering out obstacle information that meets an abnormal condition, and obtaining cleaned obstacle information;

[0033] Interpolate or extrapolate the cleaned obstacle information to obtain attributes of the time-synchronized third obstacle information.

[0034] Optionally, the matching module is specifically used to:

[0035] generating a corresponding feature code according to the obstacle feature of each obstacle in the first obstacle information and the second obstacle information;

[0036] It is determined whether the first obstacle information and the second obstacle information match according to the corresponding feature code.

[0037] Optionally, the identification module is specifically used to:

[0038] When the match is complete, taking the first obstacle information or the second obstacle information as the fusion result;

[0039] When the obstacle types do not match, the first obstacle information is corrected according to the confidence of the first obstacle information, and the fusion result is obtained by combining the corrected first obstacle information and the second obstacle information;

[0040] When the spatial coordinates do not match and the confidence level is less than a preset threshold, the first obstacle information is discarded, and the fusion result is obtained from the second obstacle information;

[0041] When the spatial coordinates do not match, and the confidence level is greater than or equal to the preset threshold, re-matching is performed according to the spatial and temporal characteristics of the obstacle to obtain a new matching result, and the fusion result is obtained based on the new matching result.

[0042] Optionally, the identification module is further used to:

[0043] Calculating the similarity of obstacle historical trajectory information of the first obstacle information and the second obstacle information;

[0044] If the similarity is greater than a preset value, the first obstacle information is discarded, and the fusion result is obtained from the second obstacle information;

[0045] If the similarity is less than or equal to the preset value, the fusion result is obtained by merging the first obstacle information and the second obstacle information. The third aspect of the present application provides a vehicle, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the program to implement the obstacle detection method for the autonomous driving vehicle as described in the above embodiment.

[0046] The fourth aspect of the present application provides a computer-readable storage medium having a computer program stored thereon, characterized in that the program is executed by a processor to implement an obstacle detection method for an autonomous driving vehicle as described in the above embodiments.

[0047] Thus, the first obstacle information collected by the V2X device of the autonomous vehicle can be obtained, and the second obstacle information collected by the autonomous vehicle's own sensor can be obtained. If the collection time of the first obstacle information and the second obstacle information is time-synchronized, the attributes of the first obstacle information and the second obstacle information are matched. Otherwise, the attributes of the third obstacle information synchronized with the first obstacle information are obtained, and the attributes of the third obstacle information and the attributes of the second obstacle information are matched. The fusion result of the first obstacle information and the second obstacle information is obtained according to the matching result, and the actual obstacles around the autonomous vehicle are identified based on the fusion result. Thus, the problem of limited detection distance, blind spots in the visual angle, and susceptibility to external environmental interference caused by only detecting by the vehicle's own sensors can be solved, which greatly improves the vehicle's perception range and perception accuracy, and improves the safety of autonomous vehicles.

[0048] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0050] Figure 1 A flowchart of an obstacle detection method for an autonomous driving vehicle provided according to an embodiment of the present application;

[0051] Figure 2 A schematic diagram of the principle of fusion of obstacle information according to an embodiment of the present application;

[0052] Figure 3 is a block diagram of an obstacle detection device for an autonomous driving vehicle according to an embodiment of the present application;

[0053] Figure 4 A schematic diagram of the structure of a vehicle provided for an application embodiment. DETAILED DESCRIPTION

[0054] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0055] The obstacle detection method, device, equipment and medium of the autonomous driving vehicle of the embodiment of the present application are described below with reference to the accompanying drawings. In view of the problem mentioned in the background technology center that only the detection by the vehicle's own sensor leads to limited detection distance, blind spots in the visual angle, and susceptibility to external environmental interference, the present application provides an obstacle detection method for an autonomous driving vehicle, in which the first obstacle information collected by the V2X device of the autonomous driving vehicle can be obtained, and the second obstacle information collected by the autonomous driving vehicle's own sensor can be obtained. If the acquisition time of the first obstacle information and the second obstacle information is time-synchronized, the attributes of the first obstacle information and the second obstacle information are matched, otherwise the attributes of the third obstacle information synchronized with the time are obtained from the first obstacle information, and the attributes of the third obstacle information and the attributes of the second obstacle information are matched, and the fusion result of the first obstacle information and the second obstacle information is obtained according to the matching result, and the actual obstacles around the autonomous driving vehicle are identified based on the fusion result. In this way, the problem that only the detection by the vehicle's own sensor leads to limited detection distance, blind spots in the visual angle, and susceptibility to external environmental interference can be solved, which greatly improves the perception range and perception accuracy of the vehicle and improves the safety of the autonomous driving vehicle.

[0056] Specifically, Figure 1 A flowchart of an obstacle detection method for an autonomous driving vehicle provided in an embodiment of the present application.

[0057] like Figure 1 As shown, the obstacle detection method of the autonomous driving vehicle includes the following steps:

[0058] In step S101, first obstacle information collected by a V2X device of the autonomous driving vehicle is obtained, and second obstacle information collected by the autonomous driving vehicle's own sensor is obtained.

[0059] Specifically, if Figure 2 As shown, the embodiment of the present application can collect the first obstacle information through the V2X device and the second obstacle information through the vehicle's own sensors, such as a positioning module, a perception module, etc., wherein the first obstacle information and the second obstacle information may include the type of obstacle, type confidence, coordinates, timestamp, speed and size, etc.

[0060] It should be noted that V2X equipment refers to vehicles equipped with this equipment. In the autonomous driving mode, it can automatically select the driving route with the best road conditions by analyzing real-time traffic information, thereby greatly alleviating traffic congestion.

[0061] In step S102, it is determined whether the collection time of the first obstacle information and the second obstacle information is synchronized.

[0062] In step S103, if the acquisition time is time-synchronized, the attributes of the first obstacle information and the second obstacle information are matched; otherwise, if the acquisition time is not time-synchronized, the attributes of the third obstacle information with time synchronization are obtained from the first obstacle information, and the attributes of the third obstacle information and the attributes of the second obstacle information are matched.

[0063] Specifically, if the first obstacle information collected by the V2X device is synchronized in time with the second obstacle information collected by the vehicle's own sensor, the obstacle attributes can be matched directly.

[0064] Optionally, in some embodiments, matching the attributes of the first obstacle information and the second obstacle information includes: generating a corresponding feature code based on the obstacle features of each obstacle in the first obstacle information and the second obstacle information; and determining whether the first obstacle information and the second obstacle information match based on the corresponding feature codes.

[0065] Specifically, when matching the attributes of the first obstacle information and the second obstacle information, the embodiment of the present application can encode all obstacles into seven-dimensional features according to their type, type confidence, coordinates, timestamp, speed and size, and determine whether the first obstacle information and the second obstacle information match according to the corresponding feature codes.

[0066] It should be noted that the attributes to be matched mainly include the spatial coordinates of the obstacle and the type of obstacle. The spatial matching is mainly based on the spatial distance of the obstacle, and the matching of the obstacle type is mainly based on the confidence of the V2X obstacle and the confidence of the vehicle's perception of the obstacle.

[0067] Furthermore, in some embodiments, obtaining attributes of third obstacle information synchronized with time from the first obstacle information includes: cleaning the first obstacle information, filtering out obstacle information that meets abnormal conditions, and obtaining cleaned obstacle information; and interpolating or extrapolating the cleaned obstacle information to obtain attributes of the third obstacle information synchronized with time.

[0068] That is to say, in the embodiment of the present application, the first obstacle information can be cleaned, the abnormal points in the historical trajectory information can be deleted, the cleaned obstacle information can be obtained, and the cleaned obstacle information can be interpolated or extrapolated to obtain the attributes of the third obstacle information synchronized with time, and then the attributes of the third obstacle information can be matched with the attributes of the second obstacle information.

[0069] In step S104, a fusion result of the first obstacle information and the second obstacle information is obtained according to the matching result, and actual obstacles around the autonomous driving vehicle are identified based on the fusion result.

[0070] Optionally, in some embodiments, a fusion result of the first obstacle information and the second obstacle information is obtained based on the matching result, including: when there is a complete match, taking the first obstacle information or the second obstacle information as the fusion result; when the obstacle type does not match, correcting the first obstacle information according to the confidence of the first obstacle information, and merging the corrected first obstacle information and the second obstacle information to obtain the fusion result; when the spatial coordinates do not match and the confidence is less than a preset threshold, discarding the first obstacle information and obtaining the fusion result from the second obstacle information; when the spatial coordinates do not match, when the confidence is greater than or equal to the preset threshold, re-matching is performed according to the spatiotemporal characteristics of the obstacle to obtain a new matching result, and obtaining the fusion result based on the new matching result.

[0071] Specifically, if the first obstacle information collected by V2X and the second obstacle information collected by the vehicle's own sensor can completely match, the obstacle information can be directly output as the final result (that is, the first obstacle information or the second obstacle information is used as the fusion result); if the first obstacle information collected by V2X and the second obstacle information collected by the vehicle's own sensor cannot match only in type, the first obstacle information is corrected according to the confidence of the first obstacle information, and then output as the final result; if the first obstacle information collected by V2X and the second obstacle information collected by the vehicle's own sensor fail to successfully match in space (that is, the spatial coordinates do not match), the confidence of the first obstacle information is further determined. If the confidence of the first obstacle information is less than a preset threshold, the first obstacle information is considered unreliable and can be directly discarded; if the confidence of the first obstacle information is greater than or equal to the preset threshold, further matching of spatiotemporal features is required.

[0072] Optionally, in some embodiments, matching is performed again according to the temporal and spatial characteristics of the obstacle to obtain a new matching result, including: calculating the similarity of the obstacle historical trajectory information of the first obstacle information and the second obstacle information; if the similarity is greater than a preset value, discarding the first obstacle information and obtaining a fusion result from the second obstacle information; if the similarity is less than or equal to the preset value, merging the first obstacle information and the second obstacle information to obtain a fusion result.

[0073] Specifically, the spatiotemporal trajectory matching is to judge the similarity of the obstacle historical trajectory information of the first obstacle information collected by V2X and the second obstacle information collected by the vehicle's own sensor. If the similarity is greater than a preset value, it is considered that the positioning information of the first obstacle information collected by V2X is inaccurate, and the first obstacle information can be directly discarded; if the similarity is less than or equal to the preset value, it is considered that the first obstacle information collected by V2X does not appear in the second obstacle information obtained by the vehicle itself, and the first obstacle information and the second obstacle information can be merged into the final fusion result.

[0074] It should be noted that the spatiotemporal trajectory matching is determined by time-weighted similarity and space-weighted similarity. Given two continuous trajectories T1: p1 = p1(t), t∈[α1, β1] and T2: p2 = p2(t), t∈[α2, β2], their overlapping time period is t∈[α, β]. Intuitively, if the distance between the points of the two trajectories is close within a longer time period or a longer distance segment, the two trajectories are considered to be similar. Based on this, the similarity of the spatiotemporal trajectory can be described by two evaluation criteria: time-weighted similarity and space-weighted similarity, which are defined as follows:

[0075] Time Weighted Similarity (TWS) is defined as:

[0076]

[0077] Among them, m(p1(t), p2(t)) represents the proximity of two points. The numerator on the right side of the formula is the integral of the proximity of the points over time, which can be regarded as the time length of the trajectory fitting. Then divided by the total duration, it is the time proportion of the trajectory fitting, that is, the time-weighted similarity. Space Weighted Similarity (SWS) is defined as:

[0078]

[0079] Among them, v1(t) refers to the speed of T1 at time t, the denominator on the right side of the formula is the length of T1, and the numerator is the product of the proximity of the point and its speed and then the integral of time, which can be regarded as the distance of trajectory fitting. The division of the two is the distance ratio of trajectory fitting, that is, the space (distance) weighted similarity.

[0080] In practice, the trajectory data is a discrete time stamp and position sequence, which cannot be integrated. Therefore, the embodiment of the present application can use the linear interpolation method to infer the position of the trajectory between two sampling times, and then use the trapezoidal formula to approximate the integral:

[0081]

[0082]

[0083] During the calculation process, the trajectory needs to be interpolated first, then the proximity of the points at the corresponding timestamps is calculated, and finally the integral is calculated, and finally the fused obstacle information is output.

[0084] Therefore, by integrating V2X technology with vehicle sensors, the shortcomings of vehicle sensors such as limited detection distance, blind spots in viewing angles, and susceptibility to external environmental interference can be overcome, greatly improving the vehicle's perception range and perception accuracy. Not only can it support vehicles to obtain more information than a single vehicle can perceive, and promote the innovation and application of autonomous driving technology; it is also conducive to building a smart transportation system, promoting the development of new models and new formats of automobiles and transportation services, and is of great significance to improving traffic efficiency, saving resources, reducing pollution, reducing accident rates, and improving traffic management.

[0085] According to the obstacle detection method of the autonomous driving vehicle proposed in the embodiment of the present application, the first obstacle information collected by the V2X device of the autonomous driving vehicle can be obtained, and the second obstacle information collected by the autonomous driving vehicle's own sensor can be obtained. If the collection time of the first obstacle information and the second obstacle information is time-synchronized, the attributes of the first obstacle information and the second obstacle information are matched, otherwise the attributes of the third obstacle information synchronized with the first obstacle information are obtained, and the attributes of the third obstacle information and the attributes of the second obstacle information are matched, and the fusion result of the first obstacle information and the second obstacle information is obtained according to the matching result, and the actual obstacles around the autonomous driving vehicle are identified based on the fusion result. In this way, the problem of limited detection distance, blind spots in the viewing angle, and susceptibility to external environmental interference caused by only detecting by the vehicle's own sensors can be solved, which greatly improves the vehicle's perception range and perception accuracy, and improves the safety of the autonomous driving vehicle.

[0086] Next, an obstacle detection device for an autonomous driving vehicle in some embodiments proposed in the embodiments of the present application will be described with reference to the accompanying drawings.

[0087] Figure 3 is a block diagram of an obstacle detection device for an autonomous driving vehicle in some embodiments of the present application.

[0088] like Figure 3 As shown, the obstacle detection device 10 of the autonomous driving vehicle includes: an acquisition module 100, a judgment module 200, a matching module 300 and an identification module 400.

[0089] The acquisition module 100 is used to acquire first obstacle information collected by a V2X device of the autonomous driving vehicle, and to acquire second obstacle information collected by a sensor of the autonomous driving vehicle itself;

[0090] The judging module 200 is used to judge whether the collection time of the first obstacle information and the second obstacle information is synchronized;

[0091] The matching module 300 is used to match the attributes of the first obstacle information and the second obstacle information if the acquisition time is time-synchronized, otherwise, if the acquisition time is not time-synchronized, obtain the attributes of the third obstacle information that is time-synchronized from the first obstacle information, and match the attributes of the third obstacle information with the attributes of the second obstacle information; and

[0092] The recognition module 400 is used to obtain a fusion result of the first obstacle information and the second obstacle information according to the matching result, and to identify actual obstacles around the autonomous driving vehicle based on the fusion result.

[0093] Optionally, the attributes of the third obstacle information synchronized with the time obtained from the first obstacle information include:

[0094] Cleaning the first obstacle information, filtering out obstacle information that meets an abnormal condition, and obtaining cleaned obstacle information;

[0095] The cleaned obstacle information is interpolated or extrapolated to obtain attributes of the time-synchronized third obstacle information.

[0096] Optionally, the matching module 300 is specifically used for:

[0097] Generate a corresponding feature code according to the obstacle feature of each obstacle in the first obstacle information and the second obstacle information;

[0098] It is determined whether the first obstacle information and the second obstacle information match according to the corresponding feature codes.

[0099] Optionally, the identification module 400 is specifically used for:

[0100] When the match is complete, the first obstacle information or the second obstacle information is used as a fusion result;

[0101] When the obstacle types do not match, the first obstacle information is corrected according to the confidence of the first obstacle information, and the corrected first obstacle information and the second obstacle information are combined to obtain a fusion result;

[0102] When the spatial coordinates do not match and the confidence level is less than a preset threshold, the first obstacle information is discarded and a fusion result is obtained from the second obstacle information;

[0103] When the spatial coordinates do not match, and the confidence is greater than or equal to a preset threshold, the obstacle is re-matched according to its spatiotemporal characteristics to obtain a new matching result, and a fusion result is obtained based on the new matching result.

[0104] Optionally, the identification module 400 is further configured to:

[0105] Calculating the similarity of obstacle historical trajectory information between the first obstacle information and the second obstacle information;

[0106] If the similarity is greater than a preset value, the first obstacle information is discarded and the fusion result is obtained from the second obstacle information;

[0107] If the similarity is less than or equal to a preset value, the first obstacle information and the second obstacle information are combined to obtain a fusion result.

[0108] It should be noted that the above explanation of the embodiment of the obstacle detection method for an autonomous driving vehicle is also applicable to the obstacle detection device for an autonomous driving vehicle of this embodiment, and will not be repeated here.

[0109] According to the obstacle detection device of the autonomous driving vehicle proposed in the embodiment of the present application, the first obstacle information collected by the V2X device of the autonomous driving vehicle can be obtained, and the second obstacle information collected by the autonomous driving vehicle's own sensor can be obtained. If the collection time of the first obstacle information and the second obstacle information is time-synchronized, the attributes of the first obstacle information and the second obstacle information are matched, otherwise the attributes of the third obstacle information synchronized with the first obstacle information are obtained, and the attributes of the third obstacle information and the attributes of the second obstacle information are matched, and the fusion result of the first obstacle information and the second obstacle information is obtained according to the matching result, and the actual obstacles around the autonomous driving vehicle are identified based on the fusion result. In this way, the problem of limited detection distance, blind spots in the visual angle, and susceptibility to external environmental interference caused by only detecting by the vehicle's own sensors can be solved, which greatly improves the vehicle's perception range and perception accuracy, and improves the safety of the autonomous driving vehicle.

[0110] Figure 4 A schematic diagram of the structure of a vehicle provided in an embodiment of the present application. The vehicle may include:

[0111] Memory 401 , processor 402 , and a computer program stored in the memory 401 and executable on the processor 402 .

[0112] When the processor 402 executes the program, the obstacle detection method for the autonomous driving vehicle provided in the above embodiment is implemented.

[0113] Furthermore, the vehicle also includes:

[0114] The communication interface 403 is used for communication between the memory 401 and the processor 402 .

[0115] The memory 401 is used to store computer programs that can be executed on the processor 402 .

[0116] The memory 401 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0117] If the memory 401, the processor 402 and the communication interface 403 are implemented independently, the communication interface 403, the memory 401 and the processor 402 can be connected to each other through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0118] Optionally, in a specific implementation, if the memory 401, the processor 402 and the communication interface 403 are integrated on a chip, the memory 401, the processor 402 and the communication interface 403 can communicate with each other through an internal interface.

[0119] The processor 402 may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0120] This embodiment also provides a computer-readable storage medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the above-mentioned obstacle detection method for an autonomous driving vehicle.

[0121] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0122] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0123] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0124] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiment, the N steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0125] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

Claims

1. An obstacle detection method for an autonomous driving vehicle, characterized in that: The following steps are involved: Obtaining first obstacle information collected by a V2X device of the autonomous driving vehicle, and obtaining second obstacle information collected by a sensor of the autonomous driving vehicle itself; Determining whether the collection time of the first obstacle information and the second obstacle information is synchronized; If the acquisition time is time-synchronized, the attributes of the first obstacle information and the second obstacle information are matched; otherwise, if the acquisition time is not time-synchronized, the attributes of the third obstacle information with time synchronization are obtained from the first obstacle information, and the attributes of the third obstacle information and the attributes of the second obstacle information are matched; as well as Obtaining a fusion result of the first obstacle information and the second obstacle information according to the matching result, and identifying actual obstacles around the autonomous driving vehicle based on the fusion result; The obtaining of the fusion result of the first obstacle information and the second obstacle information according to the matching result includes: when there is a complete match, taking the first obstacle information or the second obstacle information as the fusion result; when the obstacle type does not match, correcting the first obstacle information according to the confidence of the first obstacle information, and merging the corrected first obstacle information and the second obstacle information to obtain the fusion result; when the spatial coordinates do not match and the confidence is less than a preset threshold, discarding the first obstacle information and obtaining the fusion result from the second obstacle information; when the spatial coordinates do not match and the confidence is greater than or equal to the preset threshold, re-matching according to the temporal and spatial characteristics of the obstacle to obtain a new matching result, and obtaining the fusion result based on the new matching result; The re-matching according to the spatiotemporal characteristics of the obstacle to obtain the new matching result includes: calculating the similarity of the obstacle historical trajectory information of the first obstacle information and the second obstacle information; if the similarity is greater than a preset value, discarding the first obstacle information and obtaining the fusion result from the second obstacle information; if the similarity is less than or equal to the preset value, merging the first obstacle information and the second obstacle information to obtain the fusion result.

2. The method according to claim 1, characterized in that The attributes of the third obstacle information time-synchronized by the first obstacle information include: Cleaning the first obstacle information, filtering out obstacle information that meets an abnormal condition, and obtaining cleaned obstacle information; Interpolate or extrapolate the cleaned obstacle information to obtain attributes of the time-synchronized third obstacle information.

3. The method according to claim 1, characterized in that The matching of the attributes of the first obstacle information and the second obstacle information includes: generating a corresponding feature code according to the obstacle feature of each obstacle in the first obstacle information and the second obstacle information; It is determined whether the first obstacle information and the second obstacle information match according to the corresponding feature code.

4. An obstacle detection device for an autonomous driving vehicle, characterized in that: include: An acquisition module, used to acquire first obstacle information collected by a V2X device of the autonomous driving vehicle, and to acquire second obstacle information collected by a sensor of the autonomous driving vehicle itself; A judging module, used to judge whether the collection time of the first obstacle information and the second obstacle information is synchronized; a matching module, configured to match the attributes of the first obstacle information and the second obstacle information if the acquisition time is time-synchronized, or to obtain the attributes of third obstacle information that is time-synchronized from the first obstacle information if the acquisition time is not time-synchronized, and to match the attributes of the third obstacle information and the attributes of the second obstacle information; as well as an identification module, configured to obtain a fusion result of the first obstacle information and the second obstacle information according to the matching result, and identify actual obstacles around the autonomous driving vehicle based on the fusion result; The recognition module is specifically configured to: when there is a complete match, use the first obstacle information or the second obstacle information as the fusion result; when the obstacle types do not match, correct the first obstacle information according to the confidence of the first obstacle information, and obtain the fusion result by combining the corrected first obstacle information and the second obstacle information; When the spatial coordinates do not match and the confidence is less than a preset threshold, the first obstacle information is discarded, and the fusion result is obtained from the second obstacle information; when the spatial coordinates do not match and the confidence is greater than or equal to the preset threshold, matching is performed again according to the temporal and spatial characteristics of the obstacle to obtain a new matching result, and the fusion result is obtained based on the new matching result; The recognition module is further used to: calculate the similarity of obstacle historical trajectory information of the first obstacle information and the second obstacle information; if the similarity is greater than a preset value, discard the first obstacle information and obtain the fusion result from the second obstacle information; if the similarity is less than or equal to the preset value, combine the first obstacle information and the second obstacle information to obtain the fusion result.

5. The device according to claim 4, characterized in that The attributes of the third obstacle information time-synchronized by the first obstacle information include: Cleaning the first obstacle information, filtering out obstacle information that meets an abnormal condition, and obtaining cleaned obstacle information; Interpolate or extrapolate the cleaned obstacle information to obtain attributes of the time-synchronized third obstacle information.

6. The device according to claim 4, characterized in that The matching module is specifically used for: generating a corresponding feature code according to the obstacle feature of each obstacle in the first obstacle information and the second obstacle information; It is determined whether the first obstacle information and the second obstacle information match according to the corresponding feature code.

7. A vehicle, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the obstacle detection method for an autonomous driving vehicle as described in any one of claims 1 to 3.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the obstacle detection method for an autonomous driving vehicle as described in any one of claims 1 to 3.

Images