Pedestrian object intention determination method, device, medium and autonomous driving vehicle

Abstract

The present disclosure provides a method, apparatus, device, storage medium, program product, and autonomous driving vehicle for determining pedestrian object intention, and relates to the field of computer technology, and in particular to the fields of artificial intelligence, autonomous driving, and intelligent transportation technology. The specific implementation scheme of the pedestrian object intention determination method is as follows: clustering pedestrian object data to obtain at least one cluster, wherein each cluster includes at least one target pedestrian object data; for any cluster, determining the intention reference data of each target pedestrian object data to obtain at least one initial intention reference data; and determining the cluster intention data of the cluster based on the at least one initial intention reference data as the target intention data of each target pedestrian object in the cluster.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to the fields of artificial intelligence, autonomous driving, and intelligent transportation technology. Background Art

[0002] With the development of computer and internet technologies, autonomous driving has become a key branch of artificial intelligence. In this field, pedestrian intentions are crucial for controlling autonomous vehicles. Therefore, efficiently determining pedestrian intentions has become a pressing issue. Summary of the Invention

[0003] The present disclosure provides a method, apparatus, device, storage medium, program product, and autonomous driving vehicle for determining pedestrian object intention.

[0004] According to one aspect of the present disclosure, a method for determining pedestrian object intention is provided, including clustering pedestrian object data to obtain at least one cluster cluster, wherein each cluster cluster includes at least one target pedestrian object data; for any cluster cluster, determining the intention reference data of each target pedestrian object data to obtain at least one initial intention reference data; and determining the cluster intention data of the cluster cluster based on the at least one initial intention reference data as the target intention data of each target pedestrian object of the cluster cluster.

[0005] According to another aspect of the present disclosure, a pedestrian object intention determination device is provided, comprising: a clustering module, an initial intention reference data determination module, and a cluster intention data determination module. The clustering module is used to cluster the pedestrian object data to obtain at least one cluster cluster, wherein each cluster cluster includes at least one target pedestrian object data; the initial intention reference data determination module is used to determine the intention reference data of each target pedestrian object data in any cluster cluster to obtain at least one initial intention reference data; the cluster intention data determination module is used to determine the cluster intention data of the cluster cluster based on the at least one initial intention reference data as the target intention data of each target pedestrian object in the cluster cluster.

[0006] According to another aspect of the present disclosure, an electronic device is provided, comprising: at least one processor and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method of an embodiment of the present disclosure.

[0007] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, where the computer instructions are used to cause the computer to execute the method of the embodiment of the present disclosure.

[0008] According to another aspect of the present disclosure, a computer program product is provided, including a computer program, which implements the method of the embodiment of the present disclosure when executed by a processor.

[0009] According to another aspect of the present disclosure, an autonomous driving vehicle is provided, comprising the electronic device of an embodiment of the present disclosure.

[0010] It should be understood that the contents described in this section are not intended to identify 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 readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The accompanying drawings are provided to facilitate a better understanding of the present invention and do not constitute a limitation of the present disclosure.

[0012] Figure 1 The system architecture diagram of the object intent determination method and apparatus according to an embodiment of the present disclosure is schematically shown;

[0013] Figure 2 A flowchart of a method for determining object intention according to an embodiment of the present disclosure is schematically shown;

[0014] Figure 3A The figure schematically shows a schematic diagram of obtaining at least one cluster according to an embodiment of the present disclosure;

[0015] Figure 3B A schematic diagram schematically illustrates how a distance threshold is determined according to vehicle traffic conditions according to an embodiment of the present disclosure;

[0016] Figure 4 Schematically illustrates a schematic diagram of obtaining at least one initial intention reference data according to an embodiment of the present disclosure;

[0017] Figure 5 A schematic diagram of determining clustering intention data of cluster clusters according to an embodiment of the present disclosure is schematically shown;

[0018] Figure 6 Schematically illustrates a schematic diagram of determining clustering intention data of cluster clusters according to another embodiment of the present disclosure;

[0019] Figure 7 A block diagram schematically illustrates an apparatus for determining object intent according to an embodiment of the present disclosure; and

[0020] Figure 8 A block diagram schematically shows an electronic device that can implement the object intention determination method according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0021] The following description of exemplary embodiments of the present disclosure is made in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding. These details should be considered as merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0022] The terms used herein are only for describing specific embodiments and are not intended to limit the present disclosure. The terms "comprise," "include," etc. used herein indicate the presence of the features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0023] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0024] When expressions such as "at least one of A, B and C, etc." are used, they should generally be interpreted in accordance with the meaning of the expression commonly understood by those skilled in the art (for example, "a system having at least one of A, B and C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.).

[0025] Autonomous driving technology can be understood as a technology based on computer and artificial intelligence technologies that enables complete, safe, and efficient driving without human intervention. Autonomous vehicles are an application of autonomous technology. Autonomous vehicles must avoid obstacles such as pedestrians. Therefore, pedestrian intentions are crucial for controlling autonomous vehicles. Pedestrian intentions, for example, include whether to proceed or not.

[0026] Figure 1 The system architecture of the method and apparatus for determining pedestrian object intention according to an embodiment of the present disclosure is schematically shown. It should be noted that, Figure 1 The examples shown are merely examples of system architectures to which the embodiments of the present disclosure may be applied, to help those skilled in the art understand the technical content of the present disclosure, but do not mean that the embodiments of the present disclosure may not be used in other devices, systems, environments or scenarios.

[0027] like Figure 1As shown, the system architecture 100 according to this embodiment includes a plurality of vehicles 101, 102, 103, and a server 104. The vehicles 101, 102, 103 may be autonomous driving vehicles.

[0028] In one embodiment, vehicles 101, 102, and 103 can exchange data with server 104. For example, vehicles 101, 102, and 103 can transmit sensing data to server 104, obtain pedestrian object data based on the sensing data, and perform calculations based on the pedestrian object data to obtain target intention data for each target pedestrian object in the cluster. For example, the server can determine vehicle traffic strategy data based on the cluster intention data obtained through calculation. The vehicle traffic strategy data is used to indicate the traffic strategy of vehicles 101, 102, and 103. The traffic strategy may include, for example, whether the vehicle passes or not passes. Based on the vehicle traffic strategy data, control data for controlling the vehicle's travel can be determined. The control data may be used, for example, to control the vehicle's chassis wire control system.

[0029] In another example, the vehicles 101, 102, and 103 may perform data processing, wherein the vehicle-mounted system of the current vehicle may have a data processing function, and the vehicle-mounted system may perform calculations based on the reference data of the vehicle to obtain the control data of the current vehicle.

[0030] Exemplarily, a vehicle includes electronic equipment, including but not limited to a vehicle-mounted system, which can execute the pedestrian object intention determination method of the embodiment of the present disclosure. The vehicle-mounted system can include a chassis wire control system.

[0031] The vehicle may be an autonomous vehicle.

[0032] It should be understood that Figure 1 The number of vehicles and servers in the embodiment is merely illustrative. Any number of vehicles and servers may be provided as needed.

[0033] It should be noted that the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved in the technical solution of this disclosure are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0034] In the technical solution disclosed herein, the user's authorization or consent is obtained before obtaining or collecting the user's personal information.

[0035] The present disclosure provides a method for determining the intention of a pedestrian object. Figure 1 The system architecture of Figures 2 to 6 The pedestrian object intention determination method according to the exemplary embodiment of the present disclosure is described. The pedestrian object intention determination method of the embodiment of the present disclosure can be, for example, Figure 1The server 104 is shown to execute.

[0036] Figure 2 A flowchart of a method for determining the intention of a pedestrian object according to an embodiment of the present disclosure is schematically shown.

[0037] like Figure 2 As shown, the pedestrian object intention determination method 200 of the embodiment of the present disclosure may include, for example, operations S210 to S230.

[0038] In operation S210 , pedestrian object data is clustered to obtain at least one cluster.

[0039] Each cluster includes at least one target pedestrian object data.

[0040] For example, sensing data on the road can be obtained through sensing equipment installed in the vehicle, and pedestrian object data can be obtained from the sensing data. The following will be explained using an autonomous vehicle as an example.

[0041] In operation S220 , for any cluster, intention reference data of each target pedestrian object data is determined to obtain at least one initial intention reference data.

[0042] The target pedestrian object data can be understood as the individual pedestrian object data of the cluster. For example, in a scene with 30 pedestrians standing on the road, the pedestrian object data includes 30 individual pedestrian object data. The pedestrian object data can be clustered to obtain, for example, three clusters, each containing 6, 12, and 12 target pedestrian object data, respectively.

[0043] In operation S230 , cluster intention data of the cluster is determined based on at least one initial intention reference data as target intention data of each target pedestrian object data of the cluster.

[0044] It should be noted that the following two phenomena exist in actual scenarios.

[0045] 1) Pedestrians' behaviors affect each other. For example, if a pedestrian P1 moves slowly, the speed of pedestrian P2 behind P1 is limited by the speed of pedestrian P1, and pedestrian P2 also moves slowly.

[0046] 2) Pedestrians with the same intention are more concentrated. For example, pedestrians P3 and P4 intend to pass through. Pedestrian P5, who also intends to pass through, will move in the direction of pedestrians P3 and P4, forming a pedestrian cluster with the same intention as pedestrians P3 and P4.

[0047] Due to the above phenomenon, pedestrians clustered together are more likely to have the same intention. Therefore, the pedestrian object intention determination method of the embodiment of the present invention can divide the pedestrian object data by clustering the pedestrian object data to obtain at least one cluster cluster, and the probability of at least one target pedestrian object data in each cluster cluster having the same intention is relatively high; by determining the intention reference data of each target pedestrian object data for each cluster cluster, the at least one initial intention reference data obtained can be used to determine the cluster intention data of the cluster cluster, thereby improving the accuracy of pedestrian object intention determination; the cluster intention data can be used as the target intention data of each target pedestrian object data in the cluster cluster. For any cluster cluster, there is no need to determine the corresponding target intention data of each target pedestrian object data based on the initial intention reference data of each target pedestrian object data, which can effectively reduce the computing resources for determining the object intention.

[0048] According to another embodiment of the present disclosure, the pedestrian object intention determination method may further include: determining vehicle traffic strategy data based on the clustered intention data; and determining control data for controlling the vehicle's travel based on the vehicle traffic strategy data.

[0049] The vehicle traffic strategy data can be understood as data that characterizes the vehicle's traffic strategy. The traffic strategy may include, for example, whether the vehicle passes or not. The control data may be used, for example, to control the vehicle's chassis wire control system. It can be understood that the vehicle's traffic strategy is related to the traffic intention of the pedestrian. The pedestrian intention determination method of the embodiment of the present disclosure determines the vehicle traffic strategy data based on the clustered intention data; and the control data determined based on the vehicle traffic strategy data can control the vehicle's travel based on the pedestrian's traffic intention, which is safer and more intelligent.

[0050] Figure 3A A schematic diagram of obtaining at least one cluster in a method for determining pedestrian object intention according to another embodiment of the present disclosure is schematically shown.

[0051] like Figure 3A As shown, the following embodiments can be used to cluster pedestrian object data to obtain a specific example of at least one cluster.

[0052] In operation S311 , for any individual pedestrian object data Pi in the pedestrian object data Pd, a distance parameter between the individual pedestrian object data Pi and other individual pedestrian object data in the pedestrian object data Pd is determined.

[0053] exist Figure 3AIn the example, the pedestrian object data Pd includes individual pedestrian object data P1 to individual pedestrian object data Pn, totaling n individual pedestrian object data. For any individual pedestrian object data Pi, a total of n-1 distance parameters including distance parameter Di_1 to distance parameter Di_n (excluding distance parameter Di_i) between the individual pedestrian object data Pi and other individual pedestrian object data can be determined.

[0054] The distance parameter may be understood as a parameter characterizing the distance between individual pedestrian object data. The distance parameter may be determined based on an image coordinate system, for example.

[0055] In operation S312 , the cluster Ci corresponding to the individual pedestrian object data Pi is determined according to the distance parameter and the distance threshold Thd.

[0056] The distance threshold can be determined based on vehicle traffic conditions. For example, if the speed of cluster C closest to vehicle X is v, and the distance D between cluster C and vehicle X satisfies the requirement that vehicle X can successfully pass the road at its current speed without colliding with cluster C, the distance D between cluster C and vehicle X can be used as the vehicle traffic condition. In this case, the distance threshold can be distance D.

[0057] It can be understood that the pedestrian object intention determination method of the embodiment of the present disclosure can obtain target intention data of the target pedestrian object data, and the target intention data represents the passing intention of the target pedestrian object data. The passing intention of the target pedestrian object data includes, for example, passing or not passing.

[0058] Figure 3B The figure schematically shows a schematic diagram of determining a distance threshold according to vehicle traffic conditions.

[0059] like Figure 3B As shown, for example, along a direction A perpendicular to the vehicle's travel direction, the distance Dxc between the individual pedestrian object data Px closest to the vehicle Car and the vehicle Car is denoted by the distance Dxc. The distance Dp that satisfies the vehicle Car's passageway Ro and does not conflict with the individual pedestrian object Px can be determined based on the speeds of the individual pedestrian object Px and the vehicle Car. When the distance Dxc between the individual pedestrian object data Px and the vehicle Car is greater than the distance Dp, the vehicle can pass smoothly.

[0060] In the above example, Dp can be used as the distance threshold.

[0061] The distance threshold determined according to the vehicle's passage conditions can adapt to the following practical application scenarios: the distance between the vehicle and the nearest pedestrian does not meet the passage conditions, but the distance between the vehicle and another pedestrian meets the passage conditions. At this time, the vehicle can wait for the nearest pedestrian to pass before passing, and will not conflict with other pedestrians. It can be understood that among the multiple clusters determined according to the distance threshold Dp, the distance between different clusters is greater than the distance threshold Dp. Similar to the above practical application scenarios, for pedestrians corresponding to the cluster closest to the vehicle, the vehicle's passage strategy can be determined based on the clustering intention data of the cluster and the vehicle's passage conditions. For example, the clustering intention data of the cluster closest to the vehicle is passage, and the distance between the cluster and the vehicle does not meet the passage conditions, then the vehicle's passage strategy is not to pass. The vehicle can wait for the nearest cluster to pass before passing, and will not conflict with other clusters.

[0062] According to the pedestrian object intention determination method of the embodiment of the present disclosure, the distance threshold determined according to the vehicle traffic conditions can also ensure that when determining the vehicle's traffic decision, there will be no inconsistent vehicle traffic decisions, especially in application scenarios with large pedestrian traffic. Specifically, the clustering intention data of the cluster cluster serves as the target intention data of each target pedestrian object data of the cluster cluster. When the distance threshold is the distance Dp, the target intention data of the target pedestrian object data classified into the same cluster cluster are the same, thereby making the subsequent vehicle traffic strategies for the target pedestrian object data of the cluster cluster consistent. At any moment, the unique traffic decision of the vehicle can be determined based on the clustering intention data of the cluster cluster closest to the vehicle and the traffic conditions of the vehicle at the current moment.

[0063] The pedestrian object intention determination method of the disclosed embodiment uses the distance parameter between any individual pedestrian object data and other individual pedestrian object data as the basis for clustering the individual pedestrian object data. The clusters corresponding to the individual pedestrian objects determined based on the distance parameter and the distance threshold are more consistent with actual scenarios where the distances between clustered pedestrians are small. Because clustered pedestrians are more likely to have the same intention, the subsequent use of the clustered intent data of the cluster as the target intent data for each target pedestrian object data in the cluster is more accurate.

[0064] For example, in the above embodiment, determining the vehicle passage strategy data based on the clustering intention data may include, for example, determining the vehicle passage strategy data based on the clustering intention data and the vehicle passage condition data. The vehicle passage condition data may include, for example, the above-mentioned: the speed of cluster C closest to vehicle X is v, and the distance D between cluster C and vehicle X satisfies the requirement that vehicle X can smoothly pass the road at its current speed without colliding with cluster C. In this case, the distance D between cluster C and vehicle X may be used as the vehicle passage condition.

[0065] Figure 4 A schematic diagram of obtaining at least one initial intention reference data in a method for determining intention of a pedestrian object according to yet another embodiment of the present disclosure is schematically shown.

[0066] The pedestrian object data is obtained based on the sensing data Se, which includes image data and wireless signal data. The initial intention reference data includes at least one of the following: speed reference data, position reference data, and form reference data.

[0067] For example, the image data may be obtained by a camera provided on the autonomous vehicle. The wireless signal data may be obtained by a wireless signal device provided on the autonomous vehicle, such as a radar.

[0068] like Figure 4 As shown, the following embodiments can be used to implement, for any cluster, determining the intention reference data of each target pedestrian object data, and obtaining a specific example of at least one initial intention reference data.

[0069] In operation S421 , morphological target detection is performed on the image data G to obtain morphological reference data Rs of each target pedestrian object data.

[0070] The morphological reference data Rs is used to characterize the morphology of the target pedestrian object data, and the morphological reference data Rs is mapped to the initial intention reference data In.

[0071] For example, the morphological reference data may include data representing a pedestrian making a gesture to signal the autonomous vehicle to stop, which corresponds to initial intention reference data for the autonomous vehicle to pass. The morphological reference data may also include data representing a pedestrian's facial orientation toward the autonomous vehicle, which corresponds to initial intention reference data for the autonomous vehicle to pass.

[0072] In operation S422 , at least one of speed reference data Rv and position reference data Rp of the target pedestrian object data is determined based on the wireless signal data G.

[0073] Figure 4 An example is schematically shown in which the initial intention reference data includes speed reference data, position reference data, and form reference data.

[0074] The pedestrian object intention determination method of the embodiment of the present disclosure can obtain initial intention reference data including, for example, speed reference data, position reference data, and shape reference data through image data and wireless signal data. The initial intention reference data is used as a reference for determining intention data from multiple aspects such as speed, position, and shape. The target intention data subsequently obtained based on the initial intention reference data has higher accuracy.

[0075] Figure 5 A schematic diagram of determining clustering intention data of cluster clusters in a pedestrian object intention determination method according to yet another embodiment of the present disclosure is schematically shown.

[0076] like Figure 5 As shown, a specific example of determining clustering intention data of a cluster cluster based on at least one initial intention reference data can be implemented according to the following embodiments.

[0077] In operation S531 , at least one initial intention reference data is classified to obtain at least one intention reference category.

[0078] Figure 5 An example of determining the clustering intention data Ic_i of a certain cluster Ci is schematically shown. Figure 5 It schematically shows the cluster Ci including n initial intention reference data, namely initial intention reference data In_1 to initial intention reference data In_n, and m intention reference categories, namely intention reference categories C1 to intention reference categories Cm, obtained by classifying the initial intention reference data.

[0079] In operation S532 , an intent reference category ratio of each intent reference category is determined based on each intent reference category and at least one initial intent reference data.

[0080] The proportion of intention reference categories represents the ratio of intention reference categories to the total amount of initial intention reference data in the cluster.

[0081] Figure 5 A total of m intention reference category proportions, including intention reference category proportions R1_i to intention reference category proportions Rm_i, are schematically shown.

[0082] In operation S533 , reference data Rc of the cluster Ci is determined according to the intended reference category ratio and the ratio threshold THr.

[0083] In operation S534 , clustering intention data Ic_i is determined based on the reference data Rc of the clusters.

[0084] Exemplarily, the proportion threshold can be set by relevant personnel, or, for a cluster, the intention reference category with the highest numerical value can be selected as the proportion threshold Thr. When the intention reference category proportion is equal to the proportion threshold, the corresponding intention reference category is used as the clustering intention data of the cluster.

[0085] For example, in a scenario where the initial intent reference data includes morphological reference data, a cluster Ci includes 20 initial intent reference data, including 15 initial intent reference data representing passable traffic and 5 initial intent reference data representing non-passable traffic. After classifying the initial intent reference data, the passable intent reference category C1 and the non-passable intent reference category C2 are obtained. It can be determined that the passable intent reference category R1_i accounts for 0.75, and the non-passable intent reference category R2_i accounts for 0.25. The ratio threshold can be set to 0.75. At this time, the reference data of the cluster represents "passable traffic," and the clustered intent data of the cluster can be determined to be passable traffic.

[0086] For example, for a scenario where the initial intention reference data includes speed reference data, a cluster includes 20 initial intention reference data. The speed reference data can be segmented and classified according to their values ​​to obtain 15 initial intention reference data with speed values ​​within the 0.5m / s-1m / s segment and 5 initial intention reference data with speed values ​​within the 0-0.5m / s segment. It can be determined that the proportion R1_i of the intention reference category with speed values ​​within the 0.5m / s-1m / s segment is 0.75, and the proportion R2_i of the intention reference category with speed values ​​within the 0-0.5m / s segment is 0.25. The ratio threshold can be set to 0.75, at which point the reference data of the cluster represents "the speed of the cluster is 0.5m / s-1m / s, and the speed direction of the cluster is due north to due west". Alternatively, the maximum or mean value of the corresponding speed segment can be selected as the reference data for the cluster. For example, the maximum value of the corresponding speed segment can be used as the reference data for the cluster. For example, the reference data for the cluster can represent "the speed of the cluster is 1 m / s, and the speed direction of the cluster is due north." For example, a common speed range can be obtained based on historical statistics. If the reference data meets the conditions that the speed is within the common speed range and the speed direction is perpendicular to the vehicle's travel direction, the clustering intent data for the cluster can be determined as passing.

[0087] For the scenario where the initial intention reference data includes position reference data, segmentation and classification can be performed according to the value of the position reference data, which is similar to the scenario where the initial intention reference data includes speed reference data, and will not be repeated here.

[0088] Exemplarily, when the reference data of a cluster includes multiple of morphological reference data, speed reference data, and position reference data, each reference data can be assigned a weight according to the actual scenario, and the clustering intention data of the cluster can be determined based on the weighted sum of the three. The clustering intention data of the cluster and the corresponding target intention data have higher accuracy.

[0089] The pedestrian object intention determination method of the embodiment of the present disclosure can accurately reflect the characteristics of the cluster by classifying the initial intention reference data and determining the reference data of the cluster cluster based on the proportion. The cluster cluster intention data and target intention data determined based on the cluster cluster reference data are more accurate. The cluster cluster intention data serves as the target intention data of each target pedestrian object data of the cluster cluster, which can reduce the resources required to calculate the target intention data of each target pedestrian object.

[0090] Figure 6 A schematic diagram of determining clustering intention data of cluster clusters in a pedestrian object intention determination method according to yet another embodiment of the present disclosure is schematically shown.

[0091] The initial intention reference data includes at least one of the following: speed reference data, position reference data.

[0092] like Figure 6 As shown, a specific example of determining clustering intention data of a cluster cluster based on at least one initial intention reference data can be implemented according to the following embodiments.

[0093] In operation S631 , at least one initial intention reference data is numerically sorted to obtain an intention reference sequence Rk.

[0094] Exemplarily, the position reference data may represent the distance between the position coordinates of the pedestrian object data and the position coordinates of the autonomous driving vehicle along a direction perpendicular to the driving direction of the autonomous driving vehicle.

[0095] In operation S632 , reference data Rc of the clusters are determined according to the intended reference sequence Rk and the order threshold Thk.

[0096] Exemplarily, the order threshold may be set to 1, that is, the initial intention reference data with the highest numerical value in the intention reference sequence is used as the reference data for the clustering cluster.

[0097] In operation S633 , clustering intention data Ic_i of the cluster is determined based on the reference data Rc of the cluster.

[0098] The pedestrian object intention determination method of the embodiment of the present disclosure obtains an intention reference sequence by numerically sorting the initial intention reference data, and the reference data of the cluster cluster determined according to the sequence threshold can better characterize the characteristics of the cluster cluster. The cluster cluster intention data and target intention data determined based on the cluster cluster reference data are more accurate. The cluster cluster intention data serves as the target intention data of each target pedestrian object data of the cluster cluster, which can reduce the resources required for calculating the target intention data of each target pedestrian object.

[0099] Figure 7A block diagram of a pedestrian object intention determination device according to an embodiment of the present disclosure is schematically shown.

[0100] like Figure 7 As shown, the pedestrian object intention determination device 700 of the embodiment of the present disclosure includes, for example, a clustering module 710 , an initial intention reference data determination module 720 , and a clustering intention data determination module 730 .

[0101] A clustering module, configured to cluster the pedestrian object data to obtain at least one cluster, wherein each cluster includes at least one target pedestrian object data;

[0102] an initial intention reference data determination module, configured to determine, for any cluster, the intention reference data of each target pedestrian object data, and obtain at least one initial intention reference data; and

[0103] The clustering intention data determination module is used to determine the clustering intention data of the cluster according to at least one initial intention reference data as the target intention data of each target pedestrian object data of the cluster.

[0104] According to the pedestrian object intention determination device of the embodiment of the present disclosure, the clustering module includes: a distance parameter determination submodule and a cluster cluster determination submodule.

[0105] The distance parameter determination submodule is used to determine, for any individual pedestrian object data in the pedestrian object data, a distance parameter between the individual pedestrian object data and other individual pedestrian object data in the pedestrian object data.

[0106] The cluster determination submodule is used to determine the cluster corresponding to the individual pedestrian object data based on the distance parameter and the distance threshold, wherein the distance threshold is determined according to the vehicle traffic conditions.

[0107] According to the pedestrian object intention determination device of the embodiment of the present disclosure, the pedestrian object data is obtained based on the sensing data, and the sensing data includes image data and wireless signal data; the initial intention reference data includes at least one of the following: speed reference data, position reference data and form reference data; the initial intention reference data determination module includes at least one of the following: a first determination submodule and a second determination submodule.

[0108] The first determination submodule is used to perform morphological target detection on the image data to obtain morphological reference data of each target pedestrian object data, wherein the morphological reference data is used to characterize the morphology of the target pedestrian object data, and the morphological reference data is mapped to the initial intention reference data.

[0109] The second determining submodule is configured to determine at least one of speed reference data and position reference data of the target pedestrian object data according to the wireless signal data.

[0110] According to the pedestrian object intention determination device of an embodiment of the present disclosure, the clustering intention data determination module includes: a classification submodule, an intention reference category proportion determination submodule, a reference data first determination submodule and a clustering intention data determination submodule.

[0111] The classification submodule is used to classify at least one initial intention reference data to obtain at least one intention reference category.

[0112] The intention reference category ratio determination submodule is used to determine the intention reference category ratio of each intention reference category based on each intention reference category and at least one initial intention reference data, wherein the intention reference category ratio represents the proportion of the intention reference category to the total amount of initial intention reference data in the cluster.

[0113] The first reference data determination submodule is used to determine the reference data of the cluster according to the intended reference category proportion and the ratio threshold.

[0114] The clustering intention data determination submodule is used to determine the clustering intention data based on the reference data of the cluster cluster.

[0115] According to the pedestrian object intention determination device of an embodiment of the present disclosure, the initial intention reference data includes at least one of the following: speed reference data, position reference data; the clustered intention data determination module includes: an intention reference sequence determination submodule, a reference data second determination submodule and a clustered intention data determination submodule.

[0116] The intention reference sequence determination submodule is used to numerically sort at least one initial intention reference data to obtain an intention reference sequence.

[0117] The second reference data determination submodule is used to determine the reference data of the cluster according to the intended reference sequence and the sequence threshold.

[0118] The clustering intention data determination submodule is used to determine the clustering intention data based on the reference data of the cluster cluster.

[0119] According to an embodiment of the present disclosure, the pedestrian object intention determination device further includes: a vehicle traffic strategy data determination module and a control data determination module.

[0120] The vehicle traffic strategy data determination module is used to determine the vehicle traffic strategy data based on the clustering intention data.

[0121] The control data determination module is used to determine the control data for controlling the driving of the vehicle according to the vehicle traffic strategy data.

[0122] It should be understood that the embodiments of the device part of the present disclosure are the same or similar to the embodiments of the method part of the present disclosure, and the technical problems solved and the technical effects achieved are also the same or similar, and the present disclosure will not elaborate on them here.

[0123] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0124] According to an embodiment of the present disclosure, the present disclosure further provides an autonomous driving vehicle, which includes, for example, an electronic device, and the electronic device includes at least one processor and a memory in communication with the at least one processor. The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the above method. For example, the electronic device of the embodiment of the present disclosure is connected to Figure 8 The electronics shown are similar.

[0125] Figure 8 A schematic block diagram of an example electronic device 800 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are provided as examples only and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0126] like Figure 8 As shown, the device 800 includes a computing unit 801, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 802 or a computer program loaded from a storage unit 808 into a random access memory (RAM) 803. Various programs and data required for the operation of the device 800 can also be stored in the RAM 803. The computing unit 801, the ROM 802, and the RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0127] Various components in device 800 are connected to I / O interface 805, including an input unit 806, such as a keyboard, mouse, etc.; an output unit 807, such as various types of displays, speakers, etc.; a storage unit 808, such as a magnetic disk, optical disk, etc.; and a communication unit 809, such as a network card, modem, wireless communication transceiver, etc. The communication unit 809 allows device 800 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0128] The computing unit 801 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 801 performs the various methods and processes described above, such as the pedestrian object intention determination method. For example, in some embodiments, the pedestrian object intention determination method can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 808. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 800 via the ROM 802 and / or the communication unit 809. When the computer program is loaded into the RAM 803 and executed by the computing unit 801, one or more steps of the pedestrian object intention determination method described above can be performed. Alternatively, in other embodiments, the computing unit 801 can be configured to perform the pedestrian object intention determination method by any other suitable means (e.g., via firmware).

[0129] Various embodiments of the systems and techniques described above can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0130] The program code for implementing the method of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device so that when the program code is executed by the processor or controller, the functions / operations specified in the flow chart and / or block diagram are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0131] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in conjunction with an instruction execution system, device or equipment. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can 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 can 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.

[0132] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0133] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0134] Computer systems may include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The client and server relationship arises through computer programs running on the respective computers and having a client-server relationship to each other.

[0135] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved. This is not a limitation herein.

[0136] The above specific embodiments do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure shall be included within the scope of protection of this disclosure.

Claims

1. A method for determining pedestrian object intention, comprising: Clustering pedestrian object data to obtain at least one cluster includes: determining, for any individual pedestrian object data in the pedestrian object data, a distance parameter between the individual pedestrian object data and other individual pedestrian object data in the pedestrian object data; and determining the cluster corresponding to the individual pedestrian object data based on the distance parameter and a distance threshold, wherein the distance threshold is determined according to vehicle traffic conditions; wherein each cluster includes at least one target pedestrian object data; For any one of the clusters, determining initial intention reference data of each target pedestrian object data to obtain at least one initial intention reference data; and According to the at least one initial intention reference data, cluster intention data of the cluster is determined as target intention data of each target pedestrian object data of the cluster.

2. The method according to claim 1, wherein The pedestrian object data is obtained based on sensing data, wherein the sensing data includes image data and wireless signal data; the initial intention reference data includes at least one of the following: speed reference data, position reference data, and form reference data; For any one of the clusters, determining the initial intention reference data of each target pedestrian object data, obtaining at least one initial intention reference data includes at least one of the following: Performing morphological target detection on the image data to obtain morphological reference data of each target pedestrian object data, wherein the morphological reference data is used to characterize the morphology of the target pedestrian object data; as well as At least one of the speed reference data and the position reference data of the target pedestrian object data is determined according to the wireless signal data.

3. The method according to claim 1, wherein The determining, based on the at least one initial intention reference data, the clustering intention data of the clustering cluster comprises: classifying the at least one initial intention reference data to obtain at least one intention reference category; Determining, based on each of the intention reference categories and the at least one initial intention reference data, an intention reference category ratio of each of the intention reference categories, wherein the intention reference category ratio represents a proportion of the intention reference category to the total amount of the initial intention reference data in the cluster; Determining reference data of the cluster according to the reference category proportion and ratio threshold of the intention; and The clustering intention data is determined based on the reference data of the cluster.

4. The method according to claim 1, wherein The initial intention reference data includes at least one of the following: speed reference data, position reference data; The determining, based on the at least one initial intention reference data, the clustering intention data of the clustering cluster comprises: numerically sorting the at least one initial intention reference data to obtain an intention reference sequence; Determining reference data of the cluster according to the intended reference sequence and the sequence threshold; and The clustering intention data is determined based on the reference data of the cluster.

5. The method according to any one of claims 1 to 4, further comprising: Determining vehicle traffic strategy data based on the clustering intention data; as well as Control data for controlling the travel of the vehicle is determined based on the vehicle traffic strategy data.

6. A pedestrian object intention determination device comprising: A clustering module, configured to cluster the pedestrian object data to obtain at least one cluster, wherein each cluster includes at least one target pedestrian object data; an initial intention reference data determination module, configured to determine, for any one of the clusters, initial intention reference data for each of the target pedestrian object data, to obtain at least one initial intention reference data; and a clustering intention data determination module, configured to determine, based on the at least one initial intention reference data, clustering intention data of the cluster as target intention data of each target pedestrian object data of the cluster; Wherein, the clustering module includes: a distance parameter determination submodule, configured to determine, for any individual pedestrian object data in the pedestrian object data, a distance parameter between the individual pedestrian object data and other individual pedestrian object data in the pedestrian object data; and The cluster determination submodule is configured to determine the cluster corresponding to the individual pedestrian object data according to the distance parameter and a distance threshold, wherein the distance threshold is determined according to vehicle traffic conditions.

7. The device according to claim 6, wherein The pedestrian object data is obtained based on sensing data, wherein the sensing data includes image data and wireless signal data; the initial intention reference data includes at least one of the following: speed reference data, position reference data, and form reference data; and the initial intention reference data determination module includes at least one of the following: a first determining submodule, configured to perform morphological target detection on the image data to obtain morphological reference data of each target pedestrian object data, wherein the morphological reference data is used to characterize the morphology of the target pedestrian object data; and The second determining submodule is configured to determine at least one of the speed reference data and the position reference data of the target pedestrian object data according to the wireless signal data.

8. The device according to claim 6, wherein The clustering intention data determination module includes: a classification submodule, configured to classify the at least one initial intention reference data to obtain at least one intention reference category; an intention reference category proportion determination submodule, configured to determine the intention reference category proportion of each intention reference category based on each intention reference category and the at least one initial intention reference data, wherein the intention reference category proportion represents the proportion of the intention reference category to the total amount of the initial intention reference data in the cluster; a first reference data determination submodule, configured to determine the reference data of the cluster according to the proportion of the intended reference category and a ratio threshold; and The clustering intention data determination submodule is used to determine the clustering intention data based on the reference data of the cluster cluster.

9. The device according to claim 6, wherein The initial intention reference data includes at least one of the following: speed reference data, position reference data; The clustering intention data determination module includes: an intention reference sequence determination submodule, configured to numerically sort the at least one initial intention reference data to obtain an intention reference sequence; a second reference data determination submodule, configured to determine the reference data of the cluster according to the intended reference sequence and a sequence threshold; and The clustering intention data determination submodule is used to determine the clustering intention data based on the reference data of the cluster cluster.

10. The apparatus according to any one of claims 6 to 9, further comprising: A vehicle traffic strategy data determination module, configured to determine vehicle traffic strategy data based on the clustering intention data; as well as The control data determination module is used to determine the control data for controlling the driving of the vehicle according to the vehicle traffic strategy data.

11. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 5.

12. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 5.

13. A computer program product comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 5.

14. An autonomous driving vehicle comprising the electronic device according to claim 11.

Images