Method and apparatus for processing point cloud information

Abstract

The present disclosure provides a point cloud information processing method and device, relates to the technical field of artificial intelligence, and in particular to the technical field of automatic driving, intelligent transportation and deep learning. The specific implementation includes: analyzing point cloud information obtained by a vehicle using a radar to detect obstacles to obtain an analysis result; in response to determining that the measurement noise of the obstacle detection of the vehicle is to be updated, determining the point cloud density of the vehicle according to the analysis result; and updating the measurement noise of the vehicle according to the point cloud density. The present disclosure can dynamically determine the measurement noise of the vehicle through the point cloud density per unit area, thereby improving the accuracy and timeliness of updating the measurement noise.

Description

Technical Field

[0001] This disclosure relates to the field of artificial intelligence technology, specifically to autonomous driving, intelligent transportation, and deep learning technologies, and particularly to methods and apparatus for processing point cloud information. Background Technology

[0002] In fields such as autonomous driving, the common approach to tracking the motion trajectory of obstacles is to use the Kalman filter algorithm for calculation.

[0003] During vehicle detection using sensors, including radar, radar point cloud signals can be used as input. Upstream detection algorithms can then identify these radar point clouds as bounding boxes for objects. This allows the vehicle to detect obstacles and respond appropriately, such as braking or honking. Summary of the Invention

[0004] A method, apparatus, electronic device, and storage medium for processing point cloud information are provided.

[0005] According to a first aspect, a method for processing point cloud information is provided, comprising: parsing point cloud information obtained by a vehicle using radar to detect obstacles, and obtaining a parsing result; in response to determining that the measurement noise of the obstacle detection of the vehicle will be updated, determining the point cloud density of the vehicle based on the parsing result; and updating the measurement noise of the vehicle based on the point cloud density.

[0006] According to a second aspect, a point cloud information processing apparatus is provided, comprising: a parsing unit configured to parse point cloud information obtained by a vehicle using radar to detect obstacles, and obtain a parsing result; an execution unit configured to, in response to determining that the measurement noise of the obstacle detection of the vehicle will be updated, determine the point cloud density of the vehicle based on the parsing result; and an updating unit configured to update the measurement noise of the vehicle based on the point cloud density.

[0007] According to a third aspect, 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, the instructions being executed by the at least one processor to enable the at least one processor to perform a method according to any embodiment of a point cloud information processing method.

[0008] According to the fourth aspect, a cloud control platform is provided, including electronic devices as described in the third aspect.

[0009] According to a fifth aspect, a non-transitory computer-readable storage medium is provided storing computer instructions for causing a computer to perform a method according to any embodiment of a point cloud information processing method.

[0010] According to a sixth aspect, a computer program product is provided, including a computer program that, when executed by a processor, implements the method of any embodiment of the method for processing point cloud information.

[0011] According to the scheme disclosed herein, the measurement noise of a vehicle can be dynamically determined by the point cloud density per unit area, thereby improving the accuracy and timeliness of updating the measurement noise. Attached Figure Description

[0012] Other features, objects, and advantages of this disclosure will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0013] Figure 1 This is an exemplary system architecture diagram to which some embodiments of this disclosure can be applied;

[0014] Figure 2a This is a flowchart of an embodiment of a point cloud information processing method according to the present disclosure;

[0015] Figure 2b This is a schematic diagram of another application scenario of the point cloud information processing method according to this disclosure;

[0016] Figure 2c This is a schematic diagram of another application scenario of the point cloud information processing method according to this disclosure;

[0017] Figure 3 This is a schematic diagram of another application scenario of the point cloud information processing method according to this disclosure;

[0018] Figure 4a This is a flowchart of yet another embodiment of the point cloud information processing method according to the present disclosure;

[0019] Figure 4b This is a schematic diagram of another application scenario of the point cloud information processing method according to this disclosure;

[0020] Figure 5 This is a schematic diagram of a structure of an embodiment of a point cloud information processing apparatus according to the present disclosure;

[0021] Figure 6 This is a block diagram of an electronic device used to implement the point cloud information processing method of the embodiments of this disclosure. Detailed Implementation

[0022] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0023] The acquisition, storage, and application of user personal information involved in this technical solution comply with the provisions of relevant laws and regulations, necessary confidentiality measures have been taken, and it does not violate public order and good morals.

[0024] It should be noted that, unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other. This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0025] Figure 1 An exemplary system architecture 100 is shown, which is an embodiment of a point cloud information processing method or a point cloud information processing apparatus to which the present disclosure can be applied.

[0026] like Figure 1 As shown, system architecture 100 may include terminal device 101, network 102, and server 103. Network 102 is used as a medium to provide a communication link between terminal device 101 and server 103. Network 102 may include various connection types, such as wired or wireless communication links or fiber optic cables, etc.

[0027] Users can use terminal device 101 to interact with server 103 via network 102 to receive or send messages, etc. Various communication client applications can be installed on terminal device 101, such as video applications, live streaming applications, instant messaging tools, email clients, social media platform software, etc.

[0028] The terminal device 101 here can be either hardware or software. When the terminal device 101 is hardware, it can be various electronic devices with a display screen, including but not limited to vehicles. When the terminal device 101 is software, it can be installed in the electronic devices listed above. It can be implemented as multiple software programs or software modules (e.g., multiple software programs or software modules used to provide distributed services) or as a single software program or software module. No specific limitations are made here.

[0029] Server 103 can be a server that provides various services, such as a backend server that supports terminal device 101. The backend server can analyze and process the received point cloud information and other data, and feed back the processing results (such as point cloud density) to the terminal device.

[0030] It should be noted that the point cloud information processing method provided in this embodiment can be executed by the server 103 or the terminal device 101, and correspondingly, the point cloud information processing device can be set in the server 103 or the terminal device 101.

[0031] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0032] Continue to refer to Figure 2a The diagram illustrates a flow 200 of an embodiment of a point cloud information processing method according to the present disclosure. The point cloud information processing method includes the following steps:

[0033] Step 201: Analyze the point cloud information obtained by the vehicle using radar to detect obstacles, and obtain the analysis results.

[0034] In this embodiment, the vehicle uses radar to detect obstacles, thus obtaining point cloud information of the obstacles. The point cloud information processing method runs on an execution entity (e.g., Figure 1 The server or terminal device shown can parse the point cloud information to obtain the parsing results. Specifically, the point cloud information may include detection boxes indicating the location and size of obstacles.

[0035] Step 202: In response to determining that the measurement noise for obstacle detection of the vehicle will be updated, the point cloud density of the vehicle is determined based on the analysis result.

[0036] In this embodiment, if the execution entity determines that the measurement noise for obstacle detection of the vehicle needs to be updated, it determines the point cloud density of the vehicle based on the analysis results. Specifically, the execution entity can determine the point cloud density of the vehicle based on the analysis results in various ways. For example, the execution entity can input the analysis results into a preset model (such as various deep neural networks) and obtain the point cloud density output from the preset model. The preset model can then use the analysis results to determine the point cloud density.

[0037] Step 203: Update the measurement noise of the vehicle based on the point cloud density.

[0038] In this embodiment, the aforementioned execution entity can update the vehicle's measurement noise in various ways. For example, the execution entity can input the point cloud density into a specified model and obtain the measurement noise output from the specified model. The specified model can use the point cloud density to determine the measurement noise. The method provided by the above embodiments of this disclosure can dynamically determine the vehicle's measurement noise through the point cloud density per unit area, thereby improving the accuracy and timeliness of updating the measurement noise.

[0039] In some optional implementations of any embodiment of this disclosure, determining the point cloud density of the vehicle based on the parsing result includes: obtaining, in the parsing result, the vertical distance between the vehicle and the nearest edge of the detection box of the target obstacle; obtaining, in the parsing result, the point distance between the corner point of the detection box closest to the vehicle and the target point on the nearest edge; and determining the point cloud density of the target point based on the vertical distance and the point distance, wherein the smaller the point distance, the greater the point cloud density.

[0040] In these implementations, if the executing entity determines that the measurement noise of the obstacle detection for the vehicle needs to be updated, it can obtain the vertical distance between the nearest edge of the detection frame of the vehicle and the target obstacle in the parsing results. Measurement noise is the difference between the vehicle's detection results and the actual data, involving the overall error from the sensor to the detection model (and algorithm).

[0041] In practice, obstacle detection for vehicles generally refers to lidar point cloud detection; therefore, measurement noise can be correlated with point cloud information. Alternatively, in some cases, obstacle detection may also include image detection and other detection methods; therefore, measurement noise can also be the result of analyzing and summarizing the detection results from various methods.

[0042] The nearest edge refers to an edge within the bounding box of the target obstacle, specifically the edge closest to the vehicle. The bounding box is the detection result included in the point cloud information. The target obstacle is any obstacle detected by the vehicle. The nearest edge of the target obstacle may differ depending on the positional relationship between the vehicle and the obstacle. For example, ... Figure 2b As shown, if the vehicle is behind an obstacle, the nearest edge is the edge behind the obstacle's detection frame. Figure 2c As shown, if the vehicle is located on the side of an obstacle, the nearest side is the side of the obstacle that is closest to the vehicle, such as the left side of the obstacle in the figure. "Side" can refer to a positional difference between the center point of the vehicle and the center point of the obstacle that is no greater than a preset difference threshold.

[0043] The aforementioned execution entity can obtain the nearest corner point of the vehicle and the target point on the nearest edge from the parsing results. Then, the execution entity can determine the distance between the nearest corner point and the target point, and use this distance as the point distance. A corner point is a vertex. For example, if the vehicle is behind and to the right of an obstacle, the nearest corner point is the corner point behind and to the right of the obstacle detection box. In practice, the target point can be any point on the nearest edge whose point cloud density needs to be determined.

[0044] When determining the nearest edge and point distance, the vehicle position used can be the position of the vehicle radar, or the position of the vehicle's center of gravity or center, etc.

[0045] The aforementioned execution entity can determine the point cloud density of the target point based on vertical distance and point distance using various methods. For example, the execution entity can input the vertical distance and point distance into a preset model and obtain the point cloud density output from the preset model. This preset model is used to predict the point cloud density of the target point using the vertical distance and point distance.

[0046] These implementations can use the nearest corner point of the obstacle detection box that is closest to the vehicle as a reference point for determining the point cloud density, and add the nearest edge to the point cloud density calculation. This allows them to take advantage of the high point cloud density within the range of the nearest edge and the nearest corner point to improve the accuracy of the point cloud density calculation.

[0047] See also Figure 3 , Figure 3 This is a schematic diagram illustrating an application scenario of the point cloud information processing method according to this embodiment. Figure 3 In the application scenario, the execution entity 301 analyzes the point cloud information obtained by the vehicle using radar to detect obstacles, and obtains the analysis result 302. In response to determining that the measurement noise for obstacle detection of the vehicle will be updated, the execution entity 301 determines the point cloud density 303 of the vehicle based on the analysis result 302. The execution entity 301 updates the measurement noise 304 of the vehicle based on the point cloud density 303.

[0048] Further reference Figure 4a This illustrates a flow 400 of another embodiment of a point cloud information processing method. Flow 400 includes the following steps:

[0049] Step 401: Analyze the point cloud information obtained by the vehicle using radar to detect obstacles, and obtain the analysis result.

[0050] Step 402, in response to determining that the measurement noise for obstacle detection of the vehicle will be updated, the vertical distance between the vehicle and the nearest edge of the target obstacle's detection box in the parsing results is obtained.

[0051] Step 403: In the parsing results, obtain the point distance between the corner point closest to the vehicle and the target point on the nearest edge of the detection box.

[0052] Step 404: Determine the point cloud density of the target point based on the vertical distance and the point distance, where the smaller the point distance, the greater the point cloud density.

[0053] Step 405: Update the vehicle's measurement noise based on the point cloud density, where the measurement noise is inversely proportional to the square of the point cloud density.

[0054] In this embodiment, the execution entity can update the vehicle's measurement noise based on point cloud density in various ways. For example, the execution entity can input the point cloud density into a preset formula or model to obtain the measurement noise output from the formula or model. The preset formula or model is used to generate measurement noise using point cloud density. Furthermore, the formula or model can indicate that the measurement noise is inversely proportional to the square of the point cloud density.

[0055] In practice, the measurement noise R can be expressed as:

[0056]

[0057] in, It can be a preset coefficient, and its maximum value can be 30π / 180, so it can take values ​​(0, π / 6). This can be related to the number of laser lines (x) emitted by the vehicle's lidar. Specifically, =λπ / 180, where λ = 360 / x. The aforementioned executing entity or other electronic equipment can determine the coefficients based on the accuracy of the sensor (such as lidar) and the detection accuracy (e.g., by using a specified formula or model to determine the coefficients). This represents the point cloud density of the radar point cloud on the surface of the obstacle.

[0058] This embodiment can accurately update the measurement noise using the point cloud density through the inverse square relationship.

[0059] Optionally, the nearest corner point is in the vehicle's non-radar blind zone; the method further includes: acquiring the processing noise of obstacle detection; based on the processing noise, determining the weight of the model prediction result of the vehicle's motion state parameters in the motion state parameters used per unit time, and defining this weight as the first weight; based on the measurement noise, determining the weight of the motion state detection parameter corresponding to the obstacle detection result of the vehicle in the motion state parameters used per unit time, and defining this weight as the second weight; using the first weight and the second weight, weighting the model prediction result and the motion state detection parameter to obtain the motion state parameters used by the vehicle per unit time.

[0060] In these optional implementations, the aforementioned execution entity can acquire the processing noise. Processing noise is the noise in the predicted motion state parameters of the vehicle using a kinematic model; it represents the error of the kinematic model in a filter (such as a Kalman filter) relative to the actual motion in the physical world. The aforementioned execution entity can acquire the processing noise using a preset formula or model, or it can acquire the processing noise from other electronic devices.

[0061] Motion state parameters can include various parameters related to the vehicle's motion state, such as position and velocity. The aforementioned execution entity can determine the weight of the model prediction result of the vehicle's motion state parameters within the total number of motion state parameters used per unit time, based on the processed noise, using various methods. For example, the execution entity can obtain the correspondence between processed noise and a first weight, thereby determining the first weight corresponding to the processed noise. Alternatively, the execution entity can input the processed noise into the formula or model for determining the first weight.

[0062] Motion state detection parameters are the actual motion state parameters analyzed based on the obstacle detection results obtained from the vehicle's sensors.

[0063] The aforementioned executing entity can use various methods to determine the weight of the motion state detection parameters corresponding to the vehicle's obstacle detection results within the total motion state parameters used per unit time, based on the measured noise. For example, the executing entity can obtain the correspondence between the measured noise and the second weight, thereby determining the first weight corresponding to the processed noise. Alternatively, the executing entity can input the processed noise into a formula or model for determining the second weight. For instance, this formula or model can determine the first and second weights based on the processed noise and the measured noise.

[0064] In practice, a unit of time can be, for example, one second. Specifically, this unit of time can be the current unit of time, such as the current second, or a future unit of time, such as the next second.

[0065] Specifically, motion state parameters can be used to determine the parameters of a vehicle's perception system, thereby affecting the vehicle's safety and efficiency.

[0066] These implementation methods can improve the accuracy of determining motion state parameters by measuring and processing noise, and adjusting the parameters that are actually measured and predicted by the model when calculating motion state parameters.

[0067] In some optional implementations of any embodiment of this disclosure, the above-mentioned response to determining that the measurement noise of obstacle detection for the vehicle will be updated includes: in response to the volume of the obstacle being greater than a preset threshold, determining whether to update the measurement noise of obstacle detection for the vehicle; if the vehicle is behind the obstacle, or to the side and rear of the obstacle, then determining that the measurement noise of obstacle detection for the vehicle will be updated.

[0068] In these implementations, the obstacle can be relatively large, such as a large vehicle. If the vehicle's position relative to the obstacle is that it is behind or to the side / rear of the obstacle, an update can be performed; otherwise, if it is to the side, no update is performed. Specifically, when the positional relationship is to the side / rear, the vehicle and obstacle can be in different lanes, and the positional difference between the center point of the vehicle and the center point of the obstacle is greater than a preset difference threshold. When the positional relationship is to the side, the positional difference between the center point of the vehicle and the center point of the obstacle is not greater than a preset difference threshold.

[0069] like Figure 4b As shown in the figure, the vehicle's position, i.e., the position of the vehicle's (main vehicle's) radar, is shown. The vehicle is to the side and rear of the obstacle vehicle, specifically to its left and rear. The nearest edge of the obstacle vehicle's detection box is A. The target point on the nearest edge is b. The nearest corner point is a, with a vertical distance of D and a point distance of x. In the point cloud information, edge A has a larger amount of point cloud information. That is, under normal circumstances, the amount of point cloud information on edge A is greater than that on other edges of the detection box.

[0070] Among these implementation methods, the nearest edge with the largest amount of point cloud information is found by using the positional relationship between the vehicle and the obstacle, which can improve the accuracy of updating measurement noise.

[0071] In some optional implementations of any embodiment of this disclosure, determining the point cloud density of a target point based on vertical distance and point distance includes: determining the sum of squares of vertical distance and point distance; determining the quotient of vertical distance and sum of squares; and determining the point cloud density based on the quotient.

[0072] In these alternative implementations, the aforementioned execution entity can determine the point cloud density in various ways, such as by inputting the quotient into a model that uses the quotient to determine the point cloud density, and obtaining the point cloud density output from the model.

[0073] Specifically, see [link to relevant documentation] Figure 4b Point cloud density of target point b It can be:

[0074]

[0075] when hour, For function The derivative:

[0076]

[0077] These implementations provide a new way to determine point cloud density by accurately determining the sum of the squares of the vertical distance and the point distance.

[0078] Further reference Figure 5 As an implementation of the methods shown in the above figures, this disclosure provides an embodiment of a point cloud information processing device, which is similar to... Figure 2a Corresponding to the method embodiment shown, in addition to the features described below, the device embodiment may also include [features related to...]. Figure 2a The method embodiments shown have the same or corresponding features or effects. This device can be specifically applied to various electronic devices.

[0079] like Figure 5 As shown, the point cloud information processing apparatus 500 of this embodiment includes: a parsing unit 501, an execution unit 502, and an updating unit 503. The parsing unit 501 is configured to parse the point cloud information obtained by a vehicle using radar to detect obstacles, and obtain a parsing result; the execution unit 502 is configured to determine the point cloud density of the vehicle based on the parsing result in response to determining that the measurement noise of the obstacle detection of the vehicle will be updated; the updating unit 503 is configured to update the measurement noise of the vehicle based on the point cloud density.

[0080] In this embodiment, the specific processing of the parsing unit 501, the execution unit 502, and the updating unit 503 of the point cloud information processing device 500, and the resulting technical effects, can be referred to respectively. Figure 2a The relevant descriptions of steps 201, 202, and 203 in the corresponding embodiments will not be repeated here.

[0081] In some optional implementations of this embodiment, the execution unit is further configured to perform the following steps to determine the point cloud density of the vehicle based on the parsing result: obtaining the vertical distance between the vehicle and the nearest edge of the detection box of the target obstacle in the parsing result; obtaining the point distance between the corner point of the detection box closest to the vehicle and the target point on the nearest edge in the parsing result; and determining the point cloud density of the target point based on the vertical distance and the point distance, wherein the smaller the point distance, the greater the point cloud density.

[0082] In some optional implementations of this embodiment, the execution unit is further configured to perform the following actions in response to determining that the measurement noise for obstacle detection of the vehicle should be updated: in response to the volume of the obstacle being greater than a preset threshold, determining whether to update the measurement noise for obstacle detection of the vehicle; if the vehicle is behind the obstacle, or to the side and rear of the obstacle, then determining that the measurement noise for obstacle detection of the vehicle should be updated.

[0083] In some optional implementations of this embodiment, the execution unit is further configured to determine the point cloud density of the target point based on the vertical distance and the point distance in the following manner: determining the sum of squares of the vertical distance and the point distance; determining the quotient of the sum of squares of the vertical distance and the point distance; and determining the point cloud density based on the quotient.

[0084] In some optional implementations of this embodiment, the apparatus further includes an update unit configured to update the measurement noise of the vehicle based on the point cloud density, wherein the measurement noise is inversely proportional to the square of the point cloud density.

[0085] In some optional implementations of this embodiment, the nearest corner point is in the vehicle's non-radar blind zone; the device further includes: a processing unit configured to acquire processing noise of obstacle detection; a first calculation unit configured to determine, based on the processing noise, the weight of the model prediction result of the vehicle's motion state parameters in the motion state parameters used per unit time, and define the weight as a first weight; a second calculation unit configured to determine, based on measurement noise, the weight of the motion state detection parameter corresponding to the obstacle detection result of the vehicle in the motion state parameters used per unit time, and define the weight as a second weight; and a weighting unit configured to use the first weight and the second weight to weight the model prediction result and the motion state detection parameter to obtain the motion state parameters used by the vehicle per unit time.

[0086] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0087] Figure 6 A schematic block diagram of an example electronic device 600 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 may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0088] like Figure 6As shown, device 600 includes a computing unit 601, which can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) 602 or a computer program loaded into random access memory (RAM) 603 from storage unit 608. RAM 603 may also store various programs and data required for the operation of device 600. The computing unit 601, ROM 602, and RAM 603 are interconnected via bus 604. Input / output (I / O) interface 605 is also connected to bus 604.

[0089] Multiple components in device 600 are connected to I / O interface 605, including: input unit 606, such as keyboard, mouse, etc.; output unit 607, such as various types of monitors, speakers, etc.; storage unit 608, such as disk, optical disk, etc.; and communication unit 609, such as network card, modem, wireless transceiver, etc. Communication unit 609 allows device 600 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0090] The computing unit 601 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 601 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose 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 601 performs the various methods and processes described above, such as the point cloud information processing method. For example, in some embodiments, the point cloud information processing method can be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 608. In some embodiments, part or all of the computer program can be loaded and / or installed on device 600 via ROM 602 and / or communication unit 609. When the computer program is loaded into RAM 603 and executed by the computing unit 601, one or more steps of the point cloud information processing method described above can be performed. Alternatively, in other embodiments, the computing unit 601 can be configured to perform the point cloud information processing method by any other suitable means (e.g., by means of firmware).

[0091] Various embodiments of the systems and techniques described above herein 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), systems-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0092] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0093] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0094] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, 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 sound input, voice input, or tactile input).

[0095] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0096] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0097] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0098] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for processing point cloud information, the method comprising: The point cloud information obtained by the vehicle using radar to detect obstacles is analyzed to obtain the analysis results; In response to determining that the measurement noise for obstacle detection of the vehicle will be updated, the point cloud density of the vehicle is determined based on the analysis results; Update the vehicle's measurement noise based on the point cloud density; The step of determining the point cloud density of the vehicle based on the analysis results includes: In the analysis results, the vertical distance between the nearest edge of the vehicle and the vehicle in the detection box of the target obstacle is obtained; In the analysis results, the point distance between the corner point closest to the vehicle and the target point on the nearest edge of the detection box is obtained; The point cloud density of the target point is determined based on the vertical distance and the point distance, wherein the smaller the point distance, the greater the point cloud density.

2. The method according to claim 1, wherein, The response to determining that the measurement noise for obstacle detection of the vehicle will be updated includes: In response to the obstacle's volume being greater than a preset threshold, it is determined whether to update the measurement noise for obstacle detection of the vehicle. If the vehicle is behind or to the side and rear of the obstacle, it is determined that the measurement noise for obstacle detection of the vehicle will be updated.

3. The method according to any one of claims 1-2, wherein, Determining the point cloud density of the target point based on the vertical distance and the point distance includes: Determine the sum of the squares of the vertical distance and the point distance; Determine the quotient of the vertical distance and the sum of squares, and determine the point cloud density based on the quotient.

4. The method according to any one of claims 1-2, wherein, The step of updating the vehicle's measurement noise based on the point cloud density includes: The vehicle's measurement noise is updated based on the inverse ratio of the square of the point cloud density.

5. The method according to any one of claims 1-2, wherein, The nearest corner point is in the vehicle's non-radar blind spot; The method further includes: Acquire the processing noise of the obstacle detection; Based on the noise processing, the weight of the model prediction result of the vehicle's motion state parameters in the motion state parameters used per unit time is determined, and this weight is determined as the first weight. Based on the measured noise, the weight of the motion state detection parameter corresponding to the obstacle detection result of the vehicle in the motion state parameters used in the unit time is determined, and this weight is determined as the second weight. Using the first weight and the second weight, the model prediction result and the motion state detection parameters are weighted to obtain the motion state parameters adopted by the vehicle in the unit time.

6. A point cloud information processing apparatus, the apparatus comprising: The parsing unit is configured to analyze the point cloud information obtained by the vehicle using radar to detect obstacles and obtain the analysis results; An execution unit is configured to, in response to determining that the measurement noise for obstacle detection of the vehicle will be updated, determine the point cloud density of the vehicle based on the analysis result; The update unit is configured to update the measured noise of the vehicle based on the point cloud density; The execution unit is further configured to perform the determination of the point cloud density of the vehicle based on the parsing result in the following manner: In the analysis results, the vertical distance between the nearest edge of the vehicle and the vehicle in the detection box of the target obstacle is obtained; In the analysis results, the point distance between the corner point closest to the vehicle and the target point on the nearest edge of the detection box is obtained; The point cloud density of the target point is determined based on the vertical distance and the point distance, wherein the smaller the point distance, the greater the point cloud density.

7. The apparatus according to claim 6, wherein, The execution unit is further configured to perform the update in response to determining that the measurement noise for obstacle detection of the vehicle will be updated in the following manner: In response to the obstacle's volume being greater than a preset threshold, it is determined whether to update the measurement noise for obstacle detection of the vehicle. If the vehicle is behind or to the side and rear of the obstacle, it is determined that the measurement noise for obstacle detection of the vehicle will be updated.

8. The apparatus according to any one of claims 6-7, wherein, The execution unit is further configured to perform the determination of the point cloud density of the target point based on the vertical distance and the point distance in the following manner: Determine the sum of the squares of the vertical distance and the point distance; Determine the quotient of the vertical distance and the sum of squares, and determine the point cloud density based on the quotient.

9. The apparatus according to any one of claims 6-7, wherein, The updating unit is further configured to perform the updating of the vehicle's measurement noise based on the point cloud density in the following manner: The update unit is configured to update the vehicle's measurement noise according to the inverse ratio of the square of the point cloud density.

10. The apparatus according to any one of claims 6-7, wherein, The nearest corner point is in the vehicle's non-radar blind spot; The device further includes: The processing unit is configured to acquire the processing noise of the obstacle detection; The first calculation unit is configured to determine, based on the processed noise, the weight of the model prediction result of the vehicle's motion state parameters in the motion state parameters used per unit time, and to determine the weight as the first weight. The second calculation unit is configured to determine, based on the measurement noise, the weight of the motion state detection parameter corresponding to the obstacle detection result of the vehicle in the motion state parameter used in the unit time, and to determine the weight as the second weight. The weighting unit is configured to use the first weight and the second weight to weight the model prediction result and the motion state detection parameters to obtain the motion state parameters adopted by the vehicle in the unit time.

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 to enable the at least one processor to perform the method of any one of claims 1-5.

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

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

Images