CN116311149A

CN116311149A - Obstacle filtering method, device and storage medium

Info

Publication number: CN116311149A
Application number: CN202211711282.7A
Authority: CN
Inventors: 司马兵; 刘京凯
Original assignee: Apollo Zhilian Beijing Technology Co Ltd
Current assignee: Apollo Zhilian Beijing Technology Co Ltd
Priority date: 2022-12-29
Filing date: 2022-12-29
Publication date: 2023-06-23

Abstract

The disclosure provides a method, a device and a storage medium for filtering obstacles, relates to the technical field of artificial intelligence, and particularly relates to the technical fields of automatic driving, environment sensing and the like. The specific implementation scheme is as follows: acquiring an obstacle point cloud of at least one obstacle; determining an obstacle meeting a first preset condition among the at least one obstacle as a target obstacle; the first preset condition is used for determining suspected filterable barriers; clustering the obstacle point cloud of the target obstacle, and acquiring semantic information of the obstacle point cloud of the target obstacle; determining whether the target obstacle is a filterable obstacle based on the clustered obstacle point cloud and the semantic information; the target obstacle is filtered in response to the target obstacle being a filterable obstacle. The method disclosed by the invention improves the intelligence, the passing efficiency and the sending power of automatic driving and improves the user experience.

Description

Obstacle filtering method, device and storage medium

Technical Field

The disclosure relates to the technical field of artificial intelligence, in particular to the technical fields of automatic driving, environment sensing and the like, and particularly relates to a barrier filtering method, a barrier filtering device and a storage medium.

Background

In the driving process of an automatic driving vehicle, the obstacle is usually required to be monitored in real time so as to avoid in time. At present, a weeping willow is usually arranged above a lane, and the weeping willow is a filterable obstacle, namely: the weeping willow does not influence the traffic of vehicles, and when the vehicle detects that the obstacle is the weeping willow, the obstacle can be ignored and no corresponding avoidance measures are taken.

However, at present, the weeping willow and the ground are commonly identified as large obstacles, and the automatic driving vehicle cannot effectively distinguish the weeping willow obstacles from other obstacles, so that the filtering of the weeping willow obstacles cannot be realized, and the automatic driving vehicle is usually caused to park and avoid due to the existence of the weeping willow, so that the intelligent performance, the passing efficiency and the power for sending and reaching of the automatic driving are low, and the user experience is influenced.

Disclosure of Invention

The present disclosure provides a method and apparatus for filtering an obstacle.

According to an aspect of the present disclosure, there is provided a method of filtering an obstacle,

acquiring an obstacle point cloud of at least one obstacle;

determining an obstacle meeting a first preset condition among the at least one obstacle as a target obstacle; the first preset condition is used for determining suspected filterable barriers;

clustering the obstacle point cloud of the target obstacle, and acquiring semantic information of the obstacle point cloud of the target obstacle;

determining whether the target obstacle is a filterable obstacle based on the clustered obstacle point cloud and the semantic information;

the target obstacle is filtered in response to the target obstacle being a filterable obstacle.

Among the obstacle filtering method provided by the disclosure, the vehicle can accurately identify the filterable obstacle (such as weeping obstacle) and can filter the filterable obstacle, so that the vehicle does not need to execute avoidance measures (such as parking or detouring) on the filterable obstacle, the intelligent, passing efficiency and power transmission of automatic driving are improved, and user experience is improved. The confidence and reliability of the method for identifying the filterable obstacle are high, so that the situations of false identification and false filtration can be avoided, and the closed-loop capability of the automatic driving vehicle is effectively improved.

According to another aspect of the present disclosure, there is provided an obstacle filtering device,

the acquisition module is used for acquiring an obstacle point cloud of at least one obstacle;

a first determining module, configured to determine, as a target obstacle, an obstacle that satisfies a first preset condition among at least one obstacle; the first preset condition is used for determining suspected filterable barriers;

the processing module is used for carrying out clustering processing on the obstacle point cloud of the target obstacle and acquiring semantic information of the obstacle point cloud of the target obstacle;

the second determining module is used for determining whether the target obstacle is a filterable obstacle or not based on the clustered obstacle point cloud and the semantic information;

and the filtering module is used for responding to the target obstacle as a filterable obstacle and filtering the target obstacle.

According to an aspect of the present disclosure, an electronic device is provided, comprising at least one processor, and

a memory communicatively coupled to the at least one processor; wherein, the liquid crystal display device comprises a liquid crystal display device,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the obstacle filtering method according to the embodiments of the first aspect of the disclosure.

According to an aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform an obstacle filtering method of an embodiment of the first aspect of the present disclosure is presented.

According to an aspect of the present disclosure, a computer program product is presented, comprising a computer program which, when executed by a processor, implements the steps of the obstacle filtering method of the first aspect embodiment of the present disclosure.

It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the disclosure, nor is it intended to be used to limit the scope of the disclosure. Other features of the present disclosure will become apparent from the following specification.

Drawings

The drawings are for a better understanding of the present solution and are not to be construed as limiting the present disclosure. Wherein:

FIG. 1a is a flow chart of a method of obstacle filtering according to one embodiment of the present disclosure;

FIG. 1b is a realistic scene graph according to one embodiment of the present disclosure;

FIG. 1c is a diagram of detecting a point cloud obstacle in the actual scenario of FIG. 1b, without employing the filtering method of the present disclosure, according to one embodiment of the present disclosure;

FIG. 1d is a diagram of detecting a point cloud obstacle in the actual scenario of FIG. 1b after a filtering method according to one embodiment of the present disclosure is employed;

FIG. 2 is a block diagram of an obstacle filtering device according to one embodiment of the disclosure;

fig. 3 is a block diagram of an electronic device used to implement an embodiment of the present disclosure.

Detailed Description

Exemplary embodiments of the present disclosure are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present disclosure to facilitate understanding, and should be considered as merely exemplary. Accordingly, one of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

FIG. 1a is a flow chart of a method of filtering an obstacle according to one embodiment of the disclosure, as shown in FIG. 1a, the method comprising the steps of:

s101, obtaining an obstacle point cloud of at least one obstacle.

Alternatively, the method of the present disclosure may be an obstacle filtering strategy for autopilot performed in a particular scenario. The special scene may be: the vegetation is a luxuriant scene, such as a closed park scene and a natural tourist attraction scene which are luxuriant.

In some embodiments, the obstacle point cloud may be radar acquired on the vehicle.

S102, determining an obstacle meeting a first preset condition as a target obstacle in the at least one obstacle.

In some embodiments, the first preset condition may be used to determine a suspected filterable obstacle, which may include a roadside tree, which may include a tree with branches and leaves, e.g., the filterable obstacle may be a weeping willow obstacle.

The first preset condition may be set based on a shape feature of the weeping willow obstacle, where the shape feature of the weeping willow obstacle may be obtained through statistical analysis of actual scene data of the weeping willow.

In particular, for weeping willows, the weeping willow does not normally contact the ground when being hung from a trunk, but is suspended on the ground, that is, a certain distance exists between the tip of the weeping willow and the ground, and no object exists within the certain distance. Based on the above, when the point cloud of the weeping willow is obtained, only the point cloud corresponding to the ground and the point cloud corresponding to the weeping willow are obtained, and the distance between the ground and the weeping willow tip is not provided with an object, so that the point cloud corresponding to the weeping willow is separated into an upper layer and a lower layer when seen from the front view, and a higher height difference exists between the upper layer and the lower layer.

In addition, for weeping willows, the height is generally high because it is weeping from the trunk.

Based on this, the first preset condition described above may be set to include at least one of:

the obstacle point cloud of the obstacle is separated into an upper layer and a lower layer;

the height of the obstacle is greater than a first threshold;

the obstacle is a non-foreground obstacle;

the total number of obstacle point clouds for the obstacle is less than a second threshold.

Optionally, the aforementioned foreground obstacle is generally regular in size, and may be: vehicles, pedestrians, bicycles, etc.

And, the "total number of obstacle point clouds of the obstacle is smaller than the second threshold" is also set as the first preset condition for the convenience of reducing the subsequent processing pressure. Specifically, when the condition that the total number of obstacle point clouds of the obstacles is smaller than the second threshold value is not met, the obstacles with the larger total number of obstacle point clouds may be selected as the target obstacle, and when the vehicle subsequently processes (such as performing subsequent clustering processing) the target obstacle, the processing pressure of the vehicle is higher, so that the processing efficiency is affected, and other scheduling of the vehicle is also affected. Therefore, by setting the condition that the total number of obstacle point clouds of the obstacle is smaller than the second threshold value to screen the target obstacle, the subsequent processing pressure of the vehicle can be reduced.

As can be seen from the foregoing, in the embodiment of the present disclosure, the first preset condition is set, so that the obstacle can be first coarsely screened based on the first preset condition, and the suspected filterable obstacle (i.e. the target obstacle) is screened out, so that only whether the screened suspected filterable obstacle is the filterable obstacle or not is determined later, thereby reducing the resource and the power consumption.

S103, clustering is carried out on the obstacle point cloud of the target obstacle, and semantic information of the obstacle point cloud of the target obstacle is obtained.

Optionally, in some embodiments, the "clustering the obstacle point cloud of the target obstacle" may include the following steps:

and a, determining the number of surrounding adjacent points of each point in the obstacle point cloud.

Alternatively, the surrounding near points may be: a point at which the distance from the point is smaller than the third threshold value. The third threshold may be predefined.

Alternatively, in some embodiments, the point-to-point distance may be calculated based on the point's Z value. The accuracy of Z value calculation is higher, so that clustering is finer, and the obtained clustering result is more accurate, and therefore, when the filterable obstacle is identified based on the clustered obstacle point cloud, the identification accuracy and reliability can be ensured.

And b, determining points with the number of surrounding adjacent points less than a fourth threshold value as noise points.

Wherein the fourth threshold value may be predefined.

And c, determining points with the number of surrounding neighboring points being greater than or equal to a fourth threshold value as core points.

And d, clustering the noise points.

And e, traversing the surrounding adjacent points of the core points, and clustering the core points and the surrounding adjacent points of the core points.

And f, iteratively calculating other points which are not clustered until clustering processing is carried out on all the points so as to obtain at least one type of clustered point cloud.

In the method, the obstacle point cloud of the target obstacle is clustered, so that whether the target obstacle is a filterable obstacle or not can be accurately analyzed based on the point cloud after the clustering, and the confidence and reliability of obstacle identification are improved.

Optionally, in some embodiments, a class label may be further set for each class of clustered point clouds when performing clustering; the class labels corresponding to the different classes of the clustered point clouds are different, and the class labels are used for distinguishing the different classes of the clustered point clouds, so that the total class number of the clustered point clouds corresponding to the target obstacle can be known easily by counting the total number of the class labels of the target obstacle, and the efficiency is higher when the target obstacle is determined to be a filterable obstacle based on the total class number of the clustered point clouds corresponding to the target obstacle.

For example, the cluster point cloud corresponding to the noise may be 0, and the labels classified for the cluster point clouds of other classes may be: 1. 2, 3..i, i is the total class number of the cluster point clouds corresponding to the target obstacle.

Further, in some embodiments, the semantic information described above may be obtained from other models (e.g., models for determining semantic information). And, it may be that, for each point in the obstacle point cloud of the target obstacle, semantic information corresponding to the point is acquired, where the semantic information is used to indicate an object type to which the point may belong.

S104, determining whether the target obstacle is a filterable obstacle based on the clustered obstacle point cloud and semantic information.

Optionally, in some embodiments, in response to the clustered obstacle point cloud of the target obstacle and the semantic information of the obstacle point cloud of the target obstacle meeting a second preset condition, determining the target obstacle as a filterable obstacle.

Wherein the second preset condition may be used to determine a filterable barrier. It should be noted that, in the embodiment of the present disclosure, the filterable obstacle is substantially a weeping obstacle, and based on the content of the step S102, the characteristics of the weeping obstacle substantially include:

the point cloud is separated into an upper layer and a lower layer;

a certain height difference exists between the point clouds of the upper layer and the lower layer;

the lower layer point cloud is essentially the point cloud corresponding to the ground.

Based on this, in some embodiments, the second preset condition may include:

the clustered obstacle point cloud is separated into an upper layer and a lower layer;

the average height difference (such as average Z value height difference) between the obstacle point clouds after the upper layer and the lower layer of clustering is larger than a fifth threshold value;

in the clustered obstacle point clouds of the lower layer, the point cloud occupancy ratio of the semantic information ground is higher than a sixth threshold value.

Specifically, the above-mentioned "the clustered obstacle point cloud is separated into the upper and lower layers" may be understood as: the clustered obstacle point clouds of the target obstacle comprise two types of clustered point clouds, and the two types of clustered point clouds are separated into an upper layer and a lower layer.

In the above-mentioned "the point cloud ratio of the clustered obstacle point cloud of the lower layer, where the semantic information is the ground is higher than the sixth threshold" may be understood as: most of the clustered obstacle point clouds at the lower layer are point clouds corresponding to the ground, namely, the clustered point clouds at the lower layer are point clouds of the ground.

The above-mentioned "the average height difference between the clustered obstacle point clouds of the upper layer and the lower layer is greater than the fifth threshold" may be understood as: on the premise that the lower-layer cluster point cloud is the ground point cloud, the average geodesic height of the upper-layer cluster point cloud is higher than a fifth threshold value.

When the obstacle point cloud of a certain target obstacle and semantic information thereof meet the second preset condition, the condition that the target obstacle is a weeping obstacle is confirmed, namely, the condition that the passing of vehicles is not influenced is indicated, so that corresponding strategies can be adopted for the target obstacle subsequently to avoid the target obstacle by the vehicles, and the intelligent of automatic driving and the power transmission are improved.

S105, filtering the target obstacle in response to the target obstacle being the filterable obstacle.

Alternatively, in some embodiments, the "filtering target obstacle" may be: the information corresponding to the target obstacle is not sent to the downstream module, for example: and the output label of the target obstacle can be set to false, and the information corresponding to the target obstacle can not be sent to a downstream module after the subsequent message serialization, so that the filtering of the weeping obstacle in the sensing module is completed.

In summary, in the obstacle filtering method provided by the disclosure, the vehicle can accurately identify the filterable obstacle (such as the weeping obstacle) and can filter the filterable obstacle, so that the vehicle does not need to execute avoidance measures (such as parking or detouring) on the filterable obstacle, thereby improving the intelligence, passing efficiency and power of automatic driving and improving user experience. The confidence and reliability of the method for identifying the filterable obstacle are high, so that the situations of false identification and false filtration can be avoided, and the closed-loop capability of the automatic driving vehicle is effectively improved.

For example, fig. 1b is an actual scene graph (photographed by a camera) according to an embodiment of the present disclosure, fig. 1c is a point cloud obstacle detection graph in the actual scene of fig. 1b when a filtering method according to an embodiment of the present disclosure is not adopted, and fig. 1d is a point cloud obstacle detection graph in the actual scene of fig. 1b after a filtering method according to an embodiment of the present disclosure is adopted. As can be seen from comparing fig. 1c and fig. 1d, the filtering method of the present disclosure can successfully filter the weeping willow obstacle without displaying the weeping willow obstacle point cloud on the obstacle detection chart.

Fig. 2 is a block diagram of an obstacle filtering device according to an embodiment of the disclosure, and as shown in fig. 2, the obstacle filtering device 200 includes:

an acquisition module 210, configured to acquire an obstacle point cloud of at least one obstacle;

a first determining module 220, configured to determine, as a target obstacle, an obstacle that satisfies a first preset condition among the at least one obstacle; the first preset condition is used for determining suspected filterable barriers;

the processing module 230 is configured to perform clustering processing on the obstacle point cloud of the target obstacle, and acquire semantic information of the obstacle point cloud of the target obstacle;

a second determining module 240, configured to determine whether the target obstacle is a filterable obstacle based on the clustered obstacle point cloud and the semantic information;

a filtering module 250 for filtering the target obstacle in response to the target obstacle being a filterable obstacle.

In some implementations, the filterable barrier is a weeping barrier.

In some implementations, the first preset condition includes at least one of:

the height of the obstacle is greater than a first threshold;

the obstacle is a non-foreground obstacle;

In some implementations, the processing module is further to:

determining the number of surrounding adjacent points of each point in the obstacle point cloud, wherein the surrounding adjacent points are as follows: a point having a distance from the point less than a third threshold;

determining points with the number of surrounding adjacent points less than a fourth threshold value as noise points;

determining points with the number of surrounding neighboring points being greater than or equal to a fourth threshold value as core points;

clustering the noise points;

traversing the surrounding adjacent points of the core points, and clustering the core points and the surrounding adjacent points of the core points;

and (3) iteratively calculating other points which are not clustered until clustering processing is carried out on all the points so as to obtain at least one type of clustered point cloud.

In some implementations, the distance between points is calculated based on the Z-value of the points.

In some implementations, the apparatus is further to:

setting class labels for each class of cluster point clouds; wherein class labels corresponding to different classes of cluster point clouds are different; the class labels are used for distinguishing different classes of clustered point clouds.

In some implementations, the second determination module is further to:

responding to clustered obstacle point clouds of the target obstacle and semantic information of the clustered obstacle point clouds of the target obstacle to meet a second preset condition, and determining the target obstacle as a filterable obstacle;

the second preset condition includes:

the average height difference between the clustered obstacle point clouds of the upper layer and the lower layer is larger than a fifth threshold value;

In some implementations, the filter module is further to:

and not sending the information corresponding to the target obstacle to a downstream module.

The vehicle can accurately identify the filterable obstacle (such as the weeping obstacle) and can filter the filterable obstacle, so that the vehicle can avoid the filterable obstacle (such as parking or detouring), the intelligent, passing efficiency and power transmission of automatic driving are improved, and user experience is improved.

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

FIG. 3 illustrates a schematic block diagram of an example electronic device 300 that may be used to implement embodiments of the present disclosure. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the disclosure described and/or claimed herein.

As shown in fig. 3, the apparatus 300 includes a computing unit 301 that may perform various suitable actions and processes according to a computer program stored in a Read Only Memory (ROM) 302 or a computer program loaded from a storage unit 308 into a Random Access Memory (RAM) 303. In the RAM 303, various programs and data required for the operation of the device 300 may also be stored. The computing unit 301, the ROM 302, and the RAM 303 are connected to each other by a bus 304. An input/output (I/O) interface 305 is also connected to bus 304.

Various components in device 300 are connected to I/O interface 305, including: an input unit 306 such as a keyboard, a mouse, etc.; an output unit 307 such as various types of displays, speakers, and the like; a storage unit 308 such as a magnetic disk, an optical disk, or the like; and a communication unit 309 such as a network card, modem, wireless communication transceiver, etc. The communication unit 309 allows the device 300 to exchange information/data with other devices via a computer network such as the internet and/or various telecommunication networks.

The computing unit 301 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of computing unit 301 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 301 performs the respective methods and processes described above, such as the obstacle filtering method of the first aspect embodiment or the obstacle filtering method of the second aspect embodiment. For example, in some embodiments, the obstacle filtering method of the first aspect embodiment or the obstacle filtering method of the second aspect embodiment may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as the storage unit 308. In some embodiments, part or all of the computer program may be loaded and/or installed onto the device 300 via the ROM 302 and/or the communication unit 309. When the computer program is loaded into the RAM 303 and executed by the computing unit 301, one or more steps of the obstacle filtering method of the first aspect embodiment or the obstacle filtering method of the second aspect embodiment described above may be performed. Alternatively, in other embodiments, the computing unit 301 may be configured to perform the obstacle filtering method of the first aspect embodiment or the obstacle filtering method of the second aspect embodiment by any other suitable means (e.g. by means of firmware).

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.

Program code for carrying out methods of the present disclosure may be written in any combination of one or more programming languages. These 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 the program code, when executed by the processor or controller, causes the functions/operations specified in the flowchart and/or block diagram to be implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of this disclosure, a machine-readable medium may be a tangible medium that can include or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, 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 disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide for interaction with a user, the systems and techniques described here 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 a user; and a keyboard and pointing device (e.g., a mouse or trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, 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: local Area Networks (LANs), wide Area Networks (WANs), and the internet.

The computer system may include a client and a server. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server may be a cloud server, a server of a distributed system, or a server incorporating a blockchain.

It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps recited in the present disclosure may be performed in parallel or sequentially or in a different order, provided that the desired results of the technical solutions of the present disclosure are achieved, and are not limited herein.

The above detailed description should not be taken as limiting the scope of the present disclosure. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present disclosure are intended to be included within the scope of the present disclosure.

Claims

1. A method of filtering an obstacle, comprising:

acquiring an obstacle point cloud of at least one obstacle;

2. The method of claim 1, wherein the filterable barrier comprises a roadside tree comprising a tree having branches and leaves.

3. The method of claim 2, wherein the first preset condition comprises at least one of:

the height of the obstacle is greater than a first threshold;

the obstacle is a non-foreground obstacle;

4. The method according to claim 1 or 2, wherein the clustering the obstacle point cloud of the target obstacle comprises:

clustering the noise points;

5. The method of claim 4, wherein the point-to-point distance is calculated based on the Z-value of the point.

6. The method according to claim 4, wherein the method further comprises:

setting class labels for each class of cluster point clouds; wherein class labels corresponding to different classes of cluster point clouds are different.

7. The method of claim 2, wherein the determining whether the target obstacle is a filterable obstacle based on the clustered obstacle point cloud and the semantic information comprises:

the second preset condition includes:

8. The method of claim 1, wherein the filtering the target obstacle comprises:

9. An obstruction filtering device, comprising:

10. The device of claim 9, wherein the filterable barrier comprises a roadside tree comprising a tree having branches and leaves.

11. The apparatus of claim 9, wherein the first preset condition comprises at least one of:

the height of the obstacle is greater than a first threshold;

the obstacle is a non-foreground obstacle;

12. The apparatus of claim 9 or 10, wherein the processing module is further configured to:

clustering the noise points;

13. The apparatus of claim 12, wherein the point-to-point distance is calculated based on the Z-value of the point.

14. The apparatus of claim 12, wherein the apparatus is further configured to:

15. The apparatus of claim 10, wherein the second determination module is further configured to:

the second preset condition includes:

16. The apparatus of claim 9, wherein the filter module is further configured to:

17. An electronic device, comprising:

at least one processor; and

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-8.

18. A non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the method of any one of claims 1-8.

19. A computer program product comprising a computer program which, when executed by a processor, implements the steps of the method according to any of claims 1-8.

20. A vehicle comprising the electronic device of claim 10.