CN109633688B - LiDAR obstacle recognition method and device - Google Patents

Abstract

The present application provides a laser radar obstacle identification method and device, the method includes obtaining the information of obstacles to be identified in continuous N+1 frames around the unmanned vehicle scanned by the laser radar; Whether the obstacle to be identified enters the lidar blind area; according to the length of the obstacle to be identified entering the lidar blind area, the laser point cloud of the obstacle to be identified in the Nth frame is truncated; according to the truncated The obstacle laser point cloud is matched with the laser point cloud of the obstacle to be identified in the N+1th frame; the laser point cloud of the obstacle to be identified in the N+1th frame is completed; The laser point cloud of the obstacle to be identified is used for obstacle identification. It can identify some obstacles that enter the lidar blind area, avoiding the risk of collision caused by misidentifying their length and distance, and effectively improving the driving safety of unmanned vehicles.

Description

LiDAR obstacle recognition method and device

【Technical field】

The present application relates to the field of automatic control, in particular to a laser radar obstacle recognition method and device.

【Background technique】

In unmanned vehicles, multiple types of sensors are integrated: GPS-IMU (inertial measurement unit, Inertial Measurement Unit) integrated navigation module, cameras, lidar, millimeter wave radar and other sensors.

During the driving process of unmanned vehicles, it mainly relies on lidar to detect obstacles, and lidar can realize accurate shape perception of obstacles. However, since the lidar is installed at the front center of the top of the unmanned vehicle, its vertical field of view ranges from plus or minus 15 degrees. Blocked by the body of the unmanned vehicle, there is a blind area of vision, generally in the range of 6-7m. Therefore, the obstacle detection of existing unmanned vehicles can only detect obstacles that appear in the field of view of the lidar. As the unmanned vehicle advances, the obstacle is likely to enter the blind area of the lidar, causing the obstacle to be recognized incorrectly or lost. The self-driving vehicle misidentified the distance, identifying it as a car outside the blind spot. This affects the decision-making of the vehicle decision-making control module of the unmanned vehicle, which in turn leads to the risk of collision.

【Content of invention】

Aspects of the present application provide a laser radar obstacle recognition method and device, which are used to solve the problem of obstacle recognition errors caused by obstacles partially entering the laser radar blind zone.

In one aspect of the present application, there is provided a lidar obstacle recognition method, including:

Obtain the information of obstacles to be identified in consecutive N+1 frames around the unmanned vehicle scanned by the lidar;

According to the information of the obstacle to be identified in each frame in the first N frames in the N+1 frame, it is judged whether the obstacle to be identified in the N+1 frame enters the lidar blind area; if it enters the lidar blind area truncating the laser point cloud of the obstacle to be identified in the Nth frame according to the length of the obstacle to be identified entering the lidar blind zone;

According to the truncated laser point cloud of the obstacle to be identified and the laser point cloud of the obstacle to be identified in the N+1th frame are matched; if the matching is successful, the said to be identified in the N+1th frame Obstacle laser point cloud for completion;

Obstacle identification is performed according to the completed laser point cloud of the obstacle to be identified.

According to the above-mentioned aspect and any possible implementation, an implementation is further provided. Obtaining the information of obstacles to be identified in consecutive N+1 frames around the unmanned vehicle scanned by the lidar includes:

Obtain the point cloud of the obstacle to be identified in each frame of the first N frames in the N+1 frame, and match the point cloud of the obstacle to be identified in each frame of the first N frames in the N+1 frame , to determine the corresponding relationship of the obstacles to be recognized in each frame.

According to the above aspect and any possible implementation manner, an implementation manner is further provided, according to the information of the obstacle to be identified in each frame in the first N frames in the N+1 frame, it is judged that the N+th Whether the obstacle to be identified in 1 frame enters the lidar blind area includes:

Through the point cloud of the obstacle to be identified in each frame in the first N frames, determine the position, moving speed and moving direction of the obstacle to be identified in the Nth frame and the range of the lidar blind zone, and determine the N+th The length of the obstacle to be recognized entering the blind zone of the lidar in one frame.

According to the above aspect and any possible implementation, an implementation is further provided, truncating the laser point cloud of the obstacle to be identified in the Nth frame according to the length of the obstacle to be identified entering the blind zone of the lidar include:

The laser point cloud of the obstacle to be identified in the Nth frame is truncated according to the three-dimensional model of the blind area and the length of the identified obstacle entering the lidar blind area.

According to the above aspect and any possible implementation, an implementation is further provided. Complementing the laser point cloud of the obstacle to be identified in the N+1th frame includes:

If the obstacle to be identified in the Nth frame does not enter the lidar blind area, the obstacle to be identified in the N+1th frame enters the lidar blind area according to the laser point cloud of the obstacle to be identified in the Nth frame Part of the laser point cloud is completed;

If the obstacle to be identified in the Nth frame has entered the lidar blind area, enter the obstacle to be identified in the N+1th frame according to the completed laser point cloud of the obstacle to be identified in the Nth frame The laser point cloud of the part of the lidar blind area is completed.

Another aspect of the present application provides a laser radar obstacle recognition device, including:

The obtaining module is used to obtain the information of the obstacles to be identified in continuous N+1 frames around the unmanned vehicle scanned by the lidar;

A truncation module, configured to determine whether the obstacle to be identified in the N+1th frame enters the lidar blind spot according to the information of the obstacle to be identified in each of the first N frames in the N+1 frame; If entering the lidar blind area, the laser point cloud of the obstacle to be identified in the Nth frame is truncated according to the length of the obstacle to be identified entering the lidar blind area;

The completion module is used to match the truncated laser point cloud of the obstacle to be identified with the laser point cloud of the obstacle to be identified in the N+1th frame; if the matching is successful, the N+1th frame Complete the laser point cloud of the obstacle to be identified in ;

The identification module is configured to perform obstacle identification according to the completed laser point cloud of the obstacle to be identified.

According to the above aspect and any possible implementation manner, an implementation manner is further provided, and the acquisition module is specifically used for:

Obtain the point cloud of the obstacle to be identified in each frame of the first N frames in the N+1 frame, and match the point cloud of the obstacle to be identified in each frame of the first N frames in the N+1 frame , to determine the corresponding relationship of the obstacles to be recognized in each frame.

According to the above aspects and any possible implementation, an implementation is further provided, the truncation module is specifically used for:

Through the point cloud of the obstacle to be identified in each frame in the first N frames, determine the position, moving speed and moving direction of the obstacle to be identified in the Nth frame and the range of the lidar blind zone, and determine the N+th The length of the obstacle to be recognized entering the blind zone of the lidar in one frame.

According to the above aspect and any possible implementation manner, an implementation manner is further provided, and the truncation module is specifically further configured to:

The laser point cloud of the obstacle to be identified in the Nth frame is truncated according to the three-dimensional model of the blind area and the length of the identified obstacle entering the lidar blind area.

According to the above aspect and any possible implementation manner, an implementation manner is further provided, the completion module is specifically used for:

If the obstacle to be identified in the Nth frame does not enter the lidar blind area, the obstacle to be identified in the N+1th frame enters the lidar blind area according to the laser point cloud of the obstacle to be identified in the Nth frame Part of the laser point cloud is completed;

If the obstacle to be identified in the Nth frame has entered the lidar blind area, enter the obstacle to be identified in the N+1th frame according to the completed laser point cloud of the obstacle to be identified in the Nth frame The laser point cloud of the part of the lidar blind area is completed.

Another aspect of the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the above method is implemented.

It can be seen from the technical solution that the embodiment of the present application can identify some obstacles entering the blind area of the lidar, avoiding the risk of collision caused by misidentifying their length and distance.

【Description of drawings】

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings that need to be used in the embodiments or the description of the prior art. Obviously, the accompanying drawings in the following description are of the present application For some embodiments, those of ordinary skill in the art can also obtain other drawings based on these drawings without paying creative efforts.

FIG. 1 is a schematic flow chart of a lidar obstacle recognition method provided by an embodiment of the present application;

FIG. 2 is a schematic structural diagram of a lidar obstacle recognition device provided by an embodiment of the present application;

Figure 3 shows a block diagram of an exemplary computer system/server 012 suitable for use in implementing embodiments of the present invention.

【Detailed ways】

In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments It is a part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of this application.

Fig. 1 is a schematic diagram of a lidar obstacle recognition method provided by an embodiment of the present application, as shown in Fig. 1, including the following steps:

Step S11, obtaining the information of obstacles to be identified in consecutive N+1 frames around the unmanned vehicle scanned by the lidar;

Step S12, according to the information of the obstacle to be identified in each frame of the first N frames in the N+1 frame, it is judged whether the obstacle to be identified in the N+1th frame enters the lidar blind area; if it enters The lidar blind zone, truncating the laser point cloud of the obstacle to be identified in the Nth frame according to the length of the obstacle to be identified entering the lidar blind zone;

Step S13: Match the truncated laser point cloud of the obstacle to be identified with the laser point cloud of the obstacle to be identified in frame N+1; Complete the laser point cloud of the obstacles to be identified;

Step S14 , performing obstacle identification according to the completed laser point cloud of the obstacle to be identified.

In a preferred implementation of step S11,

The obstacle recognition method of this embodiment is applied in the technical field of unmanned driving. In unmanned driving, unmanned vehicles need to be able to automatically identify obstacles in the road, so as to make timely decisions and control during vehicle driving, so as to facilitate the safe driving of vehicles. The execution body of the obstacle recognition method in this embodiment can be an obstacle recognition device, which can be obtained by integrating multiple modules, and the obstacle recognition device can be specifically set in an unmanned vehicle to Human-driven vehicles are controlled.

The information of the obstacle to be identified in this embodiment can be obtained by laser radar scanning. The specification of lidar can adopt 16 lines, 32 lines or 64 lines and so on. The higher the number of lines, the greater the unit energy density of the lidar. In this embodiment, the laser radar mounted on the current vehicle rotates 360 times per second, and scans the information of obstacles to be identified around the current vehicle, which is a frame of information of obstacles to be identified. The information of the obstacle to be identified in this embodiment may include a point cloud of the obstacle to be identified and a reflection value of the obstacle to be identified. There may be one or more obstacles to be identified around the current vehicle. After the lidar scan, the center of mass position of the current vehicle can be used as the origin of the coordinate system, and the two directions parallel to the horizontal plane are taken as the x direction and the y direction as the length direction and the width direction, and the direction perpendicular to the ground is the z direction , as the height direction. Then, the obstacle to be identified can be identified in the coordinate system according to the relative position and distance of each point in the point cloud of the obstacle to be identified and the origin.

Preferably, the obstacles to be identified in each frame of the first N frames in the N+1 frame are matched, the corresponding relationship of the obstacles to be identified in each frame is determined, and the obstacles to be identified in each frame are determined. identify. Through the matching operation, it is possible to avoid the problem that the same obstacle to be recognized in different frames is recognized as a different obstacle when the obstacle to be recognized fails to be tracked in a complex scene (for example, there are multiple obstacles).

Preferably, the identification is performed after acquiring the information of the obstacle to be identified in each frame around the unmanned vehicle scanned by the lidar.

In this way, in the point cloud of each obstacle to be identified, the point cloud of each obstacle to be identified can be obtained according to the relative position of each point in each obstacle to be identified and the current vehicle. In addition, lidar can also detect the reflection value of each point in each obstacle to be identified. In practical applications, the coordinate system can also take the center of mass position of the lidar as the origin, and the other directions remain unchanged. The value of N in this embodiment may be selected according to actual requirements.

In a preferred implementation of step S12,

According to the information of the obstacle to be identified in each frame of the first N frames in the N+1 frame, it is judged whether the obstacle to be identified in the N+1 frame enters the lidar blind area; if the N+1 frame The obstacle to be identified enters the lidar blind area, and the laser point cloud of the obstacle to be identified in the Nth frame is truncated according to the length of the obstacle to be identified entering the lidar blind area;

Preferably, in each frame of the previous N frames, the point cloud of each obstacle to be identified is obtained, and matched with the point cloud of the obstacle to be identified in the previous frame, to determine that the same obstacle to be identified is in Point clouds in each of the previous N frames.

Through the point clouds of the same obstacle to be identified in each frame in the previous N frames, determine the position of the obstacle to be identified in the Nth frame, with the current center of mass position of the unmanned vehicle as the origin in the coordinate system, Movement speed and direction of movement.

It is judged whether the obstacle to be identified in the N+1th frame enters the lidar blind zone. In this embodiment, the laser radar blind zone is within 8 meters of the current unmanned vehicle's centroid position as the origin.

If in the Nth frame, the obstacle to be identified is already located at the edge of the lidar blind zone, and the relative speed between the unmanned vehicle and the obstacle to be identified is a positive value, then the obstacle to be identified in the N+1 frame Identify obstacles entering the lidar blind zone. Since the lidar scan cycle is fast, such as 360 times per second, within 1/360 second, the obstacle to be identified can only partially enter the blind area of the lidar, which causes the lidar to only acquire the obstacle to be identified part of the laser point cloud of the object. Part of the point cloud of the obstacle to be identified in the N+1th frame and the point cloud distribution of the obstacle to be identified in the Nth frame have changed, then the matching will fail during the matching process, and the N+1th The obstacle to be identified in the frame is determined to be a new obstacle, rather than the obstacle to be identified in the Nth frame partially entering the lidar blind area.

Therefore, according to the position, moving speed and moving direction of the obstacle to be identified in the coordinate system with the current center of mass position of the unmanned vehicle as the origin in the Nth frame, and the range of the lidar blind zone, determine the N+1th frame The length of the obstacle to be identified entering the blind zone of the lidar.

Cut off the laser point cloud of the obstacle to be identified in the Nth frame according to the length of the obstacle to be identified entering the blind zone of the laser radar in the N+1th frame, so as to enter the blind zone of the laser radar in the N+1th frame Match the laser point cloud of the obstacle to be identified.

Preferably, the laser point cloud of the obstacle to be identified in the Nth frame is intercepted into the part of the length of the lidar blind zone, and the interception takes into account the three-dimensional model of the lidar being blocked by the body of the unmanned vehicle, rather than a simple Cut vertically.

In a preferred implementation of step S13,

Matching the truncated laser point cloud of the obstacle to be identified with the laser point cloud of the obstacle to be identified in the N+1th frame;

Wherein, the laser point cloud obtained by truncating the laser point cloud of the obstacle to be identified in the Nth frame is the predicted laser point cloud of the obstacle to be identified in the N+1th frame, by combining with the N+1th Matching the laser point cloud of the actual obstacle to be identified in the frame can improve the matching accuracy and success rate.

If the matching is successful, the laser point cloud of the obstacle to be identified in the N+1th frame is completed; that is, the laser point of the part where the obstacle to be identified in the N+1th frame enters the blind area of the lidar Cloud completes.

Preferably, the completion is based on the laser point cloud of the obstacle to be identified in the Nth frame to complete the laser point cloud of the part where the obstacle to be identified in the N+1th frame enters the blind area of the lidar . Since the completion is performed on the basis of the laser point cloud of the obstacle to be recognized in the N+1th frame, the completed laser point cloud is more in line with the shape of the actual obstacle.

In a preferred implementation of step S14,

Obstacle identification is performed according to the completed laser point cloud of the obstacle to be identified.

Preferably, according to the completed laser point cloud of the obstacle to be identified, the precise length, distance and other parameters of the obstacle to be identified can be obtained, and more accurate laser light can be provided for the identification of the obstacle to be identified. The point cloud ensures the accuracy of identifying the obstacle to be identified.

In another preferred implementation of this embodiment, if in the Nth frame, the obstacle to be identified has entered part of the lidar blind area, and the relative speed between the unmanned vehicle and the obstacle to be identified is If the value is positive, the obstacle to be recognized in the N+1th frame continues to partially enter the lidar blind area. Since the lidar scan cycle is fast, such as 360 times per second, within 1/360 second, the obstacle to be identified can only partially enter the blind area of the lidar, which causes the lidar to only acquire the obstacle to be identified part of the laser point cloud of the object. Part of the point cloud of the obstacle to be identified in the N+1th frame and the point cloud distribution of the obstacle to be identified in the Nth frame have changed, then the matching will fail during the matching process, and the N+1th The obstacle to be recognized in the frame is determined to be a new obstacle, but the obstacle to be recognized in the Nth frame continues to enter the lidar blind area. Truncation, matching and completion are performed in the same manner as in the above-mentioned embodiments of the present application. Wherein, the laser point cloud of the obstacle to be identified in the Nth frame is the laser point cloud of the obstacle to be identified obtained after completing the laser point cloud of the obstacle to be identified in the Nth frame .

In another preferred implementation of this embodiment, if in the Nth frame, the obstacle to be identified has entered part of the lidar blind area, and the relative speed between the unmanned vehicle and the obstacle to be identified is Negative value, the obstacle to be recognized in frame N+1 still continues to partially enter the lidar blind area (the part that enters the lidar blind area is less or all exits the lidar blind area). Since the lidar scan cycle is fast, such as 360 times per second, within 1/360 second, the obstacle to be identified can only partially enter the blind area of the lidar, which causes the lidar to only acquire the obstacle to be identified part of the laser point cloud of the object. Part of the point cloud of the obstacle to be identified in the N+1th frame and the point cloud distribution of the obstacle to be identified in the Nth frame have changed, then the matching will fail during the matching process, and the N+1th The obstacle to be recognized in the frame is determined to be a new obstacle, but the obstacle to be recognized in the Nth frame continues to enter the lidar blind zone. Using the same method as the above-mentioned embodiment of the present application, by completing the laser point cloud of the obstacle to be identified in the Nth frame, and the laser point cloud of the obstacle to be identified in the N+1th frame Matching, after the matching is successful, complete the laser point cloud of the obstacle to be identified in the N+1th frame, perform obstacle recognition according to the completed obstacle laser point cloud, and judge whether it affects The movement of a human-driven vehicle. Wherein, the laser point cloud of the obstacle to be identified in the Nth frame is the laser point cloud of the obstacle to be identified obtained after completing the laser point cloud of the obstacle to be identified in the Nth frame .

By adopting the technical solutions provided by the above embodiments, some obstacles entering the blind area of the lidar can be identified, avoiding the risk of collision caused by misidentifying their length and distance, and effectively improving the driving safety of unmanned vehicles.

It should be noted that for the foregoing method embodiments, for the sake of simple description, they are expressed as a series of action combinations, but those skilled in the art should know that the present application is not limited by the described action sequence. Depending on the application, certain steps may be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification belong to preferred embodiments, and the actions and modules involved are not necessarily required by this application.

The above is the introduction about the method embodiment, and the solution of the present invention will be further described through the device embodiment below.

Fig. 2 is a schematic structural diagram of a lidar obstacle recognition device provided by an embodiment of the present application, as shown in Fig. 2 , including:

Obtaining module 21, is used for obtaining the information of the obstacle to be identified of the consecutive N+1 frames around the laser radar scanning unmanned vehicle;

The truncation module 22 is used to determine whether the obstacle to be identified in the N+1th frame enters the lidar blind area according to the information of the obstacle to be identified in each frame of the first N frames in the N+1 frame ; If entering the lidar blind area, truncate the laser point cloud of the obstacle to be identified in the Nth frame according to the length of the obstacle to be identified entering the lidar blind area;

The completion module 23 is used to match the laser point cloud of the obstacle to be identified in the N+1th frame according to the truncated laser point cloud of the obstacle to be identified; if the matching is successful, the N+1th Complete the laser point cloud of the obstacle to be identified in the frame;

The identification module 24 is configured to perform obstacle identification according to the completed laser point cloud of the obstacle to be identified.

In a preferred implementation of the acquisition module 21,

The obstacle recognition method of this embodiment is applied in the technical field of unmanned driving. In unmanned driving, unmanned vehicles need to be able to automatically identify obstacles in the road, so as to make timely decisions and control during vehicle driving, so as to facilitate the safe driving of vehicles. The execution body of the obstacle recognition method in this embodiment can be an obstacle recognition device, which can be obtained by integrating multiple modules, and the obstacle recognition device can be specifically set in an unmanned vehicle to Human-driven vehicles are controlled.

The information of the obstacle to be identified in this embodiment can be obtained by laser radar scanning. The specification of lidar can adopt 16 lines, 32 lines or 64 lines and so on. The higher the number of lines, the greater the unit energy density of the lidar. In this embodiment, the laser radar mounted on the current vehicle rotates 360 times per second, and scans the information of obstacles to be identified around the current vehicle, which is a frame of information of obstacles to be identified. The information of the obstacle to be identified in this embodiment may include a point cloud of the obstacle to be identified and a reflection value of the obstacle to be identified. There may be one or more obstacles to be identified around the current vehicle. After the lidar scan, the center of mass position of the current vehicle can be used as the origin of the coordinate system, and the two directions parallel to the horizontal plane are taken as the x direction and the y direction as the length direction and the width direction, and the direction perpendicular to the ground is the z direction , as the height direction. Then, the obstacle to be identified can be identified in the coordinate system according to the relative position and distance of each point in the point cloud of the obstacle to be identified and the origin.

Preferably, the obstacles to be identified in each frame of the first N frames in the N+1 frame are matched, the corresponding relationship of the obstacles to be identified in each frame is determined, and the obstacles to be identified in each frame are determined. identify. Through the matching operation, it is possible to avoid the problem that the same obstacle to be recognized in different frames is recognized as a different obstacle when the obstacle to be recognized fails to be tracked in a complex scene (for example, there are multiple obstacles).

Preferably, the identification is performed after acquiring the information of the obstacle to be identified in each frame around the unmanned vehicle scanned by the lidar.

In this way, in the point cloud of each obstacle to be identified, the point cloud of each obstacle to be identified can be obtained according to the relative position of each point in each obstacle to be identified and the current vehicle. In addition, lidar can also detect the reflection value of each point in each obstacle to be identified. In practical applications, the coordinate system can also take the center of mass position of the lidar as the origin, and the other directions remain unchanged. The value of N in this embodiment may be selected according to actual requirements.

In a preferred implementation of the truncation module 22,

According to the information of the obstacle to be identified in each frame of the first N frames in the N+1 frame, it is judged whether the obstacle to be identified in the N+1 frame enters the lidar blind area; if the N+1 frame The obstacle to be identified enters the lidar blind area, and the laser point cloud of the obstacle to be identified in the Nth frame is truncated according to the length of the obstacle to be identified entering the lidar blind area;

Preferably, in each frame of the previous N frames, the point cloud of each obstacle to be identified is obtained, and matched with the point cloud of the obstacle to be identified in the previous frame, to determine that the same obstacle to be identified is in Point clouds in each of the previous N frames.

Through the point clouds of the same obstacle to be identified in each frame in the previous N frames, determine the position of the obstacle to be identified in the Nth frame, with the current center of mass position of the unmanned vehicle as the origin in the coordinate system, Movement speed and direction of movement.

It is judged whether the obstacle to be identified in the N+1th frame enters the lidar blind zone. In this embodiment, the laser radar blind zone is within 8 meters of the current unmanned vehicle's centroid position as the origin.

If in the Nth frame, the obstacle to be identified is already located at the edge of the lidar blind zone, and the relative speed between the unmanned vehicle and the obstacle to be identified is a positive value, then the obstacle to be identified in the N+1 frame Identify obstacles entering the lidar blind zone. Since the lidar scan cycle is fast, such as 360 times per second, within 1/360 second, the obstacle to be identified can only partially enter the blind area of the lidar, which causes the lidar to only acquire the obstacle to be identified part of the laser point cloud of the object. Part of the point cloud of the obstacle to be identified in the N+1th frame and the point cloud distribution of the obstacle to be identified in the Nth frame have changed, then the matching will fail during the matching process, and the N+1th The obstacle to be identified in the frame is determined to be a new obstacle, rather than the obstacle to be identified in the Nth frame partially entering the lidar blind area.

Therefore, according to the position, moving speed and moving direction of the obstacle to be identified in the coordinate system with the current center of mass position of the unmanned vehicle as the origin in the Nth frame, and the range of the lidar blind zone, determine the N+1th frame The length of the obstacle to be identified entering the blind zone of the lidar.

Cut off the laser point cloud of the obstacle to be identified in the Nth frame according to the length of the obstacle to be identified entering the blind zone of the laser radar in the N+1th frame, so as to enter the blind zone of the laser radar in the N+1th frame Match the laser point cloud of the obstacle to be identified.

Preferably, the laser point cloud of the obstacle to be identified in the Nth frame is intercepted into the part of the length of the lidar blind zone, and the interception takes into account the three-dimensional model of the lidar being blocked by the body of the unmanned vehicle, rather than a simple Cut vertically.

In a preferred implementation of the completion module 23,

Matching the truncated laser point cloud of the obstacle to be identified with the laser point cloud of the obstacle to be identified in the N+1th frame;

Wherein, the laser point cloud obtained by truncating the laser point cloud of the obstacle to be identified in the Nth frame is the predicted laser point cloud of the obstacle to be identified in the N+1th frame, by combining with the N+1th Matching the laser point cloud of the actual obstacle to be identified in the frame can improve the matching accuracy and success rate.

If the matching is successful, the laser point cloud of the obstacle to be identified in the N+1th frame is completed; that is, the laser point of the part where the obstacle to be identified in the N+1th frame enters the blind area of the lidar Cloud completes.

Preferably, the completion is based on the laser point cloud of the obstacle to be identified in the Nth frame to complete the laser point cloud of the part where the obstacle to be identified in the N+1th frame enters the blind area of the lidar . Since the completion is performed on the basis of the laser point cloud of the obstacle to be recognized in the N+1th frame, the completed laser point cloud is more in line with the shape of the actual obstacle.

In a preferred implementation of the identification module 24,

Obstacle identification is performed according to the completed laser point cloud of the obstacle to be identified.

Preferably, according to the completed laser point cloud of the obstacle to be identified, the precise length, distance and other parameters of the obstacle to be identified can be obtained, and more accurate laser light can be provided for the identification of the obstacle to be identified. The point cloud ensures the accuracy of identifying the obstacle to be identified.

In another preferred implementation of this embodiment, if in the Nth frame, the obstacle to be identified has entered part of the lidar blind area, and the relative speed between the unmanned vehicle and the obstacle to be identified is If the value is positive, the obstacle to be recognized in the N+1th frame continues to partially enter the lidar blind area. Since the lidar scan cycle is fast, such as 360 times per second, within 1/360 second, the obstacle to be identified can only partially enter the blind area of the lidar, which causes the lidar to only acquire the obstacle to be identified part of the laser point cloud of the object. Part of the point cloud of the obstacle to be identified in the N+1th frame and the point cloud distribution of the obstacle to be identified in the Nth frame have changed, then the matching will fail during the matching process, and the N+1th The obstacle to be recognized in the frame is determined to be a new obstacle, but the obstacle to be recognized in the Nth frame continues to enter the lidar blind area. Truncation, matching and completion are performed in the same manner as in the above-mentioned embodiments of the present application. Wherein, the laser point cloud of the obstacle to be identified in the Nth frame is the laser point cloud of the obstacle to be identified obtained after completing the laser point cloud of the obstacle to be identified in the Nth frame .

In another preferred implementation of this embodiment, if in the Nth frame, the obstacle to be identified has entered part of the lidar blind area, and the relative speed between the unmanned vehicle and the obstacle to be identified is Negative value, the obstacle to be recognized in frame N+1 still continues to partially enter the lidar blind area (the part that enters the lidar blind area is less or all exits the lidar blind area). Since the lidar scan cycle is fast, such as 360 times per second, within 1/360 second, the obstacle to be identified can only partially enter the blind area of the lidar, which causes the lidar to only acquire the obstacle to be identified part of the laser point cloud of the object. Part of the point cloud of the obstacle to be identified in the N+1th frame and the point cloud distribution of the obstacle to be identified in the Nth frame have changed, then the matching will fail during the matching process, and the N+1th The obstacle to be recognized in the frame is determined to be a new obstacle, but the obstacle to be recognized in the Nth frame continues to enter the lidar blind area. Using the same method as the above-mentioned embodiment of the present application, by completing the laser point cloud of the obstacle to be identified in the Nth frame, and the laser point cloud of the obstacle to be identified in the N+1th frame Matching, after the matching is successful, complete the laser point cloud of the obstacle to be identified in the N+1th frame, perform obstacle recognition according to the completed obstacle laser point cloud, and judge whether it affects The movement of a human-driven vehicle. Wherein, the laser point cloud of the obstacle to be identified in the Nth frame is the laser point cloud of the obstacle to be identified obtained after completing the laser point cloud of the obstacle to be identified in the Nth frame .

By adopting the technical solutions provided by the above embodiments, it is possible to identify some of the obstacles entering the blind area of the laser radar, avoiding the risk of collisions caused by misidentifying their lengths and distances, and effectively improving the safety of unmanned vehicles. driving safety of the vehicle.

In the above-mentioned embodiments, descriptions of each embodiment have their own emphases, and for parts not described in detail in a certain embodiment, reference may be made to relevant descriptions of other embodiments.

In the several embodiments provided in this application, it should be understood that the disclosed methods and devices may be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or May be integrated into another system, or some features may be ignored, or not implemented. In another point, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, and the indirect coupling or communication connection of devices or units may be in electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit. The integrated unit can be implemented in the form of hardware, or in the form of hardware plus software functional units.

Figure 3 shows a block diagram of an exemplary computer system/server 012 suitable for use in implementing embodiments of the present invention. The computer system/server 012 shown in FIG. 3 is only an example, and should not limit the functions and scope of use of the embodiments of the present invention.

As shown in Figure 3, computer system/server 012 takes the form of a general-purpose computing device. Components of computer system/server 012 may include, but are not limited to: one or more processors or processing units 016, system memory 028, bus 018 connecting various system components including system memory 028 and processing unit 016.

Bus 018 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus structures. These architectures include, by way of example, but are not limited to Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MAC) bus, Enhanced ISA bus, Video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect ( PCI) bus.

Computer system/server 012 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by computer system/server 012 and include both volatile and nonvolatile media, removable and non-removable media.

System memory 028 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 030 and/or cache memory 032 . The computer system/server 012 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. By way of example only, storage system 034 may be used to read from and write to non-removable, non-volatile magnetic media (not shown in FIG. 3, commonly referred to as a "hard drive"). Although not shown in Figure 3, a disk drive for reading and writing to removable non-volatile disks (e.g. "floppy disks") may be provided, as well as for removable non-volatile optical disks (e.g. CD-ROM, DVD-ROM or other optical media) CD-ROM drive. In these cases, each drive may be connected to bus 018 via one or more data media interfaces. Memory 028 may include at least one program product having a set (eg, at least one) of program modules configured to perform the functions of various embodiments of the present invention.

A program/utility 040 having a set (at least one) of program modules 042, such as may be stored in memory 028, such program modules 042 including - but not limited to - an operating system, one or more application programs, other program Modules and program data, each or some combination of these examples may include the implementation of the network environment. Program modules 042 generally perform the functions and/or methods of the described embodiments of the present invention.

The computer system/server 012 can also communicate with one or more external devices 014 (such as keyboards, pointing devices, displays 024, etc.). In the present invention, the computer system/server 012 communicates with external radar devices, and can also communicate with one or Devices that enable a user to interact with the computer system/server 012, and/or communicate with any device that enables the computer system/server 012 to communicate with one or more other computing devices (e.g., network cards, modems, etc.) communication. Such communication may occur through input/output (I/O) interface 022 . Also, the computer system/server 012 can also communicate with one or more networks (eg, a local area network (LAN), a wide area network (WAN) and/or a public network such as the Internet) through the network adapter 020 . As shown in FIG. 3 , network adapter 020 communicates with other modules of computer system/server 012 via bus 018 . It should be appreciated that although not shown in FIG. 3, other hardware and/or software modules may be used in conjunction with computer system/server 012, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems , tape drives, and data backup storage systems.

The processing unit 016 executes the functions and/or methods in the described embodiments of the present invention by running the programs stored in the system memory 028 .

The above-mentioned computer program can be set in a computer storage medium, that is, the computer storage medium is encoded with a computer program, and when the program is executed by one or more computers, one or more computers can execute the computer programs shown in the above-mentioned embodiments of the present invention. Method flow and/or device operation.

With the development of time and technology, the meaning of medium has become more and more extensive, and the transmission path of computer programs is no longer limited to tangible media, and can also be downloaded directly from the Internet. Any combination of one or more computer readable medium(s) may be utilized. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (non-exhaustive list) of computer readable storage media include: electrical connections with one or more leads, 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 device, magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

A computer readable signal medium may include a data signal carrying computer readable program code in baseband or as part of a carrier wave. Such propagated data signals may take many forms, including - but not limited to - electromagnetic signals, optical signals, or any suitable combination of the foregoing. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. .

Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including - but not limited to - wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

Computer program code for carrying out the operations of the present invention may be written in one or more programming languages, or combinations thereof, including object-oriented programming languages—such as Java, Smalltalk, C++, and conventional Procedural Programming Language - such as "C" or a similar programming language. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In cases involving a remote computer, the remote computer can be connected to the user computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (such as via the Internet using an Internet service provider). connect).

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, rather than limiting them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: it can still Modifications are made to the technical solutions described in the foregoing embodiments, or equivalent replacements are made to some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present application.

Claims (12)

1. A laser radar obstacle recognition method, comprising:

acquiring information of obstacles to be identified of continuous N +1 frames around the unmanned vehicle scanned by the laser radar;

judging whether the obstacle to be identified in the (N + 1) th frame enters a laser radar blind area or not according to the information of the obstacle to be identified in each frame in the previous (N) frames in the (N + 1) th frame; if the obstacle to be recognized enters the laser radar blind area, cutting off the laser point cloud of the obstacle to be recognized in the Nth frame according to the length of the obstacle to be recognized entering the laser radar blind area;

matching the intercepted laser point cloud of the obstacle to be identified with the laser point cloud of the obstacle to be identified in the (N + 1) th frame; if the matching is successful, completing the laser point cloud of the obstacle to be recognized in the (N + 1) th frame;

and identifying the obstacle according to the supplemented obstacle laser point cloud to be identified.

2. The method of claim 1, wherein obtaining information that the lidar scans for consecutive N +1 frames of obstacles to be identified around the unmanned vehicle comprises:

and obtaining point clouds of obstacles to be identified in each frame in the first N frames in the N +1 frames, matching the point clouds of the obstacles to be identified in each frame in the first N frames in the N +1 frames, and determining the corresponding relation of the obstacles to be identified in each frame.

3. The method according to claim 1, wherein determining whether the obstacle to be identified in the N +1 th frame enters the lidar blind area according to the information of the obstacle to be identified in each of the first N frames of the N +1 frames comprises:

and determining the position, the moving speed and the moving direction of the obstacle to be recognized in the Nth frame and the range of a laser radar blind area through the point cloud in each frame of the previous N frames of the obstacle to be recognized, and determining the length of the obstacle to be recognized entering the laser radar blind area in the (N + 1) th frame.

4. The method of claim 1, wherein truncating the obstacle laser point cloud to be identified in the Nth frame according to the length of the obstacle to be identified entering the lidar blind zone comprises:

and cutting the laser point cloud of the barrier to be recognized in the Nth frame according to the three-dimensional model of the blind area and the length of the barrier to be recognized entering the laser radar blind area.

5. The method of claim 1, wherein completing the obstacle laser point cloud to be identified in the (N + 1) th frame comprises:

if the obstacle to be identified in the Nth frame does not enter the laser radar blind area, completing the laser point cloud of the part of the obstacle to be identified in the (N + 1) th frame, which enters the laser radar blind area, according to the laser point cloud of the obstacle to be identified in the Nth frame;

and if the obstacle to be identified in the Nth frame enters the laser radar blind area, completing the laser point cloud of the part of the obstacle to be identified in the (N + 1) th frame, which enters the laser radar blind area, according to the laser point cloud after the completion of the obstacle to be identified in the Nth frame.

6. A laser radar obstacle recognition apparatus, comprising:

the system comprises an acquisition module, a display module and a control module, wherein the acquisition module is used for acquiring information of obstacles to be identified of continuous N +1 frames around a laser radar scanning unmanned vehicle;

the truncation module is used for judging whether the obstacle to be identified in the (N + 1) th frame enters a laser radar blind area or not according to the information of the obstacle to be identified in each frame in the previous (N) frames in the (N + 1) th frame; if the obstacle to be recognized enters the laser radar blind area, cutting off the laser point cloud of the obstacle to be recognized in the Nth frame according to the length of the obstacle to be recognized entering the laser radar blind area;

the completion module is used for matching the intercepted laser point cloud of the obstacle to be identified with the laser point cloud of the obstacle to be identified in the (N + 1) th frame; if the matching is successful, completing the laser point cloud of the obstacle to be recognized in the (N + 1) th frame;

and the identification module is used for identifying the obstacle according to the supplemented obstacle laser point cloud to be identified.

7. The apparatus of claim 6, wherein the obtaining module is specifically configured to:

and obtaining point clouds of obstacles to be identified in each frame in the first N frames in the N +1 frames, matching the point clouds of the obstacles to be identified in each frame in the first N frames in the N +1 frames, and determining the corresponding relation of the obstacles to be identified in each frame.

8. The apparatus of claim 6, wherein the truncation module is specifically configured to:

and determining the position, the moving speed and the moving direction of the obstacle to be recognized in the Nth frame and the range of a laser radar blind area through the point cloud in each frame of the previous N frames of the obstacle to be recognized, and determining the length of the obstacle to be recognized entering the laser radar blind area in the (N + 1) th frame.

9. The apparatus of claim 6, wherein the truncation module is further specifically configured to:

and cutting the laser point cloud of the barrier to be recognized in the Nth frame according to the three-dimensional model of the blind area and the length of the barrier to be recognized entering the laser radar blind area.

10. The apparatus of claim 6, wherein the completion module is specifically configured to:

if the obstacle to be identified in the Nth frame does not enter the laser radar blind area, completing the laser point cloud of the part of the obstacle to be identified in the (N + 1) th frame, which enters the laser radar blind area, according to the laser point cloud of the obstacle to be identified in the Nth frame;

and if the obstacle to be identified in the Nth frame enters the laser radar blind area, completing the laser point cloud of the part of the obstacle to be identified in the (N + 1) th frame, which enters the laser radar blind area, according to the laser point cloud after the completion of the obstacle to be identified in the Nth frame.

11. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program implements the method of any one of claims 1 to 5.

12. A computer-readable storage medium, on which a computer program is stored, which program, when being executed by a processor, carries out the method according to any one of claims 1 to 5.