Obstacle sensing method, device, electronic device, storage medium and vehicle

Abstract

The present disclosure relates to an obstacle perception method, device, electronic device, storage medium and vehicle. The method determines the number of times an obstacle in each grid in the semantic map grid appears in the multi-frame point cloud data by obtaining multi-frame point cloud data of a semantic map grid and the weight values ​​corresponding to each frame of point cloud data in the multi-frame point cloud data. The method calculates the number of weighted obstacles in each grid in the semantic map grid based on the weight values ​​corresponding to each frame of point cloud data in the multi-frame point cloud data and the number of times an obstacle in each grid in the semantic map grid appears in the multi-frame point cloud data. When the number of weighted obstacles in a grid is greater than a preset number threshold, it is determined that an obstacle appears in the grid in the semantic map grid. The embodiments of the present disclosure can avoid reporting that an obstacle is detected when the obstacle only appears in one frame of point cloud data, thereby reducing false detections to a certain extent and improving the accuracy of obstacle perception.

Description

Technical Field

[0001] The present disclosure relates to the field of autonomous driving technology, and in particular to an obstacle perception method, device, electronic device, storage medium, and vehicle. Background Art

[0002] With the development of unmanned driving technology, self-driving vehicles will gradually be promoted in people's daily lives.

[0003] An autonomous vehicle is a vehicle that can start, drive, and stop without a driver. Autonomous driving technology relies on the vehicle's ability to perceive surrounding obstacles. Existing perception technologies typically require multiple frames of point cloud data to achieve obstacle perception.

[0004] Existing multi-frame detection methods, for multi-frame point cloud data at the same location, will report an obstacle as detected if only one frame detects an obstacle at that location. However, the obstacle detected in that frame is likely due to noise or other factors. Therefore, existing multi-frame detection methods are prone to false detections. Moreover, the more frames they rely on, the more false detections they will cause, resulting in low obstacle perception accuracy. Summary of the Invention

[0005] In order to solve the above technical problems, the present disclosure provides an obstacle perception method, device, electronic device, storage medium and vehicle to reduce false detection and improve the accuracy of obstacle perception.

[0006] In a first aspect, an embodiment of the present disclosure provides an obstacle perception method, the method comprising:

[0007] Obtaining multi-frame point cloud data of a semantic map grid and weight values ​​corresponding to each frame of point cloud data in the multi-frame point cloud data;

[0008] Determining the number of times an obstacle within each grid in the semantic map grid appears in the multiple frames of point cloud data;

[0009] According to the weight values ​​corresponding to each frame of point cloud data in the multi-frame point cloud data and the number of times an obstacle in each grid in the semantic map grid appears in the multi-frame point cloud data, the number of times the obstacle in each grid in the semantic map grid has a weight is calculated. When the number of times the obstacle in the grid has a weight is greater than a preset number threshold, it is determined that an obstacle appears in the grid in the semantic map grid.

[0010] In a second aspect, an embodiment of the present disclosure provides an obstacle sensing device, comprising:

[0011] an acquiring unit, configured to acquire multiple frames of point cloud data of a semantic map grid and weight values ​​corresponding to each frame of point cloud data in the multiple frames of point cloud data;

[0012] A first determining unit is configured to determine the number of times an obstacle in each grid in the semantic map grid appears in the multi-frame point cloud data;

[0013] The second determination unit is used to calculate the number of times an obstacle in each grid in the semantic map grid has a weight based on the weight value corresponding to each frame of point cloud data in the multi-frame point cloud data and the number of times an obstacle in each grid in the semantic map grid appears in the multi-frame point cloud data. When the number of times an obstacle in a grid has a weight is greater than a preset number threshold, it is determined that an obstacle appears in the grid in the semantic map grid.

[0014] In a third aspect, an embodiment of the present disclosure provides an electronic device, including:

[0015] Memory;

[0016] processor; and

[0017] computer programs;

[0018] The computer program is stored in the memory and is configured to be executed by the processor to implement the method as described in the first aspect.

[0019] In a fourth aspect, an embodiment of the present disclosure provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program is executed by a processor to implement the method described in the first aspect.

[0020] In a fifth aspect, an embodiment of the present disclosure further provides a computer program product, which includes a computer program or instructions, and when the computer program or instructions are executed by a processor, the obstacle perception method as described above is implemented.

[0021] In a sixth aspect, an embodiment of the present disclosure further provides a vehicle, comprising:

[0022] Memory;

[0023] processor; and

[0024] computer programs;

[0025] The computer program is stored in the memory and is configured to be executed by the processor to implement the obstacle perception method described above.

[0026] The obstacle perception method, apparatus, electronic device, storage medium, and vehicle provided in the embodiments of the present disclosure obtain multi-frame point cloud data of a semantic map grid and the weight values ​​corresponding to each frame of point cloud data in the multi-frame point cloud data, and further determine the number of times an obstacle within each grid in the semantic map grid appears in the multi-frame point cloud data. Furthermore, based on the weight values ​​corresponding to each frame of point cloud data in the multi-frame point cloud data and the number of times an obstacle within each grid in the semantic map grid appears in the multi-frame point cloud data, the number of weighted obstacles within each grid in the semantic map grid is calculated. When the number of weighted obstacles within a grid exceeds a preset number threshold, it is determined that an obstacle exists in that grid in the semantic map grid. Based on the number of times an obstacle in each grid in the semantic map grid appears in multiple frames of point cloud data, combined with the weight value corresponding to each frame of point cloud data in the multiple frames of point cloud data, the number of times the obstacle in each grid in the semantic map grid has a weight can be calculated. When the number of times the obstacle in the grid has a weight is greater than the preset number threshold, it is determined that an obstacle appears in the grid in the semantic map grid. This can avoid reporting that an obstacle is detected when the obstacle only appears in one frame of point cloud data, reduce false detection to a certain extent, and thus improve the accuracy of obstacle perception. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0028] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0029] Figure 1 A flowchart of the obstacle perception method provided in an embodiment of the present disclosure;

[0030] Figure 2 A flowchart of an obstacle perception method provided by another embodiment of the present disclosure;

[0031] Figure 3 A flowchart of an obstacle perception method provided by another embodiment of the present disclosure;

[0032] Figure 4 A flowchart of an obstacle perception method provided by another embodiment of the present disclosure;

[0033] Figure 5 A schematic diagram of the structure of an obstacle sensing device provided in an embodiment of the present disclosure;

[0034] Figure 6 A schematic diagram of the structure of an electronic device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0035] In order to more clearly understand the above-mentioned objectives, features and advantages of the present disclosure, the scheme of the present disclosure will be further described below. It should be noted that the embodiments of the present disclosure and the features therein can be combined with each other in the absence of conflict.

[0036] The following description sets forth many specific details to facilitate a full understanding of the present disclosure. However, the present disclosure may also be implemented in other ways than those described herein. It is apparent that the embodiments in the specification are only some of the embodiments of the present disclosure, not all of them. The specific embodiments described herein are intended only to explain the present disclosure and are not intended to limit the present disclosure. All other embodiments derived by persons of ordinary skill in the art based on the described embodiments of the present disclosure are intended to fall within the scope of protection of the present disclosure.

[0037] The embodiments of the present disclosure provide an obstacle perception method, which is described below in conjunction with specific embodiments.

[0038] Figure 1 Flowchart of the obstacle perception method provided by the embodiment of the present disclosure. The method can be applied to an on-board terminal, which can calculate the number of weighted obstacles in each grid in the semantic map grid based on the weight values ​​corresponding to each frame of point cloud data in the multi-frame point cloud data and the number of times the obstacle in each grid in the semantic map grid appears in the multi-frame point cloud data. When the number of weighted obstacles in the grid is greater than the preset number threshold, it is determined that an obstacle appears in the grid in the semantic map grid, thereby avoiding the situation where the obstacle is detected when the obstacle only appears in one frame of point cloud data, thereby improving the accuracy of obstacle perception. It can be understood that the obstacle perception method provided by the embodiment of the present disclosure can also be applied in other scenarios.

[0039] Below Figure 1 The obstacle perception method shown in FIG. 1 is introduced, and the specific steps of the method are as follows:

[0040] S101, obtaining multi-frame point cloud data of a semantic map grid and weight values ​​corresponding to each frame of point cloud data in the multi-frame point cloud data.

[0041] For example, the vehicle is equipped with a laser radar, and point cloud data can be obtained through the laser radar. The vehicle-mounted terminal can obtain point cloud data of the semantic map grid based on the laser radar. The semantic map grid is a map grid in the world coordinate system. The vehicle-mounted terminal can obtain multi-frame point cloud data and obtain the weight value corresponding to each frame of point cloud data in the multi-frame point cloud data. The purpose of obtaining multi-frame point cloud data is to perceive obstacles in combination with multi-frame point cloud data to reduce the occurrence of false detection. In some embodiments, the vehicle-mounted terminal is pre-set with a weight value corresponding to each frame of point cloud data in the multi-frame point cloud data. The closer the frame is to the current moment, the higher the weight value. For example, the vehicle-mounted terminal obtains eight frames of point cloud data, the weight value of the point cloud data of the current frame can be 3, the weight value of the point cloud data of the previous frame of the current frame can be 2, and the weight values ​​of the other six frames of point cloud data can all be 1. This is just an example and is not specifically limited.

[0042] In some embodiments, the method of obtaining multi-frame point cloud data of a semantic map grid and the weight values ​​corresponding to each frame of point cloud data in the multi-frame point cloud data includes: obtaining multi-frame point cloud data; matching each frame of point cloud data in the multi-frame point cloud data with the semantic map grid to obtain multi-frame point cloud data of the semantic map grid; for each frame of point cloud data in the multi-frame point cloud data of the semantic map grid, if the absolute value of the difference between the acquisition time of the frame point cloud data and the current time is less than or equal to a preset first time threshold, then determining that the weight value corresponding to the frame point cloud data is a first weight value; if the absolute value of the difference between the acquisition time of the frame point cloud data and the current time is greater than the preset first time threshold and less than the preset second time threshold, then determining that the weight value corresponding to the frame point cloud data is a second weight value; if the absolute value of the difference between the acquisition time of the frame point cloud data and the current time is greater than or equal to the preset second time threshold, then determining that the weight value corresponding to the frame point cloud data is a third weight value.

[0043] For example, if the vehicle-mounted terminal acquires eight frames of point cloud data, and the absolute value of the difference between the acquisition time of the first frame and the current time is less than or equal to the preset first time threshold, the weight value corresponding to the frame of point cloud data is determined to be the first weight value; if the absolute value of the difference between the acquisition time of the second frame of point cloud data and the current time is greater than the preset first time threshold and less than the preset second time threshold, the weight value corresponding to the second frame of point cloud data is determined to be the second weight value; if the absolute value of the difference between the acquisition time of the other six frames of point cloud data and the current time is greater than or equal to the preset second time threshold, the weight value corresponding to the other six frames of point cloud data is determined to be the third weight value. Optionally, the first weight value can be 3, the second weight value can be 2, and the third weight value can be 1. The first weight value, the second weight value, and the third weight value can be set by the user, and the set weight value and setting method are not limited here.

[0044] S102: Determine the number of times an obstacle in each grid in the semantic map grid appears in the multi-frame point cloud data.

[0045] After the vehicle terminal obtains multiple frames of point cloud data for the semantic map grid, it determines the number of times an obstacle within each grid in the semantic map grid appears in the multiple frames of point cloud data. For example, if the vehicle terminal obtains eight frames of point cloud data, it determines that an obstacle within grid A of the semantic map appears six times, within grid B eight times, within grid C once, within grid D twice, and within grid E zero times.

[0046] S103. Calculate the number of weighted obstacles in each grid in the semantic map grid based on the weight values ​​corresponding to each frame of point cloud data in the multi-frame point cloud data and the number of times an obstacle in each grid in the semantic map grid appears in the multi-frame point cloud data. When the number of weighted obstacles in the grid is greater than a preset number threshold, determine that an obstacle appears in the grid in the semantic map grid.

[0047] Based on the number of times an obstacle appears in each grid cell of the semantic map in multiple frames of point cloud data, combined with the weight values ​​corresponding to each frame of the multi-frame point cloud data, the vehicle terminal can calculate the number of weighted obstacle occurrences within each grid cell of the semantic map. When the number of weighted obstacle occurrences within a grid cell exceeds a preset threshold, it is determined that an obstacle exists in that grid cell within the semantic map, and the grid cell within the semantic map cell where the obstacle appears can be further determined. Furthermore, an image of the obstacle perception can be output within the semantic map grid, allowing for intuitive perception of the obstacle distribution.

[0048] The disclosed embodiment obtains multi-frame point cloud data of a semantic map grid and the weight values ​​corresponding to each frame of point cloud data in the multi-frame point cloud data, and further determines the number of times an obstacle within each grid in the semantic map grid appears in the multi-frame point cloud data. Furthermore, based on the weight values ​​corresponding to each frame of point cloud data in the multi-frame point cloud data and the number of times an obstacle within each grid in the semantic map grid appears in the multi-frame point cloud data, the weighted number of times an obstacle within each grid in the semantic map grid appears is calculated. When the weighted number of times an obstacle within a grid appears is greater than a preset number threshold, it is determined that an obstacle appears in that grid in the semantic map grid. Based on the number of times an obstacle in each grid in the semantic map grid appears in multiple frames of point cloud data, combined with the weight value corresponding to each frame of point cloud data in the multiple frames of point cloud data, the number of times the obstacle in each grid in the semantic map grid has a weight can be calculated. When the number of times the obstacle in the grid has a weight is greater than the preset number threshold, it is determined that an obstacle appears in the grid in the semantic map grid. This can avoid reporting that an obstacle is detected when the obstacle only appears in one frame of point cloud data, reduce false detection to a certain extent, and thus improve the accuracy of obstacle perception.

[0049] Figure 2 This is a flow chart of an obstacle perception method provided by another embodiment of the present disclosure, such as Figure 2 As shown, the method includes the following steps:

[0050] S201 , obtaining multi-frame point cloud data of a semantic map grid and weight values ​​corresponding to each frame of point cloud data in the multi-frame point cloud data.

[0051] Specifically, the implementation process and principle of S201 and S101 are the same and will not be repeated here.

[0052] S202: Determine the number of times an obstacle in each grid in the semantic map grid appears in the multi-frame point cloud data.

[0053] Specifically, the implementation process and principle of S202 are the same as those of S102 and will not be repeated here.

[0054] In some embodiments, S202 includes but is not limited to S2021, S2022, and S2023:

[0055] S2021. Determine, based on the multi-frame point cloud data, an obstacle detection status within each grid in the semantic map grid corresponding to each frame of point cloud data in the multi-frame point cloud data;

[0056] After acquiring multiple frames of point cloud data of the semantic map grid, the vehicle terminal determines the obstacle detection status of each grid in the semantic map grid corresponding to each frame of point cloud data, that is, determines whether an obstacle appears in each grid.

[0057] S2022. For each frame of point cloud data in the multiple frames of point cloud data, if an obstacle is detected within a grid, the number of times the grid corresponds to the point cloud data in the frame is recorded as 1;

[0058] For each frame of point cloud data in the multi-frame point cloud data, if the vehicle-mounted terminal detects an obstacle within the grid, the number of times the grid corresponds to the point cloud data in the frame is recorded as 1.

[0059] S2023. If no obstacle is detected in the grid, the number of times the grid corresponds to the point cloud data of this frame is recorded as 0.

[0060] If the vehicle terminal detects that there is no obstacle within a grid, the number of times that grid appears in that frame of point cloud data is recorded as 0. In this way, the number of times an obstacle appears in each grid in each frame of point cloud data can be determined, avoiding the situation where an obstacle is reported as detected when the obstacle only appears in one frame of point cloud data, that is, only appears once, and reducing false detections caused by factors such as noise.

[0061] S203: Construct a target bit table based on the number of times an obstacle in each grid appears in the multi-frame point cloud data.

[0062] After determining the number of times an obstacle within each grid in the semantic map appears in multiple frames of point cloud data, the vehicle terminal can construct a target bit table based on the number of times the obstacle appears within each grid. The number of times the obstacle appears within each grid can be filled in the corresponding position of the target bit table. The target bit table can represent the number of times the obstacle within each grid in the semantic map appears in multiple frames of point cloud data.

[0063] S204 : Convert the target bit table into a weighted bit table according to the weight value corresponding to each frame of point cloud data in the multiple frames of point cloud data.

[0064] The vehicle terminal can convert the target bit table into a weighted bit table based on the weight values ​​corresponding to each frame of point cloud data in the multi-frame point cloud data. For example, if the vehicle terminal obtains six frames of point cloud data, the weight value of the point cloud data of the current frame can be 3, the weight value of the point cloud data of the previous frame of the current frame can be 2, and the weight values ​​of the other four frames of point cloud data can all be 1. Among them, the number of times an obstacle appears in grid A in the six frames of point cloud data is 1, 0, 1, 1, 0, and 1 respectively. Since the number of obstacles in the six frames of point cloud data is 4, the value of grid A in the corresponding position of the target bit table is 4. The further to the left the six frames of point cloud data are, the closer they are to the current moment. That is, the number of times an obstacle appears in the point cloud data of the current frame is 1, the number of times an obstacle appears in the point cloud data of the previous frame of the current frame is 0, and the number of times an obstacle appears in the other four frames of point cloud data is 1, 1, 0, and 1, respectively. Multiplying the number of times an obstacle appears in each frame of point cloud data by the weight value corresponding to that frame and then adding the results together can give the weighted number of times. Therefore, the value of the corresponding position of grid A in the weighted bit table is 1*3+0*2+1*1+1*1+0*1+1*1=6. For another example, if the vehicle terminal acquires eight frames of point cloud data, the weight value of the point cloud data of the current frame can be 3, the weight value of the point cloud data of the previous frame of the current frame can be 2, and the weight values ​​of the other six frames of point cloud data can all be 1. The number of times an obstacle appears within grid B in the eight frames of point cloud data is 0, 1, 0, 1, 1, 1, 1, 1, respectively. Since the obstacle appears 6 times in the six frames of point cloud data, the value of grid B in the corresponding position in the target bit table is 6. Since the number of obstacles in the point cloud data of the current frame is 0, the number of obstacles in the point cloud data before the current frame is 1, and the number of obstacles in the other six frames of point cloud data are 0, 1, 1, 1, 1, 1, respectively, the value of grid B in the corresponding position in the weighted bit table is 0*3+1*2+0*1+1*1+1*1+1*1+1*1+1*1=7. Based on the number of obstacles in each frame of point cloud data and the weight value corresponding to each frame of point cloud data, the on-board terminal can replace the value in the corresponding position in the target bit table with the weighted value, thereby converting the target bit table into a weighted bit table.

[0065] S205 : Determine whether an obstacle appears in each grid in the semantic map grid according to the weighted bit table.

[0066] After obtaining the weighted bit table, the vehicle terminal can determine whether an obstacle exists in each grid in the semantic map based on the weighted bit table. In some embodiments, when a value in the weighted bit table exceeds a preset number threshold, the vehicle terminal determines that an obstacle exists in the grid corresponding to the value, and can further determine the grid in the semantic map where the obstacle exists.

[0067] The embodiment of the present disclosure determines the number of times an obstacle in each grid in the semantic map grid appears in the multi-frame point cloud data by obtaining the multi-frame point cloud data of the semantic map grid and the weight values ​​corresponding to each frame of point cloud data in the multi-frame point cloud data. Furthermore, based on the number of times an obstacle in each grid appears in the multi-frame point cloud data, a target bit table is constructed, and the target bit table is converted into a weighted bit table according to the weight values ​​corresponding to each frame of point cloud data in the multi-frame point cloud data. Then, whether an obstacle appears in each grid in the semantic map grid is determined based on the weighted bit table. In at least one embodiment of the present disclosure, the number of times an obstacle appears in each grid in each frame of point cloud data can be determined, so as to avoid reporting that an obstacle is detected when the obstacle only appears in one frame of point cloud data, that is, only appears once, thereby reducing false detections caused by factors such as noise. Based on the number of times an obstacle appears in each frame of point cloud data and the corresponding weight value for each frame of point cloud data, the value at the corresponding position in the target bit table is replaced with a weighted value, thereby converting the target bit table into a weighted bit table. When the value in the weighted bit table exceeds a preset threshold, the vehicle terminal determines that an obstacle has appeared in the grid corresponding to the value position. This can then determine the grid in the semantic map where the obstacle appears, further improving the accuracy of obstacle perception. Even better, an image of obstacle perception can be output within the semantic map grid, allowing for intuitive perception of the distribution of obstacles.

[0068] Figure 3 This is a flow chart of an obstacle perception method provided by another embodiment of the present disclosure, such as Figure 3 As shown, the method includes the following steps:

[0069] S301 , obtaining multi-frame point cloud data of a semantic map grid and weight values ​​corresponding to each frame of point cloud data in the multi-frame point cloud data.

[0070] Specifically, the implementation process and principle of S301 and S101 are the same and will not be described in detail here.

[0071] S302: Determine the number of times an obstacle in each grid in the semantic map grid appears in the multi-frame point cloud data.

[0072] Specifically, the implementation process and principle of S302 and S102 are the same and will not be described in detail here.

[0073] S303 : Based on the number of times an obstacle in each grid in the semantic map grid appears in the initial frame, the number of times the obstacle in each grid appears in the initial frame is used as an initial bit value, and an initial bit table is generated based on the initial bit value.

[0074] Based on the initial frame point cloud data, the vehicle terminal can determine the number of times an obstacle appears in each grid cell of the semantic map in the initial frame. Furthermore, the number of times an obstacle appears in each grid cell in the initial frame is used as the initial bit value. Based on this initial bit value, an initial bit table is generated. This is more intuitive and makes it easier to find the number of times an obstacle appears in a particular grid cell.

[0075] S304: Determine the total number of times that obstacles in each grid in the semantic map grid appear in other frames.

[0076] The vehicle terminal can determine the total number of times that obstacles in each grid in the semantic map grid appear in other frames based on the point cloud data of other frames.

[0077] S305 : Add the total number of times an obstacle in each grid in the semantic map appears in other frames to the initial bit value corresponding to the grid to obtain a target bit value corresponding to each grid.

[0078] The vehicle terminal adds the total number of times that obstacles in each grid in the semantic map grid appear in other frames to the initial bit value corresponding to the grid to obtain the target bit value corresponding to each grid.

[0079] S306 : Replace the initial bit values ​​in the initial bit table based on the target bit values ​​corresponding to each grid to obtain a target bit table.

[0080] After the on-board terminal obtains the target bit value corresponding to each grid, it replaces the initial bit value in the initial bit table with the target bit value corresponding to each grid to obtain the target bit table. The initial frame point cloud data is iterated into multi-frame point cloud data. The disclosed embodiment can achieve efficient query and iteration by storing information on whether there are obstacles in the semantic map grid in the historical frame by bit. For example, the number of times obstacles appear in six frames of point cloud data is currently obtained. If the number of times obstacles appear in eight frames of point cloud data is desired, it is only necessary to update the number of times obstacles appear in the other two frames of point cloud data to the target bit table, which is simpler and faster.

[0081] S307 : Convert the target bit table into a weighted bit table according to the weight value corresponding to each frame of point cloud data in the multiple frames of point cloud data.

[0082] Specifically, the implementation process and principle of S307 and S204 are the same and will not be repeated here.

[0083] S308: Determine whether an obstacle appears in each grid in the semantic map according to the weighted bit table.

[0084] Specifically, the implementation process and principle of S308 and S205 are the same and will not be repeated here.

[0085] In some embodiments, S308 includes but is not limited to S3081 and S3082:

[0086] S3081. Perform convolution calculation on the weighted bit values ​​in the weighted bit table to obtain convolution values ​​of the weighted bit values.

[0087] For example, the vehicle terminal performs convolution calculations on the weighted bits in the weighted bit table according to a 3*3 grid to obtain the convolution value of the weighted bit value. It is understood that convolution calculations can also be performed according to other sizes without limitation, and the preset thresholds corresponding to convolution calculations of different sizes are also different.

[0088] S3082: If the convolution value is greater than a preset threshold, it is determined that there is an obstacle in the grid corresponding to the convolution value of the weighted bit value.

[0089] If the convolution value is greater than a preset threshold, it is determined that an obstacle exists within the grid corresponding to the convolution value of the weighted bit value. The preset threshold can be set arbitrarily and is not limited. In this embodiment, a preset threshold of 8 is used as an example. When the convolution value is greater than 8, it is determined that an obstacle exists within the grid corresponding to the convolution value of the weighted bit value. When the convolution value is less than or equal to 8, it is determined that there is no obstacle within the grid corresponding to the convolution value of the weighted bit value, further improving the accuracy of obstacle detection.

[0090] The disclosed embodiment determines the number of times an obstacle appears in each grid cell of the semantic map grid in the multi-frame point cloud data by obtaining multi-frame point cloud data and the weight values ​​corresponding to each frame of point cloud data in the multi-frame point cloud data. Then, based on the number of times each obstacle appears in the initial frame, the number of times each obstacle appears in the initial frame is used as an initial bit value. An initial bit table is generated based on the initial bit value, and the total number of times each obstacle appears in the semantic map grid in other frames is determined. Furthermore, the total number of times each obstacle appears in the semantic map grid in other frames is added to the initial bit value corresponding to the grid to obtain a target bit value for each grid cell. The initial bit value in the initial bit table is replaced based on the target bit value for each grid cell to obtain a target bit table. Furthermore, based on the weight values ​​corresponding to each frame of point cloud data in the multi-frame point cloud data, the target bit table is converted into a weighted bit table. Based on the weighted bit table, whether an obstacle appears in each grid cell of the semantic map grid is determined. The disclosed embodiment can achieve efficient query and iteration by storing information on whether there are obstacles in the semantic map grid in the historical frame on a bit-by-bit basis. This can avoid reporting that an obstacle is detected when the obstacle only appears in one frame of point cloud data, that is, only appears once. This reduces false detections caused by factors such as noise and further improves the accuracy of obstacle perception.

[0091] Figure 4 This is a flow chart of an obstacle perception method provided by another embodiment of the present disclosure, such as Figure 4 As shown, the method includes the following steps:

[0092] S401: Obtain multi-frame point cloud data of a semantic map grid and weight values ​​corresponding to each frame of point cloud data in the multi-frame point cloud data.

[0093] Specifically, the implementation process and principle of S401 and S101 are the same and will not be described in detail here.

[0094] S402: Determine the number of times an obstacle in each grid in the semantic map grid appears in the multi-frame point cloud data.

[0095] Specifically, the implementation process and principle of S402 and S102 are the same and will not be repeated here.

[0096] S403: Construct a target bit table based on the number of times an obstacle in each grid appears in the multi-frame point cloud data.

[0097] Specifically, the implementation process and principle of S403 and S203 are the same, and will not be repeated here.

[0098] S404. According to the weight value corresponding to each frame of point cloud data in the multiple frames of point cloud data, the number of times the obstacle in each grid in the semantic map grid appears in each frame of point cloud data is multiplied by the weight value corresponding to the frame of point cloud data to obtain the weighted number of times corresponding to each frame of point cloud data.

[0099] For example, if the vehicle terminal acquires eight frames of point cloud data, the weight value of the point cloud data of the current frame can be 3, the weight value of the point cloud data of the previous frame can be 2, and the weight values ​​of the other six frames of point cloud data can all be 1. Within the B grid, the number of times obstacles appear in the eight frames of point cloud data is 0, 1, 0, 1, 1, 1, 1, and 1, respectively. Since the number of times obstacles appear in the point cloud data of the current frame is 0, the weighted number of times the current frame of point cloud data corresponds to is 0*3=0; the number of times obstacles appear in the point cloud data of the previous frame of the current frame is 1, the weighted number of times the current frame of point cloud data corresponds to is 1*2=2; the number of times obstacles appear in the other six frames of point cloud data is 0, 1, 1, 1, 1, and 1, and the corresponding weighted numbers are 0*1=0, 1*1=1, 1*1=1, 1*1=1, 1*1=1, and 1*1=1, respectively. Thus, the weighted number of times each frame of point cloud data corresponds to is obtained.

[0100] S405 : Add the weighted times corresponding to each frame of point cloud data to obtain the weighted bit value corresponding to each grid.

[0101] The terminal adds up the weighted times corresponding to each frame of point cloud data to obtain the weighted bit value corresponding to each grid. For example, as in S404, the value of the corresponding position of grid B in the weighted bit table is 0+2+0+1+1+1+1+1=7.

[0102] S406 : Replace the bit value corresponding to each grid in the target bit table with the weighted bit value corresponding to each grid to obtain a weighted bit table.

[0103] For example, the on-board terminal acquires six frames of point cloud data. Within grid A, the number of times an obstacle appears in the six frames is 1, 0, 1, 1, 0, and 1, respectively. Since the obstacle appears four times in the six frames, the value of the corresponding position in the target bit table for grid A is 4. The further to the left in the six frames of point cloud data, the closer it is to the current moment. That is, the number of obstacle appearances in the current frame's point cloud data is 1, the number of obstacle appearances in the previous frame's point cloud data is 0, and the number of obstacle appearances in the other four frames is 1, 1, 0, and 1, respectively. Multiplying the number of obstacle appearances in each frame of point cloud data by the corresponding weight value and adding the results yields the weighted number of times. Therefore, the value of the corresponding position in the weighted bit table for grid A is 1*3+0*2+1*1+1*1+0*1+1*1=6. The weighted bit value 6 corresponding to grid A is then used to replace the bit value 4 corresponding to grid A in the target bit table. Similarly, the replacement is performed for the other grids.

[0104] S407: Determine whether an obstacle appears in each grid in the semantic map according to the weighted bit table.

[0105] Specifically, the implementation process and principle of S407 and S205 are the same and will not be repeated here.

[0106] S408: Outputting an image of obstacle perception in the semantic map grid.

[0107] After the on-board terminal obtains the weighted bit table, it can determine the weighted number of times an obstacle appears in each grid in the semantic map grid based on the correspondence between the position of each weighted bit value and the grid in the semantic map grid, and then based on the weighted bit value in the weighted bit table, thereby outputting an image of obstacle perception in the semantic map grid, and intuitively perceiving the distribution of obstacles.

[0108] The disclosed embodiment determines the number of times an obstacle appears in each grid in the semantic map grid by obtaining multi-frame point cloud data of the semantic map grid and the weight values ​​corresponding to each frame of the multi-frame point cloud data. Then, based on the number of times an obstacle appears in each grid in the multi-frame point cloud data, a target bit table is constructed. Furthermore, based on the weight values ​​corresponding to each frame of the multi-frame point cloud data, the number of times an obstacle appears in each grid in the semantic map grid in each frame of the point cloud data is multiplied by the weight value corresponding to the frame of the point cloud data to obtain the weighted number of times corresponding to each frame of the point cloud data. The weighted number of times corresponding to each frame of the point cloud data is added to obtain the weighted bit value corresponding to each grid. The bit value corresponding to each grid in the target bit table is replaced with the weighted bit value corresponding to each grid to obtain a weighted bit table. Based on the weighted bit table, whether an obstacle appears in each grid in the semantic map grid is determined, and an obstacle perception image is output in the semantic map grid. This embodiment can determine the number of times an obstacle appears within each grid in each frame of point cloud data. This prevents reporting an obstacle detection when the obstacle appears only once in one frame of point cloud data, thus reducing false detections due to factors such as noise. It also outputs an image of the obstacle perception within the semantic map grid, allowing for intuitive perception of the obstacle distribution.

[0109] Figure 5 This is a schematic diagram of the structure of the obstacle sensing device provided in the embodiment of the present disclosure. The obstacle sensing device can be the vehicle-mounted device as described in the above embodiment, or the obstacle sensing device can be a component or assembly in the vehicle-mounted device. The obstacle sensing device provided in the embodiment of the present disclosure can execute the processing flow provided in the obstacle sensing method embodiment, such as Figure 5 As shown, the obstacle perception device 50 includes: an acquisition unit 51 for acquiring multi-frame point cloud data of a semantic map grid and a weight value corresponding to each frame of point cloud data in the multi-frame point cloud data; a first determination unit 52 for determining the number of times an obstacle in each grid in the semantic map grid appears in the multi-frame point cloud data; a second determination unit 53 for calculating the number of weighted obstacles in each grid in the semantic map grid according to the weight value corresponding to each frame of point cloud data in the multi-frame point cloud data and the number of times an obstacle in each grid in the semantic map grid appears in the multi-frame point cloud data. When the number of weighted obstacles in a grid is greater than a preset number threshold, it is determined that an obstacle appears in the grid in the semantic map grid.

[0110] Optionally, when the acquisition unit 51 acquires multi-frame point cloud data of a semantic map grid and the weight values ​​corresponding to each frame of point cloud data in the multi-frame point cloud data, it is specifically used to: acquire multi-frame point cloud data; match each frame of point cloud data in the multi-frame point cloud data with the semantic map grid to obtain multi-frame point cloud data of the semantic map grid; for each frame of point cloud data in the multi-frame point cloud data of the semantic map grid, if the absolute value of the difference between the acquisition time of the frame point cloud data and the current time is less than or equal to a preset first time threshold, then determine that the weight value corresponding to the frame point cloud data is a first weight value; if the absolute value of the difference between the acquisition time of the frame point cloud data and the current time is greater than the preset first time threshold and less than the preset second time threshold, then determine that the weight value corresponding to the frame point cloud data is a second weight value; if the absolute value of the difference between the acquisition time of the frame point cloud data and the current time is greater than or equal to the preset second time threshold, then determine that the weight value corresponding to the frame point cloud data is a third weight value.

[0111] Optionally, when determining the number of times an obstacle in each grid in the semantic map grid appears in the multi-frame point cloud data, the first determination unit 52 is specifically used to: determine the obstacle detection status in each grid in the semantic map grid corresponding to each frame of point cloud data in the multi-frame point cloud data based on the multi-frame point cloud data; for each frame of point cloud data in the multi-frame point cloud data, if an obstacle is detected in the grid, the number of times the grid appears in the frame of point cloud data is recorded as 1; if no obstacle is detected in the grid, the number of times the grid appears in the frame of point cloud data is recorded as 0.

[0112] Optionally, the device also includes: a construction unit 54 and a conversion unit 55; the construction unit 54 is used to construct a target bit table based on the number of times the obstacle in each grid appears in the multi-frame point cloud data; the conversion unit 55 is used to convert the target bit table into a weighted bit table according to the weight values ​​corresponding to each frame of point cloud data in the multi-frame point cloud data; the second determination unit 53 calculates the number of weighted obstacles in each grid in the semantic map grid according to the weight values ​​corresponding to each frame of point cloud data in the multi-frame point cloud data and the number of times the obstacle in each grid in the semantic map grid appears in the multi-frame point cloud data, and when the number of weighted obstacles in the grid is greater than a preset number threshold, it is determined that an obstacle appears in the grid in the semantic map grid, specifically for: determining whether an obstacle appears in each grid in the semantic map grid according to the weighted bit table.

[0113] Optionally, when constructing the target bit table based on the number of times the obstacles in each grid appear in the multi-frame point cloud data, the construction unit 54 is specifically used to: take the number of times the obstacles in each grid in the semantic map grid appear in the initial frame as the initial bit value, and generate an initial bit table based on the initial bit value; determine the total number of times the obstacles in each grid in the semantic map grid appear in other frames; add the total number of times the obstacles in each grid in the semantic map grid appear in the other frames to the initial bit value corresponding to the grid to obtain the target bit value corresponding to each grid; and replace the initial bit value in the initial bit table based on the target bit value corresponding to each grid to obtain the target bit table.

[0114] Optionally, when the conversion unit 55 converts the target bit table into a weighted bit table according to the weight values ​​corresponding to each frame of point cloud data in the multi-frame point cloud data, it is specifically used to: multiply the number of times the obstacle in each grid in the semantic map grid appears in each frame of point cloud data by the weight value corresponding to the frame of point cloud data according to the weight values ​​corresponding to each frame of point cloud data, to obtain the weighted number of times corresponding to each frame of point cloud data; add the weighted number of times corresponding to each frame of point cloud data to obtain the weighted bit value corresponding to each grid; and replace the bit value corresponding to each grid in the target bit table with the weighted bit value corresponding to each grid to obtain a weighted bit table.

[0115] Optionally, when the second determination unit 53 determines whether an obstacle appears in each grid in the semantic map grid based on the weighted bit table, it is specifically used to: perform a convolution calculation on the weighted bit value in the weighted bit table to obtain the convolution value of the weighted bit value; if the convolution value is greater than a preset threshold, it is determined that there is an obstacle in the grid corresponding to the convolution value of the weighted bit value.

[0116] Optionally, the device further includes: an output unit 56; the output unit 56 is configured to output an image of obstacle perception in the semantic map grid.

[0117] Figure 5 The obstacle sensing device of the illustrated embodiment can be used to implement the technical solution of the above-mentioned method embodiment. Its implementation principle and technical effects are similar and will not be repeated here.

[0118] Figure 6 This is a schematic diagram of the structure of an electronic device in the embodiment of the present disclosure. Figure 6 , which shows a structural diagram of an electronic device 600 suitable for implementing the embodiments of the present disclosure. Figure 6 The electronic device shown is only an example and should not limit the functions and scope of use of the embodiments of the present disclosure.

[0119] like Figure 6 As shown, the electronic device 600 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes to implement the obstacle perception method of the embodiment described in the present disclosure according to the program stored in the read-only memory (ROM) 602 or the program loaded from the storage device 608 to the random access memory (RAM) 603. Various programs and data required for the operation of the electronic device 600 are also stored in the RAM 603. The processing device 601, ROM 602, and RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0120] Typically, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device 600 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 6 The electronic device 600 is shown with various devices, but it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed instead.

[0121] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart, thereby implementing the obstacle perception method described above. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 609, or installed from the storage device 608, or installed from the ROM 602. When the computer program is executed by the processing device 601, the above-mentioned functions defined in the method of the embodiment of the present disclosure are performed.

[0122] An embodiment of the present disclosure also provides a vehicle, comprising: a memory; a processor; and a computer program; wherein the computer program is stored in the memory and is configured to be executed by the processor to implement the obstacle perception method as described above.

[0123] It should be noted that the computer-readable medium mentioned above in the present disclosure may be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or component, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, 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, device, or component. In the present disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. Such a propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to wires, optical cables, RF (radio frequency), etc., or any suitable combination thereof.

[0124] In some embodiments, the client and server can communicate using any currently known or future developed network protocol, such as HTTP (HyperText Transfer Protocol), and can be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0125] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0126] The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device:

[0127] Obtaining multi-frame point cloud data of a semantic map grid and weight values ​​corresponding to each frame of point cloud data in the multi-frame point cloud data;

[0128] Determining the number of times an obstacle within each grid in the semantic map grid appears in the multiple frames of point cloud data;

[0129] According to the weight values ​​corresponding to each frame of point cloud data in the multi-frame point cloud data and the number of times an obstacle in each grid in the semantic map grid appears in the multi-frame point cloud data, the number of times the obstacle in each grid in the semantic map grid has a weight is calculated. When the number of times the obstacle in the grid has a weight is greater than a preset number threshold, it is determined that an obstacle appears in the grid in the semantic map grid.

[0130] Optionally, when the above one or more programs are executed by the electronic device, the electronic device may also execute other steps described in the above embodiments.

[0131] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages, or a combination thereof, including, but not limited to, object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on the remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0132] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the module, program segment, or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of the boxes in the block diagram and / or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0133] The units involved in the embodiments described in this disclosure may be implemented in software or hardware, wherein the name of a unit does not necessarily limit the unit itself.

[0134] The functions described above herein may be performed, at least in part, by one or more hardware logic components. For example, and without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chip (SOCs), complex programmable logic devices (CPLDs), and the like.

[0135] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in conjunction with an instruction execution system, device or equipment. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium can include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0136] The above description is merely a preferred embodiment of the present disclosure and an illustration of the technical principles employed. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by the specific combination of the above-mentioned technical features, but also includes other technical solutions formed by any combination of the above-mentioned technical features or their equivalents without departing from the above-mentioned disclosed concepts. For example, a technical solution formed by replacing the above-mentioned features with (but not limited to) technical features with similar functions disclosed in this disclosure.

[0137] In addition, although each operation is described in a specific order, this should not be understood as requiring these operations to be performed in the specific order shown or in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details have been included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0138] Although the subject matter has been described in language specific to structural features and / or methodological logical acts, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are merely example forms of implementing the claims.

Claims

1. An obstacle perception method, characterized in that: The method comprises: Obtaining multi-frame point cloud data of a semantic map grid and weight values ​​corresponding to each frame of point cloud data in the multi-frame point cloud data; Determining the number of times an obstacle within each grid in the semantic map grid appears in the multiple frames of point cloud data; According to the weight values ​​corresponding to each frame of point cloud data in the multi-frame point cloud data and the number of times an obstacle in each grid in the semantic map grid appears in the multi-frame point cloud data, the number of times the obstacle in each grid in the semantic map grid has a weight is calculated. When the number of times the obstacle in the grid has a weight is greater than a preset number threshold, it is determined that an obstacle appears in the grid in the semantic map grid.

2. The method according to claim 1, characterized in that The acquiring of multi-frame point cloud data of the semantic map grid and the weight values ​​corresponding to each frame of point cloud data in the multi-frame point cloud data include: Get multi-frame point cloud data; Matching each frame of point cloud data in the multiple frames of point cloud data with the semantic map grid to obtain multiple frames of point cloud data of the semantic map grid; For each frame of point cloud data in the multiple frames of point cloud data of the semantic map grid, if the absolute value of the difference between the acquisition time of the frame of point cloud data and the current time is less than or equal to a preset first time threshold, determining the weight value corresponding to the frame of point cloud data to be a first weight value; If the absolute value of the difference between the acquisition time of the frame point cloud data and the current time is greater than the preset first time threshold and less than the preset second time threshold, determining the weight value corresponding to the frame point cloud data to be the second weight value; If the absolute value of the difference between the acquisition time of the frame point cloud data and the current time is greater than or equal to the preset second time threshold, the weight value corresponding to the frame point cloud data is determined to be a third weight value.

3. The method according to claim 1, characterized in that Determining the number of times an obstacle in each grid in the semantic map grid appears in the multi-frame point cloud data includes: Determine, based on the multiple frames of point cloud data, an obstacle detection condition within each grid in the semantic map grid corresponding to each frame of point cloud data in the multiple frames of point cloud data; For each frame of point cloud data in the multiple frames of point cloud data, if an obstacle is detected in a grid, the number of times the grid corresponds to the point cloud data in the frame is recorded as 1; If no obstacle is detected in the grid, the number of times the grid corresponds to the point cloud data of this frame is recorded as 0.

4. The method according to claim 1, wherein After determining the number of times an obstacle in each grid in the semantic map grid appears in the multiple frames of point cloud data, the method further includes: Constructing a target bit table based on the number of times the obstacle in each grid appears in the multi-frame point cloud data; Converting the target bit table into a weighted bit table according to the weight values ​​corresponding to each frame of point cloud data in the multiple frames of point cloud data; The method further comprises calculating the number of weighted obstacle occurrences in each grid in the semantic map grid according to the weight values ​​corresponding to each frame of point cloud data in the multiple frames of point cloud data and the number of occurrences of obstacles in each grid in the semantic map grid in the multiple frames of point cloud data, and determining that an obstacle occurs in the grid in the semantic map grid when the number of weighted obstacle occurrences in the grid is greater than a preset number threshold. According to the weighted bit table, it is determined whether an obstacle appears in each grid in the semantic map grid.

5. The method according to claim 4, characterized in that The constructing a target bit table based on the number of times the obstacle in each grid appears in the multi-frame point cloud data includes: According to the number of times an obstacle in each grid of the semantic map grid appears in the initial frame, the number of times the obstacle in each grid appears in the initial frame is used as an initial bit value, and an initial bit table is generated based on the initial bit value; Determining the total number of times an obstacle in each grid of the semantic map appears in other frames; Adding the total number of times an obstacle in each grid in the semantic map appears in the other frames to the initial bit value corresponding to the grid to obtain a target bit value corresponding to each grid; The initial bit values ​​in the initial bit table are replaced based on the target bit values ​​corresponding to each grid to obtain a target bit table.

6. The method according to claim 4, characterized in that The converting the target bit table into a weighted bit table according to the weight values ​​corresponding to each frame of point cloud data in the multiple frames of point cloud data includes: According to the weight values ​​corresponding to each frame of point cloud data in the multiple frames of point cloud data, the number of times an obstacle in each grid in the semantic map grid appears in each frame of point cloud data is multiplied by the weight value corresponding to the frame of point cloud data to obtain the weighted number of times corresponding to each frame of point cloud data; Adding the weighted times corresponding to each frame of point cloud data to obtain the weighted bit value corresponding to each grid; The bit value corresponding to each grid in the target bit table is replaced with the weighted bit value corresponding to each grid to obtain a weighted bit table.

7. The method according to claim 4, characterized in that The determining, based on the weighted bit table, whether an obstacle appears in each grid in the semantic map grid includes: Performing a convolution calculation on the weighted bit values ​​in the weighted bit table to obtain a convolution value of the weighted bit values; If the convolution value is greater than a preset threshold, it is determined that there is an obstacle in the grid corresponding to the convolution value of the weighted bit value.

8. The method according to claim 1, characterized in that When the number of weighted obstacles in the grid is greater than a preset number threshold, after determining that an obstacle appears in the grid in the semantic map grid, the method further includes: An image of obstacle perception is output in the semantic map grid.

9. An obstacle sensing device, characterized in that: The device comprises: an acquiring unit, configured to acquire multiple frames of point cloud data of a semantic map grid and weight values ​​corresponding to each frame of point cloud data in the multiple frames of point cloud data; A first determining unit is configured to determine the number of times an obstacle in each grid in the semantic map grid appears in the multi-frame point cloud data; The second determination unit is used to calculate the number of times an obstacle in each grid in the semantic map grid has a weight based on the weight value corresponding to each frame of point cloud data in the multi-frame point cloud data and the number of times an obstacle in each grid in the semantic map grid appears in the multi-frame point cloud data. When the number of times an obstacle in a grid has a weight is greater than a preset number threshold, it is determined that an obstacle appears in the grid in the semantic map grid.

10. An electronic device, characterized in that: include: Memory; processor; as well as computer programs; The computer program is stored in the memory and configured to be executed by the processor to implement the method according to any one of claims 1 to 8.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

12. A vehicle, characterized in that: Comprising the obstacle sensing device as claimed in claim 9.

Images