Information acquisition method and apparatus, and computer device and storage medium

Abstract

The present application relates to an information acquisition method and apparatus, and a computer device and a storage medium. The method comprises: acquiring a segmentation result of a target environment image collected by a camera of a target vehicle, and a detected three-dimensional lane line point cloud of a target lane line on a target road on which the target vehicle travels; on the basis of the current pose of the camera, the segmentation result and the detected three-dimensional lane line point cloud, determining whether there is a target pitch angle related to the target environment image; and when it is determined that there is a target pitch angle related to the target environment image, using a target corrected pose corresponding to the target pitch angle to transform the position of a traffic cone in the target environment image in an image coordinate system of the target environment image into a corrected position of the traffic cone in a vehicle body coordinate system.

Description

Information acquisition methods, devices, computer equipment and storage media

[0001] This application claims priority to Chinese Patent Application No. 202510085155.8, filed on January 20, 2025, entitled "Information Acquisition Method, Apparatus, Computer Equipment and Storage Medium", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This disclosure relates to, but is not limited to, the field of vehicle technology, such as information acquisition methods, devices, computer equipment, and storage media. Background Technology

[0003] Obtaining the location of traffic cones is a key factor that autonomous driving systems must consider when making driving decisions in scenarios such as road construction detours and temporary road closures.

[0004] Typically, information such as laser reflection intensity and laser point cloud density is obtained from the lidar on the vehicle to determine the position of the cone. Based on this information, the position of the cone is predicted. Summary of the Invention

[0005] The following is an overview of the subject matter described in detail herein. This overview is not intended to limit the scope of the claims.

[0006] This disclosure provides an information acquisition method, apparatus, computer device, and storage medium to address the problem of improving the accuracy of the acquired cone position.

[0007] The technical solution adopted in the embodiments of this disclosure is as follows:

[0008] An information acquisition method, comprising:

[0009] The segmentation result of the target environment image captured by the camera of the target vehicle and the detected three-dimensional lane line point cloud of the target lane line on the target road on which the target vehicle is traveling are obtained. The segmentation result indicates multiple lane line pixels of the target lane line, and the detected three-dimensional lane line point cloud is detected based on the target environment image.

[0010] Based on the current pose of the camera, the segmentation result, and the detected 3D lane line point cloud, it is determined whether there is a target pitch angle related to the target environment image. The target projected 3D lane line point cloud obtained by inversely projecting the multiple lane line pixels onto the vehicle coordinate system of the target vehicle using the target corrected pose corresponding to the target pitch angle has an error less than an error threshold with the detected 3D lane line point cloud. The target corrected pose is obtained by replacing the current pitch angle in the current pose of the camera with the target pitch angle.

[0011] When the target pitch angle is determined to exist, the position of the cone in the target environment image in the image coordinate system of the target environment image is transformed into the corrected position of the cone in the vehicle coordinate system by using the target corrected pose.

[0012] Based on the aforementioned technical means, the segmentation results of the target environment image captured by the target vehicle's camera and the detected 3D lane line point cloud of the target lane line on the target road on which the target vehicle is traveling are used as information to obtain the position of the cone. The segmentation results of the target environment image captured by the target vehicle's camera and the detected 3D lane line point cloud of the target lane line on the target road on which the target vehicle is traveling can indicate the pixel features such as the position and type of the cone in the environment image. Even when the distance between the vehicle and the cone is far, the full range of pixel features of the cone can still be obtained. That is, the segmentation results of the target environment image captured by the target vehicle's camera and the detected 3D lane line point cloud of the target lane line on the target road on which the target vehicle is traveling can indicate the full range of pixel features of the cone. The amount of information involved in determining the position of the cone is large, which improves the accuracy of the obtained cone position and avoids the situation where the accuracy of the cone position is not affected by the distance between the vehicle and the cone.

[0013] The information acquisition method provided in this disclosure considers that at the moment the transformation of the cone's position in the target environment image's image coordinate system to its position in the target vehicle's vehicle coordinate system begins, the actual pose of the target vehicle's camera may differ from its current pose. The intrinsic parameters of the camera pose are fixed, while changes in the extrinsic parameters are typically caused by changes in the pitch angle. Differences in the pose of the same camera at different times are usually due to differences in the pitch angle at those times. Directly using the current pose of the target vehicle's camera to transform the cone's position in the target environment image's image coordinate system to its position in the target vehicle's vehicle coordinate system would lead to a decrease in the accuracy of the cone's position.

[0014] The information acquisition method provided in this disclosure determines whether a target pitch angle related to the target environment image exists based on the current pose of the target vehicle's camera, the segmentation result, and the detected 3D lane line point cloud. This is equivalent to determining whether there is a more accurate pitch angle, i.e., a target pitch angle, relative to the current pose of the target vehicle's camera. When a more accurate pitch angle is determined, the current pitch angle in the target vehicle's camera's current pose is replaced with the target pitch angle to obtain a more accurate target correction pose relative to the current pose of the target vehicle's camera. This more accurate target correction pose is used to transform the position of the cone in the target environment image's image coordinate system to the corrected position of the cone in the target vehicle's vehicle coordinate system. This obtains a more accurate corrected position of the cone in the target vehicle's vehicle coordinate system, improving the accuracy of the obtained cone position.

[0015] The method may further include: determining the outline of the target cone based on the segmentation result of the target environment image, wherein the target cone is any cone in the target environment image;

[0016] The position of the target cone in the image coordinate system is determined by the intersection of the outline of the target cone and the perpendicular bisector of the detection box surrounding the target cone.

[0017] The method may also include:

[0018] Performing cone tracking corresponding to the target environment image, the cone tracking includes:

[0019] A tracking operation corresponding to the target environment image is performed, and after the tracking operation, target particles are found from the target particle set, and the positions of the target particles are used as the tracking positions of the corresponding cones in the target environment image. The tracking positions are then output. The target particle set is initialized to an initial particle set before the first cone tracking operation. The initial particle set is obtained by generating particles based on the corresponding positions of the cones in the vehicle coordinate system in the environment image used for particle initialization. The tracking operation includes:

[0020] When there is a first particle from the target particle set in the target region associated with the target cone, the weight of the first particle is updated according to the distance between the first particle and the target cone, the weight of the target cone, and the current weight of the first particle. When there is a second particle in the target particle set, the weight of the second particle is reduced. The second particle is a particle that is not in the target region associated with any cone in the target environment image.

[0021] When there are no particles from the target particle set within the target area, new particles generated based on the corresponding position of the target cone in the vehicle coordinate system will be added to the target particle set.

[0022] The method may also include:

[0023] When the target cone has a corrected position in the vehicle coordinate system, the first weight is used as the weight of the target cone;

[0024] When the target cone does not have a corrected position in the vehicle coordinate system, the second weight is used as the weight of the target cone, wherein the second weight is less than the first weight.

[0025] The method may also include:

[0026] The position of the center point is the corresponding position of the target cone in the vehicle coordinate system, and a circular area with a preset first radius is defined as the target area related to the target cone.

[0027] Finding target particles from a set of target particles can include:

[0028] If there are particles with a weight less than 0 in the target particle set, remove the particles with a weight less than 0 from the target particle set.

[0029] Multiple candidate particles are identified from the target particle set, and the weights of the candidate particles are greater than the weight threshold.

[0030] The target particle is identified from multiple candidate particles.

[0031] Finding the target particle from multiple candidate particles can include:

[0032] The candidate particles are sorted from largest to smallest according to their weights to obtain the order of the candidate particles.

[0033] The search operation is performed iteratively until a target number of target particles are found, where the target number is the number of cones in the target environment image. The search operation includes:

[0034] Determine whether there are other particles in the target region related to the candidate particle targeted by the search operation that have not been associated with the candidate particle. The candidate particle targeted by the first search operation is the particle with the largest weight among multiple candidate particles. The candidate particle targeted by the non-first search operation is the next candidate particle after the candidate particle targeted by the previous search operation of the non-first search operation. The next candidate particle after the candidate particle targeted by the previous search operation is indicated by the order.

[0035] If there are other particles in the target region related to the candidate particle targeted by the search operation that have not been associated with the candidate particle, the candidate particle targeted by the search operation is determined as the target particle, and other particles in the target region related to the candidate particle targeted by the search operation that have not been associated with the candidate particle are associated with the candidate particle targeted by the search operation, and the next candidate particle of the candidate particle targeted by the search operation is determined as the candidate particle targeted by the next search operation.

[0036] If there are no other particles in the target region related to the candidate particle targeted by the search operation that have not been associated with the candidate particle, the next candidate particle targeted by the search operation is determined as the candidate particle targeted by the next search operation, wherein the next candidate particle targeted by the search operation is indicated by the order.

[0037] The method may also include:

[0038] The position of the center point is the position of the candidate particle targeted by the search operation in the vehicle coordinate system, and a circular area with a preset second radius is determined as the target area related to the candidate particle targeted by the search operation.

[0039] An information acquisition device, comprising:

[0040] The acquisition unit is configured to acquire the segmentation result of the target environment image captured by the camera of the target vehicle and the detected three-dimensional lane line point cloud of the target lane line on the target road on which the target vehicle is traveling, wherein the segmentation result indicates multiple lane line pixels of the target lane line and the detected three-dimensional lane line point cloud is detected based on the target environment image.

[0041] The determining unit is configured to determine whether there is a target pitch angle related to the target environment image based on the current pose of the camera, the segmentation result, and the detected 3D lane line point cloud. The error between the target projected 3D lane line point cloud obtained by inversely projecting the plurality of lane line pixels onto the vehicle coordinate system of the target vehicle using the target corrected pose corresponding to the target pitch angle and the detected 3D lane line point cloud is less than an error threshold. The target corrected pose is obtained by replacing the current pitch angle in the current pose of the camera with the target pitch angle.

[0042] The conversion unit is configured to, when the target pitch angle is determined to exist, use the target corrected pose to transform the position of the cone in the target environment image in the image coordinate system of the target environment image to the corrected position of the cone in the vehicle coordinate system.

[0043] The information acquisition device may also include:

[0044] The cone position acquisition unit is configured to determine the outline of the target cone based on the segmentation result of the target environment image, wherein the target cone is any cone in the target environment image; and to determine the position of the intersection point of the outline of the target cone and the perpendicular bisector of the detection box surrounding the target cone as the position of the target cone in the image coordinate system of the target environment image.

[0045] The information acquisition device may also include:

[0046] The tracking unit is configured to perform barrel tracking corresponding to a target environment image, the barrel tracking including:

[0047] A tracking operation corresponding to the target environment image is performed, and after the tracking operation, target particles are found from the target particle set, and the positions of the target particles are used as the tracking positions of the corresponding cones in the target environment image. The tracking positions are then output. The target particle set is initialized to an initial particle set before the first cone tracking operation. The initial particle set is obtained by generating particles based on the corresponding positions of the cones in the vehicle coordinate system in the environment image used for particle initialization. The tracking operation includes:

[0048] When there is a first particle from the target particle set in the target region associated with the target cone, the weight of the first particle is updated according to the distance between the first particle and the target cone, the weight of the target cone, and the current weight of the first particle. When there is a second particle in the target particle set, the weight of the second particle is reduced. The second particle is a particle that is not in the target region associated with any cone in the target environment image.

[0049] When there are no particles from the target particle set within the target area, new particles generated based on the corresponding position of the target cone in the vehicle coordinate system will be added to the target particle set.

[0050] The information acquisition device may also include:

[0051] The weighting unit is configured to use a first weight as the weight of the target cone when the target cone has a corrected position in the vehicle coordinate system; and to use a second weight as the weight of the target cone when the target cone does not have a corrected position in the vehicle coordinate system, wherein the second weight is less than the first weight.

[0052] The information acquisition device may also include:

[0053] The first target area determination unit is configured to determine a circular area with a center point at the corresponding position of the target cone in the vehicle coordinate system and a radius of a preset first radius as the target area associated with the target cone.

[0054] The tracking unit can also be configured to delete particles with weights less than 0 from the target particle set when there are particles with weights less than 0 in the target particle set; find multiple candidate particles from the target particle set, where the weights of the candidate particles are greater than a weight threshold; and find the target particle from the multiple candidate particles.

[0055] The tracking unit can also be configured to sort multiple candidate particles according to their weights from largest to smallest to obtain the order of the candidate particles; iteratively perform a search operation until a target number of target particles are found, where the target number is the number of cones in the target environment image. The search operation includes: determining whether there are other particles in the target region related to the candidate particle targeted by the search operation that have not been associated with the candidate particle, wherein the candidate particle targeted in the first search operation is the particle with the largest weight among the multiple candidate particles, and the candidate particle targeted in subsequent search operations is the next candidate particle targeted in the previous search operation of the subsequent search operation, wherein the next candidate particle targeted in the previous search operation is indicated by the order; in the target region related to the candidate particle targeted by the search operation If there are no other particles in the target area that have not been associated with the candidate particle, the candidate particle targeted by the search operation is determined as the target particle, and other particles in the target area related to the candidate particle targeted by the search operation that have not been associated with the candidate particle are associated with the candidate particle targeted by the search operation, and the next candidate particle of the candidate particle targeted by the search operation is determined as the candidate particle targeted by the next search operation; if there are no other particles in the target area related to the candidate particle targeted by the search operation that have not been associated with the candidate particle, the next candidate particle of the candidate particle targeted by the search operation is determined as the candidate particle targeted by the next search operation, wherein the next candidate particle of the candidate particle targeted by the search operation is indicated by the sequence.

[0056] The information acquisition device may also include:

[0057] The second target region determination unit is configured to determine a circular region with a center point that is the position of the candidate particle targeted by the search operation in the vehicle coordinate system and a radius of a preset second radius as the target region related to the candidate particle targeted by the search operation.

[0058] The beneficial effects of the embodiments disclosed herein are as follows:

[0059] The segmentation results of the target environment image captured by the target vehicle's camera and the detected 3D lane line point cloud of the target lane line on the target road on which the target vehicle is traveling are used as information to obtain the position of the cone. The segmentation results of the target environment image captured by the target vehicle's camera and the detected 3D lane line point cloud of the target lane line on the target road on which the target vehicle is traveling can indicate the pixel features such as the position and type of the cone in the environment image. Even when the distance between the vehicle and the cone is far, the full range of pixel features of the cone can still be obtained. That is, the segmentation results of the target environment image captured by the target vehicle's camera and the detected 3D lane line point cloud of the target lane line on the target road on which the target vehicle is traveling can indicate the full range of pixel features of the cone. The information involved in determining the position of the cone has a large amount of information, which improves the accuracy of the obtained cone position and avoids the situation where the accuracy of the cone position is not affected by the distance between the vehicle and the cone.

[0060] The information acquisition method provided in this disclosure considers that at the moment the transformation of the cone's position in the target environment image's image coordinate system to its position in the target vehicle's vehicle coordinate system begins, the actual pose of the target vehicle's camera may differ from its current pose. The intrinsic parameters of the camera pose are fixed, while changes in the extrinsic parameters are typically caused by changes in the pitch angle. Differences in the pose of the same camera at different times are usually due to differences in the pitch angle at those times. Directly using the current pose of the target vehicle's camera to transform the cone's position in the target environment image's image coordinate system to its position in the target vehicle's vehicle coordinate system would lead to a decrease in the accuracy of the cone's position.

[0061] The information acquisition method provided in this disclosure determines whether a target pitch angle related to the target environment image exists based on the current pose of the target vehicle's camera, the segmentation result, and the detected 3D lane line point cloud. This is equivalent to determining whether there is a more accurate pitch angle, i.e., a target pitch angle, relative to the current pose of the target vehicle's camera. When a more accurate pitch angle is determined, the current pitch angle in the target vehicle's camera's current pose is replaced with the target pitch angle to obtain a more accurate target correction pose relative to the current pose of the target vehicle's camera. This more accurate target correction pose is used to transform the position of the cone in the target environment image's image coordinate system to the corrected position of the cone in the target vehicle's vehicle coordinate system. This obtains a more accurate corrected position of the cone in the target vehicle's vehicle coordinate system, improving the accuracy of the obtained cone position.

[0062] After reading and understanding the accompanying diagrams and detailed descriptions, the other aspects can be understood. Attached Figure Description

[0063] Figure 1 is a flowchart illustrating the information acquisition method provided in an embodiment of this disclosure;

[0064] Figure 2 is a flowchart illustrating another information acquisition method provided in an embodiment of this disclosure;

[0065] Figure 3 is an example of the effect of generating particles based on the corrected position of the cone in the vehicle's coordinate system.

[0066] Figure 4 is a schematic diagram of the effect of performing tracking operations corresponding to the target environment image;

[0067] Figure 5 is a schematic diagram illustrating the effect of performing a search operation as an example.

[0068] Figure 6 is a schematic diagram of the hardware structure of a computer device provided in an embodiment of this disclosure. Embodiments of the present invention

[0069] In implementing the embodiments of this disclosure, it was found that when the distance between the vehicle and the traffic cone is large, the laser reflection intensity obtained by the vehicle is low, and the density of the laser point cloud obtained by the vehicle is sparse. This results in less information about the features used to determine the position of the traffic cone, and thus lower accuracy in determining the position of the traffic cone. Improving the accuracy of the determined position of the traffic cone has become a problem that needs to be solved.

[0070] The embodiments of this disclosure will be described below with reference to the accompanying drawings and examples. Those skilled in the art can easily understand other advantages and effects of the embodiments of this disclosure from the content disclosed in this specification. The embodiments of this disclosure can also be implemented or applied through other different methods, and the details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the embodiments of this disclosure. It should be understood that the embodiments are only for illustrating technical solutions and are not intended to limit the scope of protection of the embodiments of this disclosure.

[0071] The illustrations provided in the following embodiments are only schematic representations of the embodiments of this disclosure. Therefore, the drawings only show the components related to the embodiments of this disclosure and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0072] Referring to Figure 1, it shows a flowchart of the information acquisition method provided in an embodiment of the present disclosure.

[0073] In step S101, the segmentation results of the target environment image captured by the camera of the target vehicle and the detected three-dimensional lane line point cloud of the target lane line on the target road on which the target vehicle is traveling are obtained.

[0074] The target vehicle can be any vehicle to which the information acquisition method provided in this embodiment can be applied. The target environment image is captured by the target vehicle's camera. The target environment image describes the environment in which the target vehicle is located.

[0075] The method provided in this disclosure can be performed within a target time period. Each environmental image acquired within the target time period includes traffic cones on the road where the target vehicle is traveling. The target environmental image can be any environmental image acquired within the target time period. For each environmental image acquired within the target time period, at least steps S101-S102 can be performed. The start time of the target time period can be the acquisition time of the first environmental image acquired within the target time period, i.e., the earliest acquisition time of the environmental image, and the end time of the target time period can be the acquisition time of the last environmental image acquired within the target time period, i.e., the latest acquisition time of the environmental image.

[0076] In the disclosed embodiments, the cones in the target environment image may be detected by a target detection network.

[0077] As an example, object detection networks include the YOLO series, Faster R-CNN, etc.

[0078] In this embodiment of the disclosure, the target lane line can be a lane line randomly selected from the lane lines on both sides of the lane where the target vehicle is traveling. Alternatively, the target lane line can be a lane line randomly selected from the lane lines detected on the target road where the target vehicle is traveling.

[0079] The segmentation results of the target environment image indicate: multiple lane line pixels of the target lane line on the target road.

[0080] In other words, for a target lane line on a target road, the segmentation result of the target environment image indicates which pixels in the target environment image are lane line pixels of the target lane line.

[0081] Among them, each of the multiple lane line pixels of the target lane line has coordinates in the image coordinate system of the target environment image.

[0082] In step S101, the target environment image can be input into the instance segmentation network, and the instance segmentation network can output the segmentation result of the target environment image.

[0083] In this embodiment of the disclosure, the three-dimensional lane line point cloud detected by the target lane line is detected based on the target environment image by a neural network for detecting lane lines.

[0084] In step S101, the target environment image can be input into the three-dimensional lane detection network, and the lane detection network outputs the detected three-dimensional lane point cloud of the target lane line on the target road on which the target vehicle is traveling.

[0085] The detected 3D lane line point cloud includes multiple 3D points of the target lane line on the target road where the target vehicle is traveling. These 3D points of the target lane line are located in the vehicle's coordinate system.

[0086] In one possible implementation, the 3D lane detection network is the Bev lane detection network.

[0087] In step S102, based on the current pose of the target vehicle's camera, the segmentation results of the target environment image, and the detected three-dimensional lane line point cloud of the target lane line, it is determined whether there is a target pitch angle related to the target environment image.

[0088] The current pose of the target vehicle's camera can refer to the last determined pose of the target vehicle's camera before the start of S102.

[0089] In this embodiment of the disclosure, the target corrected pose corresponding to the target pitch angle associated with the target environment image is obtained by replacing the current pitch angle in the current pose of the target vehicle's camera with the target pitch angle associated with the target environment image.

[0090] The error between the target projected 3D lane line point cloud obtained by inversely projecting multiple lane line pixels of the target lane line onto the vehicle coordinate system of the target vehicle using the target pitch angle corresponding to the target environment image and the detected 3D lane line point cloud is less than the error threshold.

[0091] Alternatively, for a pitch angle, if the error between the projected 3D lane line point cloud of the target lane line in the vehicle coordinate system of the target vehicle and the detected 3D lane line point cloud of the target lane line in the target corrected pose corresponding to the pitch angle is less than the error threshold, then the pitch angle is determined as the target pitch angle related to the target environment image.

[0092] As an example, to determine whether a target pitch angle exists related to the target environment image, a pitch angle lookup operation can be performed iteratively until the target pitch angle is found or the number of times the pitch angle lookup operation is performed reaches a threshold. Each pitch angle lookup operation targets a reference pitch angle. The reference pitch angle for the first pitch angle lookup operation is the current pitch angle of the target vehicle's camera. The reference pitch angle for the i-th pitch angle lookup operation after the first pitch angle lookup operation is the pitch angle of the sub-interval endpoint of the 3D lane line point cloud that has the smallest error with the detected 3D lane line point cloud, as determined by the (i-1)-th pitch angle lookup operation.

[0093] In this example, the k-th pitch angle lookup operation includes: subtracting a preset angle from the reference pitch angle targeted by the k-th pitch angle lookup operation to obtain the left endpoint pitch angle corresponding to the k-th pitch angle lookup operation; and adding the preset angle to the reference pitch angle targeted by the k-th pitch angle lookup operation to obtain the right endpoint pitch angle corresponding to the k-th pitch angle lookup operation. Here, k-th refers to any operation other than the last one.

[0094] In this example, the first pitch angle interval is divided into multiple first sub-intervals, and each endpoint of each of these first sub-intervals is used as the pitch angle endpoint of the sub-interval corresponding to the k-th pitch angle lookup operation. The two endpoints of this first pitch angle interval are the left endpoint pitch angle corresponding to the k-th pitch angle lookup operation and the reference pitch angle for the k-th pitch angle lookup operation, respectively. The second pitch angle interval is divided into multiple second sub-intervals, and the two endpoints of this second pitch angle interval are the right endpoint pitch angle corresponding to the k-th pitch angle lookup operation and the reference pitch angle for the k-th pitch angle lookup operation, respectively.

[0095] In this example, each endpoint of each of the multiple second sub-intervals is used as the pitch angle of the sub-interval endpoint corresponding to the pitch angle lookup operation executed for the kth time.

[0096] In this example, the pose corresponding to the pitch angle of each sub-interval endpoint is determined for the k-th pitch angle lookup operation. The pose corresponding to the pitch angle of the sub-interval endpoint is obtained by replacing the current pitch angle in the current pose of the target vehicle's camera with the pose corresponding to the pitch angle of the sub-interval endpoint.

[0097] In this example, for each sub-interval endpoint pitch angle corresponding to the pitch angle lookup operation executed for the kth time, multiple lane line pixels of the target lane line are inversely projected onto the vehicle coordinate system of the target vehicle using the pose corresponding to the pitch angle of the sub-interval endpoint, to obtain the projected 3D lane line point cloud corresponding to the pitch angle of the sub-interval endpoint, and the error between the projected 3D lane line point cloud corresponding to the pitch angle of the sub-interval endpoint and the detected 3D lane line point cloud is determined.

[0098] In this example, the projected 3D lane line point cloud corresponding to the sub-interval endpoint pitch angle with the smallest error between it and the detected 3D lane line point cloud is denoted as 3D lane line point cloud j. If the error between 3D lane line point cloud j and the detected 3D lane line point cloud is less than the error threshold, the pitch angle of the sub-interval endpoint corresponding to 3D lane line point cloud j is determined as the target pitch angle related to the target environment image. If the error between 3D lane line point cloud j and the detected 3D lane line point cloud is not less than the error threshold, the pitch angle of the sub-interval endpoint corresponding to 3D lane line point cloud j is determined as the reference pitch angle for the (k+1)th pitch angle lookup operation.

[0099] In this embodiment of the disclosure, for a pitch angle, in order to determine the error between the projected 3D lane line point cloud obtained by inversely projecting multiple lane line pixels of the target lane line onto the vehicle coordinate system of the target vehicle at the pose corresponding to the pitch angle and the 3D points in the detected 3D lane line point cloud, the projected 3D lane line point cloud and the detected 3D lane line point cloud can be sampled using the same sampling method, such as uniform sampling, to obtain multiple sampling points of the projected 3D lane line point cloud and multiple sampling points of the detected 3D lane line point cloud. Each sampling point of the projected 3D lane line point cloud corresponds to a pair of sampling points, resulting in multiple pairs of sampling points. The pair of sampling points corresponding to one sampling point of the projected 3D lane line point cloud includes: one sampling point of the projected 3D lane line point cloud and the sampling point in the detected 3D lane line point cloud that is closest to one sampling point of the projected 3D lane line point cloud. For each pair of sampling points, the distance between the two sampling points in the pair is calculated to obtain the distance of the pair. The distances of each pair of sampling points are summed to obtain the error.

[0100] In step S103, when it is determined that there is a target pitch angle related to the target environment image, the position of the cone in the target environment image in the image coordinate system of the target environment image is transformed into the corrected position of the cone in the vehicle coordinate system of the target vehicle by using the target corrected pose corresponding to the target pitch angle related to the target environment image.

[0101] For a cone in the target environment image, the position of the cone in the image coordinate system of the target environment image is: the position of the center point of the detection box surrounding the cone output by the target detection network in the image coordinate system of the target environment image.

[0102] In this embodiment of the disclosure, the current pitch angle in the current pose of the target vehicle's camera can be replaced with a target pitch angle related to the target environment image to obtain the target corrected pose corresponding to the target pitch angle related to the target environment image. Alternatively, the target corrected pose corresponding to the target pitch angle related to the target environment image can be obtained using the following formula:

[0103]

[0104] Candidate pitch angle corresponding to candidate corrected pose =

[0105] in, This represents the current position and orientation of the target vehicle. The target pitch angle is related to the target environment image.

[0106] Referring to Figure 2, a flowchart of another information acquisition method provided in an embodiment of this disclosure is shown.

[0107] In step S201, the segmentation results of the target environment image and the detected three-dimensional lane line point cloud of the target lane line on the target road on which the target vehicle is traveling are obtained.

[0108] Another information acquisition method provided in this disclosure embodiment can be performed within a target time period. Each environmental image captured by the target vehicle's camera within the target time period includes traffic cones on the road where the target vehicle is traveling. The target environmental image can be any environmental image captured within the target time period. For each environmental image captured within the target time period, at least steps S201, S202, and S204 can be executed. The start time of the target time period can be the acquisition time of the first environmental image captured within the target time period, i.e., the earliest acquisition time of the environmental image. The end time of the target time period can be the acquisition time of the last environmental image captured within the target time period, i.e., the latest acquisition time of the environmental image.

[0109] Within the target time period, the first environmental image can be used as the environmental image for particle initialization. For each other environmental image acquired within the target time period besides the first environmental image, cone tracking is performed corresponding to that other image, i.e., for that other environmental image, step S204 is performed.

[0110] In step S202, based on the current pose of the target vehicle's camera and the detected three-dimensional lane line point cloud of the target lane line on the target road on which the target vehicle is traveling, it is determined whether there is a target pitch angle related to the target environment image.

[0111] In step S203, when it is determined that there is a target pitch angle related to the target environment image, the position of the cone in the target environment image in the image coordinate system of the target environment image is transformed into the corrected position of the cone in the vehicle coordinate system of the target vehicle by using the target corrected pose corresponding to the target pitch angle related to the target environment image.

[0112] In one possible implementation, the outline of the target cone is determined based on the segmentation result of the target environment image, wherein the target cone is any cone in the target environment image; the position of the intersection of the outline of the target cone and the perpendicular bisector of the detection box surrounding the target cone is determined as the position of the target cone in the image coordinate system of the target environment image.

[0113] In step S204, cone tracking corresponding to the target environment image is performed.

[0114] Considering the possibility of a longitudinal deviation between the corrected position of the cone in the target environment image under the vehicle's coordinate system and its actual position in the target environment image under the target vehicle's coordinate system, to ensure a stable output of the cone's position in the target vehicle's coordinate system, step S204 is executed to track the cone's position in the target vehicle's coordinate system, i.e., to perform cone tracking corresponding to the target environment image, obtaining and outputting the stable position of the cone, i.e., the tracked position of the target cone.

[0115] Step S204 includes: Step S2041-Step S2042.

[0116] In step S2041, a tracking operation corresponding to the target environment image is performed.

[0117] In step S2042, the target particle is found from the target particle set, and the position of the target particle is used as the tracking position of the corresponding cone in the target environment image. The tracking position of the corresponding cone in the target environment image is then output.

[0118] In one possible implementation, for a target particle, the cone whose position is closest to the target particle in the target environment image in the vehicle coordinate system of the target vehicle can be determined, and the position of the target particle can be used as the tracking position of the cone whose position is closest to the target particle in the target environment image in the vehicle coordinate system of the target vehicle.

[0119] In this embodiment of the disclosure, the particles in the target particle set have positions in the vehicle coordinate system of the target vehicle.

[0120] The target particle set does not refer to a specific set of particles. It should be understood as a variable. Because barrel tracking is performed sequentially for each other environment image, the particles in the target particle set will change. In embodiments of this disclosure, the composition of the target particle set should be determined in conjunction with the context when performing operations involving the target particle set.

[0121] In this embodiment of the disclosure, the target particle set is initialized to an initial particle set before the first cone tracking is performed.

[0122] The first cone tracking can refer to the cone tracking corresponding to the second environmental image acquired within the target time period. In other words, cone tracking begins with the second environmental image acquired within the target time period. All environmental images acquired within the target time period are sorted from earliest to latest acquisition time to obtain the order of all environmental images acquired within the target time period. The first environmental image acquired within the target time period is the earliest acquired environmental image, and the last environmental image acquired within the target time period is the latest acquired environmental image.

[0123] In the disclosed embodiment, the initial particle set is obtained by generating particles based on the corresponding position of the cone in the vehicle coordinate system of the target vehicle in the environmental image used for particle initialization.

[0124] The environmental image used for particle initialization can be the first environmental image acquired within the target time period.

[0125] In this embodiment of the disclosure, for each cone in the first environmental image acquired within the target time period, if the corrected position of the cone in the target vehicle's vehicle coordinate system is obtained, then the corresponding position of the cone in the target vehicle's vehicle coordinate system is: the corrected position of the cone in the target vehicle's vehicle coordinate system; if the corrected position of the cone in the target vehicle's vehicle coordinate system is not obtained, then the corresponding position of the cone in the target vehicle's vehicle coordinate system is: the original position of the cone in the target vehicle's vehicle coordinate system. The original position of the cone in the target vehicle's vehicle coordinate system refers to the position obtained by transforming the position of the cone in the image coordinate system of the target environmental image to the vehicle coordinate system of the target vehicle, based on the current pose of the target vehicle.

[0126] In this embodiment of the disclosure, in order to obtain an initial particle set, for each cone in the first environmental image captured by the camera of the target vehicle within a target time period, particles are generated based on the corresponding position of the cone in the vehicle's coordinate system. The initial particle set is composed of the particles generated based on the corresponding positions of each cone in the first environmental image captured within the target time period in the vehicle's coordinate system.

[0127] In this embodiment of the disclosure, for a cone i in the target environment image, when generating particles based on the corresponding position of the cone i in the vehicle coordinate system of the target vehicle, a square with a preset side length can be determined, whose center point is the corresponding position of the cone i in the vehicle coordinate system. The corresponding position of the cone i in the vehicle coordinate system, the corner points of the square, and the center of each side of the square are respectively used as particle positions, for a total of 9 particle positions. For each particle position, a particle is generated at that particle position. A total of 9 particles are generated.

[0128] Wherein, cone i is any cone in any environmental image collected within the target time period.

[0129] In this embodiment of the disclosure, step S2041 includes: step S20411 and step S20412.

[0130] In step S20411, when there is a first particle from the target particle set in the target area associated with the target cone, the weight of the first particle is updated according to the distance between the first particle and the target cone, the weight of the target cone, and the current weight of the first particle. When there is a second particle in the target particle set, the weight of the second particle is reduced. The second particle is a particle that is not in the target area associated with any cone in the target environment image.

[0131] In step S20412, when no particles from the target particle set exist within the target area related to the target cone, new particles generated based on the corresponding position of the target cone in the vehicle's coordinate system are added to the target particle set. Specifically, when generating new particles based on the corresponding position of the target cone in the vehicle's coordinate system, a square with a preset side length and a center point corresponding to the target cone's position in the vehicle's coordinate system is defined. The corresponding position of the target cone in the vehicle's coordinate system, the corner points of this square, and the center of each side of the square are respectively taken as particle positions. For each particle position, a particle is generated at that position, and this particle is considered a new particle. This particle is then added to the target particle set. Thus, the target particle set is updated. When the update is complete, the target particle set consists of all particles in the target particle set at the start of step S2041 and each new particle.

[0132] The target cone is any cone in the target environment image. Particles that are in the target region associated with the target cone and come from the target particle set can be called the first particle, and particles that are not in the target region associated with any cone in the target environment image can be called the second particle.

[0133] The target area associated with the target cone is an area whose center point is located at the corresponding position of the target cone in the vehicle coordinate system of the target vehicle and whose shape is a preset shape.

[0134] For each first particle, update the weight of the first particle based on its distance from the target cone, the weight of the target cone, and its current weight.

[0135] The current weight of the first particle can refer to the weight of the first particle at the moment when the weight of the first particle is updated based on the distance between the first particle and the target cone, the weight of the target cone, and the current weight of the first particle.

[0136] For a first particle, the distance between the first particle and the target cone can refer to the distance between the position of the first particle in the vehicle's coordinate system and the corresponding position of the target cone in the vehicle's coordinate system.

[0137] In this embodiment of the disclosure, each cone in the target environment image has a weight.

[0138] In one possible implementation, each cone in the target environment image has the same preset weight.

[0139] In another possible implementation, for a cone in the target environment image, when the cone has a corrected position in the vehicle coordinate system of the target vehicle, a first weight is used as the weight of the cone; when the cone does not have a corrected position in the vehicle coordinate system of the target vehicle, a second weight is used as the weight of the target cone, wherein the second weight is less than the first weight.

[0140] The second weight is less than the first weight because it takes into account that when the cone has a corrected position in the target vehicle's coordinate system, the higher the reliability of the cone's corrected position in the target vehicle's coordinate system as the tracking position, the more important the corrected position of the cone in the target vehicle's coordinate system is in the cone tracking process. Therefore, a higher weight is assigned to the cone, improving the accuracy of the corresponding results obtained from the cone's corrected position in the target vehicle's coordinate system in cone tracking, and improving the accuracy of the cone's tracking position obtained through cone tracking.

[0141] As an example, a cone with a corrected position in the target vehicle's body coordinate system has a weight of 1, while a cone without a corrected position in the target vehicle's body coordinate system has a weight of 0.5.

[0142] For a cone in the target environment image, each particle generated based on the position of the cone has an initial weight. That is, for a particle, the weight of the particle is the initial weight after the particle is generated and before the first update of the particle's weight.

[0143] In the disclosed embodiment, for a cone in the target environment image, the weight of a particle generated based on the cone's corresponding position in the target vehicle's coordinate system is less than the cone's weight. The weight of the particle generated based on the cone's corresponding position in the target vehicle's coordinate system can be a multiple of the cone's weight. Among all particles generated based on the cone's corresponding position in the target vehicle's coordinate system, the particle whose position corresponds to the cone's position in the target vehicle's coordinate system has a greater weight than the weights of all other particles. The weight of the particle whose position corresponds to the cone's position in the target vehicle's coordinate system can be a multiple of the weights of all other particles.

[0144] As an example, for a cone in the target environment image, the weight of the cone is denoted as w. The weight of the particle at the corresponding position of the cone in the vehicle coordinate system of the target vehicle is 4*w / 9, and the weight of other particles is w / 9.

[0145] Referring to Figure 3, it shows an example of the effect of generating particles based on the corrected position of the cone in the target vehicle's body coordinate system.

[0146] In this example, based on the corrected position of a cone in the target environment image within the vehicle's coordinate system, a total of 9 particles are generated. A square with a preset side length is defined, with its center point at the corresponding position of the cone in the vehicle's coordinate system. The cone's position in the vehicle's coordinate system, the corner points of the square, and the center of each side of the square are used as particle positions, resulting in a total of 9 particle positions. For each particle position, a particle is generated at that position, for a total of 9 particles. The cone's weight is denoted as w. The weight of particle 301 at the cone's corresponding position in the vehicle's coordinate system is 4*w / 9, and the weight of other particles 302 is w / 9.

[0147] In one possible implementation, in step S20411, for a first particle, updating the weight of the first particle according to the distance between the first particle and the target cone, the weight of the target cone, and the current weight of the first particle includes: determining the coefficient corresponding to the distance between the first particle and the target cone based on the set correlation between the distance between the particle and the target cone and the coefficient; multiplying the coefficient corresponding to the distance between the first particle and the target cone, the weight of the target cone, and a preset constant to obtain a product; adding the product to the current weight of the first particle to obtain the updated weight of the first particle; and updating the weight of the first particle to the updated weight of the first particle.

[0148] As an example, for a first particle within the target region associated with the target cone, the weight of that first particle can be updated using the following formula:

[0149]

[0150] in, The weight of the target cone, Let be the distance between the first particle and the target cone. This represents the current weight of the first particle. This is the updated weight for the first particle.

[0151] In step S20412, for a second particle, reducing the weight of the second particle is achieved by subtracting a preset weight reduction amount from the current weight of the second particle. Here, the current weight of the second particle is the weight of the second particle at the moment the weight reduction begins.

[0152] As an example, the reduction could be 0.2.

[0153] In step S20412, when there are no particles from the target particle set in the target area, new particles generated according to the corresponding position of the target cone in the vehicle coordinate system are added to the target particle set.

[0154] In one possible implementation of step S2042, multiple third particles are identified from the target particle set, with each third particle having a weight greater than a weight threshold. Multiple fourth particles are then identified from these third particles. These fourth particles are sorted in descending order of their weights. After sorting, each of the top target number of fourth particles is identified as a target particle, where the target number is the number of cones in the target environment image. For a third particle, if it satisfies a first condition, it is identified as a fourth particle. The first condition includes: there are other particles within the target region associated with the third particle, and these other particles within the target region associated with the third particle do not appear in other target regions. Here, the other particles are particles other than the third particle, and the other target regions are regions associated with particles other than the third particle. For a third particle, if the third particle satisfies the second condition, then the third particle is determined to be the fourth particle. The second condition includes: there are other particles in the target area related to the third particle; the corresponding other particles in the target area related to the third particle appear in the corresponding other target area; and the weight of the third particle is greater than the weight of the other particles related to the corresponding other target area.

[0155] In this embodiment of the disclosure, for a particle, the region associated with the particle is a region whose center point is located in the vehicle coordinate system of the target vehicle and whose shape is a preset shape.

[0156] As an example, the preset shape is a circle.

[0157] Referring to Figure 4, it shows a schematic diagram of the effect of performing a tracking operation corresponding to the target environment image.

[0158] Figure 4 shows particle 401, whose position corresponds to the location of cone 1 in the vehicle's coordinate system. The target region associated with cone 401 is a circular region 402. Particle 403 is the first particle from the target particle set within the target region associated with the target cone, and particle 404 is the second particle. Figure 4 also shows particle 405, whose position corresponds to the location of cone 2 in the vehicle's coordinate system. No particles from the target particle set exist within the target region associated with cone 2. New particles are generated based on the location of cone 2 in the target vehicle's coordinate system, and Figure 4 shows the new particle 406. The new particles generated based on the location of cone 2 in the target vehicle's coordinate system are added to the target particle set. This updates the particles in the target particle set.

[0159] In another possible implementation of step S2042, step S2042 includes: step S20421.

[0160] In step S20421, when there are particles with a weight less than 0 in the target particle set, the particles with a weight less than 0 are deleted from the target particle set; candidate particles with a weight greater than the weight threshold are found from the target particle set; and the target particle is found from multiple candidate particles.

[0161] As an example, the weight threshold is 1.

[0162] In one possible implementation, finding the target particle from multiple candidate particles includes: sorting the candidate particles according to their weights from largest to smallest to obtain the order of the candidate particles; iteratively performing the search operation until a target number of target particles are found, where the target number is the number of cones in the target environment image. The i-th search operation includes: determining whether there are other particles not associated with the candidate particle in the target region related to the candidate particle targeted by the i-th search operation, wherein the candidate particle targeted by the first search operation is the particle with the largest weight among the multiple candidate particles, and the candidate particles targeted by subsequent search operations are the next candidate particles after the candidate particles targeted by the previous search operation after the previous search operation. Subsequent search operations are any search operations other than the first search operation, and the next candidate particles after the candidate particles targeted by the previous search operation are indicated by this order; if there are other particles not associated with the candidate particle in the target region related to the candidate particle targeted by the i-th search operation, the candidate particle targeted by the i-th search operation is determined as the target particle. Furthermore, other particles within the target region related to the candidate particle targeted by the i-th search operation that have not been associated with the candidate particle are associated with the candidate particle targeted by the i-th search operation. The next candidate particle targeted by the i-th search operation is determined as the candidate particle targeted by the next search operation, i.e., the (i+1)-th search operation. If there are no other particles within the target region related to the candidate particle targeted by the i-th search operation that have not been associated with the candidate particle, the next candidate particle targeted by the i-th search operation is determined as the candidate particle targeted by the next search operation, i.e., the (i+1)-th search operation. The next candidate particle targeted by the i-th search operation is indicated by this order.

[0163] Among them, the target region associated with a candidate particle is: a region whose center point is located in the vehicle coordinate system of the target vehicle and whose shape is a preset shape.

[0164] In one possible implementation, a circular region with a center point located at the position of the candidate particle targeted by the search operation in the vehicle coordinate system of the target vehicle and a preset second radius is defined as the target region associated with the candidate particle targeted by the search operation.

[0165] As an example, the second radius is preset to 0.9m.

[0166] Referring to Figure 5, it shows a schematic diagram of the effect of performing an example of a lookup operation.

[0167] Schematic diagram of the effect. Figure 5 shows candidate particles 501, 503, and 505. The weight of candidate particle 501 is 1.34. The weight of candidate particle 503 is 1.31. The weight of candidate particle 505 is 1.21.

[0168] In this example, the first search operation targets candidate particle 501, the second search operation targets candidate particle 502, and the third search operation targets candidate particle 503.

[0169] During the first search operation, it was determined that there were other particles in the target region 502 that were not associated with the candidate particle 501. Therefore, the candidate particle 501 was determined as the target particle.

[0170] During the second search operation, it was determined that there were no other particles in the target region 504 related to candidate particle 503 that had not been associated with the candidate particle. Therefore, during the second search operation, candidate particle 503 was not identified as the target particle.

[0171] During the third search operation, it was determined that there were other particles in the target region 506 related to the candidate particle 505 that had not been associated with the candidate particle. Therefore, the candidate particle 505 was determined as the target particle.

[0172] This disclosure also provides an information acquisition device configured to implement the above-described method embodiments and implementation methods, which will not be repeated hereafter. As used below, the term "unit" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments can be implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated. The device in this disclosure is presented in the form of a functional unit, which refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that execute one or more software or fixed programs, and / or other devices that can provide the above-described functions.

[0173] The information acquisition device includes:

[0174] The acquisition unit is configured to acquire the segmentation result of the target environment image captured by the camera of the target vehicle and the detected three-dimensional lane line point cloud of the target lane line on the target road on which the target vehicle is traveling, wherein the segmentation result indicates multiple lane line pixels of the target lane line and the detected three-dimensional lane line point cloud is detected based on the target environment image.

[0175] The determining unit is configured to determine whether there is a target pitch angle related to the target environment image based on the current pose of the camera, the segmentation result, and the detected 3D lane line point cloud. The error between the target projected 3D lane line point cloud obtained by inversely projecting the plurality of lane line pixels onto the vehicle coordinate system of the target vehicle using the target corrected pose corresponding to the target pitch angle and the detected 3D lane line point cloud is less than an error threshold. The target corrected pose is obtained by replacing the current pitch angle in the current pose of the camera with the target pitch angle.

[0176] The conversion unit is configured to, when the target pitch angle is determined to exist, use the target corrected pose to transform the position of the cone in the target environment image in the image coordinate system of the target environment image to the corrected position of the cone in the vehicle coordinate system.

[0177] The information acquisition device may also include:

[0178] The cone position acquisition unit is configured to determine the outline of the target cone based on the segmentation result of the target environment image, wherein the target cone is any cone in the target environment image; and to determine the position of the intersection point of the outline of the target cone and the perpendicular bisector of the detection box surrounding the target cone as the position of the target cone in the image coordinate system of the target environment image.

[0179] The information acquisition device may also include:

[0180] The tracking unit is configured to perform barrel tracking corresponding to a target environment image, the barrel tracking including:

[0181] A tracking operation corresponding to the target environment image is performed, and after the tracking operation, target particles are found from the target particle set, and the positions of the target particles are used as the tracking positions of the corresponding cones in the target environment image. The tracking positions are then output. The target particle set is initialized to an initial particle set before the first cone tracking operation. The initial particle set is obtained by generating particles based on the corresponding positions of the cones in the vehicle coordinate system in the environment image used for particle initialization. The tracking operation includes:

[0182] When there is a first particle from the target particle set in the target region associated with the target cone, the weight of the first particle is updated according to the distance between the first particle and the target cone, the weight of the target cone, and the current weight of the first particle. When there is a second particle in the target particle set, the weight of the second particle is reduced. The second particle is a particle that is not in the target region associated with any cone in the target environment image.

[0183] When there are no particles from the target particle set within the target area, new particles generated based on the corresponding position of the target cone in the vehicle coordinate system will be added to the target particle set.

[0184] The information acquisition device may also include:

[0185] The weighting unit is configured to use a first weight as the weight of the target cone when the target cone has a corrected position in the vehicle coordinate system; and to use a second weight as the weight of the target cone when the target cone does not have a corrected position in the vehicle coordinate system, wherein the second weight is less than the first weight.

[0186] The information acquisition device may also include:

[0187] The first target area determination unit is configured to determine a circular area with a center point at the corresponding position of the target cone in the vehicle coordinate system and a radius of a preset first radius as the target area associated with the target cone.

[0188] The tracking unit can also be configured to delete particles with weights less than 0 from the target particle set when there are particles with weights less than 0 in the target particle set; find multiple candidate particles from the target particle set, where the weights of the candidate particles are greater than a weight threshold; and find the target particle from the multiple candidate particles.

[0189] The tracking unit can also be configured to sort multiple candidate particles according to their weights from largest to smallest to obtain the order of the candidate particles; iteratively perform a search operation until a target number of target particles are found, where the target number is the number of cones in the target environment image. The search operation includes: determining whether there are other particles in the target region related to the candidate particle targeted by the search operation that have not been associated with the candidate particle, wherein the candidate particle targeted in the first search operation is the particle with the largest weight among the multiple candidate particles, and the candidate particle targeted in subsequent search operations is the next candidate particle targeted in the previous search operation of the subsequent search operation, wherein the next candidate particle targeted in the previous search operation is indicated by the order; in the target region related to the candidate particle targeted by the search operation If there are no other particles in the target area that have not been associated with the candidate particle, the candidate particle targeted by the search operation is determined as the target particle, and other particles in the target area related to the candidate particle targeted by the search operation that have not been associated with the candidate particle are associated with the candidate particle targeted by the search operation, and the next candidate particle of the candidate particle targeted by the search operation is determined as the candidate particle targeted by the next search operation; if there are no other particles in the target area related to the candidate particle targeted by the search operation that have not been associated with the candidate particle, the next candidate particle of the candidate particle targeted by the search operation is determined as the candidate particle targeted by the next search operation, wherein the next candidate particle of the candidate particle targeted by the search operation is indicated by the sequence.

[0190] The information acquisition device may also include:

[0191] The second target region determination unit is configured to determine a circular region with a center point that is the position of the candidate particle targeted by the search operation in the vehicle coordinate system and a radius of a preset second radius as the target region related to the candidate particle targeted by the search operation.

[0192] Referring to Figure 6, Figure 6 is a schematic diagram of the hardware structure of a computer device according to an embodiment of this disclosure. The computer device includes one or more processors 10, a memory 20, and interfaces configured to connect the various components, including high-speed interfaces and low-speed interfaces. The various components are interconnected via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on an external input / output device (such as a display device coupled to the interface). In some alternative embodiments, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules. Similarly, multiple devices can be connected, each providing some necessary operations (e.g., as a server array, a set of blade servers, or a multiprocessor system). The processor 10 can be a central processing unit, a network processor, or a combination thereof. The processor 10 may also include hardware chips. The hardware chips can be application-specific integrated circuits (ASICs), programmable logic devices (PLDs), or combinations thereof. The programmable logic devices can be complex programmable logic devices (CLPs), field-programmable gate arrays (FPGAs), general-purpose array logic (GPRS), or any combination thereof. The memory 20 stores instructions executable by at least one processor 10 to perform the methods shown in the above embodiments. The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on vehicle usage, etc. Furthermore, the memory 20 may include high-speed random access memory and non-transient memory, such as at least one disk storage device, flash memory device, or other non-transient solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, which can be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof. The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk, or solid-state drive; the memory 20 may also include combinations of the above types of memory. The computer device also includes an input device 30 and an output device 40. The processor 10, memory 20, input device 30, and output device 40 may be connected via a bus or other means. The input device 30 can receive input digital or character information, and generate key signal inputs related to user settings and function control of the computer device, such as touch screen, keypad, mouse, trackpad, touchpad, joystick, one or more mouse buttons, trackball, joystick, etc.The output device 40 may include a display device, an auxiliary lighting device (e.g., an LED), and a haptic feedback device (e.g., a vibration motor). The display device includes, but is not limited to, liquid crystal displays, light-emitting diodes, displays, and plasma displays. In some alternative embodiments, the display device may be a touchscreen.

[0193] This disclosure also provides a computer-readable storage medium in which the methods described in this disclosure can be implemented in hardware or firmware, or implemented as recordable on a storage medium, or implemented as computer code originally stored on a remote storage medium or a non-transitory machine-readable storage medium and subsequently stored on a local storage medium after being downloaded over a network. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; the storage medium can also include combinations of the above types of memory. It is understood that a computer, processor, microprocessor controller, or programmable hardware includes storage components capable of storing or receiving software or computer code that, when accessed and executed by the computer, processor, or hardware, implements the methods shown in the above embodiments.

[0194] A portion of the embodiments disclosed herein can be applied as a computer program product, such as computer instructions. When executed by a computer, these instructions, through the operation of the computer, can invoke or provide methods and / or technical solutions according to the embodiments of this disclosure. Those skilled in the art will understand that the forms in which computer instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Accordingly, the ways in which computer instructions are executed by a computer include, but are not limited to: the computer directly executing the instruction; or the computer compiling the instruction and then executing the corresponding compiled program; or the computer reading and executing the instruction; or the computer reading and installing the instruction and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0195] The above embodiments are merely illustrative of the technical solutions, and the protection scope of this disclosure is not limited thereto. Equivalent substitutions or modifications made by those skilled in the art based on the embodiments of this disclosure are all within the protection scope of this disclosure.

[0196] The various modules, sub-modules, units, and sub-units mentioned in the above technical description do not refer solely to functional divisions or virtual components at the pure software logic level. Rather, they can be implemented as physical devices, components, parts, or combinations thereof with specific hardware or physical structures. They can also be implemented using a combination of hardware and software, i.e., hardware executes corresponding software instructions to achieve the corresponding function. The aforementioned modules, sub-modules, units, and sub-units can be exemplified as Application-Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), Programmable Logic Devices (PLDs), Discrete Logic Circuits, Central Processing Units (CPUs), Microprocessors, Microcontrollers, Microcontrollers, Digital Signal Processors (DSPs), Dedicated Function Chips, Storage Devices, Interface Circuits, and other hardware entities or hardware component combinations with functions such as data processing, signal transmission, and logical operations. Furthermore, the aforementioned hardware entities can be equipped with corresponding software programs, firmware, or instruction sets to form a functional carrier that combines hardware foundation and software execution capabilities.

Claims

1. An information acquisition method, comprising: The segmentation result of the target environment image captured by the camera of the target vehicle and the detected three-dimensional lane line point cloud of the target lane line on the target road on which the target vehicle is traveling are obtained. The segmentation result indicates multiple lane line pixels of the target lane line, and the detected three-dimensional lane line point cloud is detected based on the target environment image. Based on the current pose of the camera, the segmentation result, and the detected 3D lane line point cloud, it is determined whether there is a target pitch angle related to the target environment image. The target projected 3D lane line point cloud obtained by inversely projecting the multiple lane line pixels onto the vehicle coordinate system of the target vehicle using the target corrected pose corresponding to the target pitch angle has an error less than an error threshold with the detected 3D lane line point cloud. The target corrected pose is obtained by replacing the current pitch angle in the current pose of the camera with the target pitch angle. When the target pitch angle is determined to exist, the position of the cone in the target environment image in the image coordinate system of the target environment image is transformed into the corrected position of the cone in the vehicle coordinate system by using the target corrected pose.

2. The method according to claim 1, further comprising: Based on the segmentation results of the target environment image, the outline of the target cone is determined, where the target cone is any cone in the target environment image; The position of the target cone in the image coordinate system is determined by the intersection of the outline of the target cone and the perpendicular bisector of the detection box surrounding the target cone.

3. The method according to claim 1 or 2, further comprising: Performing cone tracking corresponding to the target environment image, the cone tracking includes: A tracking operation corresponding to the target environment image is performed, and after the tracking operation, target particles are found from the target particle set, and the positions of the target particles are used as the tracking positions of the corresponding cones in the target environment image. The tracking positions are then output. The target particle set is initialized to an initial particle set before the first cone tracking operation. The initial particle set is obtained by generating particles based on the corresponding positions of the cones in the vehicle coordinate system in the environment image used for particle initialization. The tracking operation includes: When there is a first particle from the target particle set in the target region associated with the target cone, the weight of the first particle is updated according to the distance between the first particle and the target cone, the weight of the target cone, and the current weight of the first particle. When there is a second particle in the target particle set, the weight of the second particle is reduced. The second particle is a particle that is not in the target region associated with any cone in the target environment image. When there are no particles from the target particle set in the target area related to the target cone, new particles generated according to the corresponding position of the target cone in the vehicle coordinate system will be added to the target particle set.

4. The method according to claim 3, further comprising: When the target cone has a corrected position in the vehicle coordinate system, the first weight is used as the weight of the target cone; When the target cone does not have a corrected position in the vehicle coordinate system, the second weight is used as the weight of the target cone, wherein the second weight is less than the first weight.

5. The method according to claim 3, further comprising: The position of the center point is the corresponding position of the target cone in the vehicle coordinate system, and a circular area with a preset first radius is defined as the target area related to the target cone.

6. The method according to claim 3, wherein, The target particles identified from the target particle set include: If there are particles with a weight less than 0 in the target particle set, remove the particles with a weight less than 0 from the target particle set. Multiple candidate particles are identified from the target particle set, where the weight of each candidate particle is greater than a weight threshold. The target particle is identified from multiple candidate particles.

7. The method according to claim 6, wherein, Finding the target particle from multiple candidate particles includes: The candidate particles are sorted from largest to smallest according to their weights to obtain the order of the candidate particles. The search operation is performed iteratively until a target number of target particles are found, where the target number is the number of cones in the target environment image. The search operation includes: Determine whether there are other particles in the target region related to the candidate particle targeted by the search operation that have not been associated with the candidate particle. The candidate particle targeted by the first search operation is the particle with the largest weight among multiple candidate particles. The candidate particle targeted by the non-first search operation is the next candidate particle after the candidate particle targeted by the previous search operation of the non-first search operation. The next candidate particle after the candidate particle targeted by the previous search operation is indicated by the order. If there are other particles in the target region related to the candidate particle targeted by the search operation that have not been associated with the candidate particle, the candidate particle targeted by the search operation is determined as the target particle, and other particles in the target region related to the candidate particle targeted by the search operation that have not been associated with the candidate particle are associated with the candidate particle targeted by the search operation, and the next candidate particle of the candidate particle targeted by the search operation is determined as the candidate particle targeted by the next search operation. If there are no other particles in the target region related to the candidate particle targeted by the search operation that have not been associated with the candidate particle, the next candidate particle targeted by the search operation is determined as the candidate particle targeted by the next search operation, wherein the next candidate particle targeted by the search operation is indicated by the order.

8. The method according to claim 7, further comprising: The position of the center point is the position of the candidate particle targeted by the search operation in the vehicle coordinate system, and a circular area with a preset second radius is determined as the target area related to the candidate particle targeted by the search operation.

9. An information acquisition device, comprising: The acquisition unit is configured to acquire the segmentation result of the target environment image captured by the camera of the target vehicle and the detected three-dimensional lane line point cloud of the target lane line on the target road on which the target vehicle is traveling, wherein the segmentation result indicates multiple lane line pixels of the target lane line and the detected three-dimensional lane line point cloud is detected based on the target environment image. The determining unit is configured to determine whether there is a target pitch angle related to the target environment image based on the current pose of the camera, the segmentation result, and the detected 3D lane line point cloud. The error between the target projected 3D lane line point cloud obtained by inversely projecting the plurality of lane line pixels onto the vehicle coordinate system of the target vehicle using the target corrected pose corresponding to the target pitch angle and the detected 3D lane line point cloud is less than an error threshold. The target corrected pose is obtained by replacing the current pitch angle in the current pose of the camera with the target pitch angle. The conversion unit is configured to, when the target pitch angle is determined to exist, use the target corrected pose to transform the position of the cone in the target environment image in the image coordinate system of the target environment image to the corrected position of the cone in the vehicle coordinate system.

10. A computer device installed in a vehicle, comprising: A memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, the processor executing the computer instructions to perform the method of any one of claims 1 to 8.

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

12. A computer program product comprising computer instructions for causing a computer to perform the method of any one of claims 1 to 8.

Images