A SLAM backend pose optimization method based on graph neural network

Abstract

The present invention relates to the technical field of simultaneous positioning and map construction, and discloses a SLAM backend posture optimization method based on a graph neural network, comprising the following steps: extracting key feature points of an image stream continuously input into a SLAM system and recording feature descriptors; projecting the key feature points into a three-dimensional space, obtaining corresponding points of left and right camera images through stereo matching and triangulating them to obtain the three-dimensional coordinates of the key feature points; selecting key frames, recording the key feature points and feature descriptors in the key frames, and judging whether the number of key frames is greater than a preset threshold; determining the co-viewing relationship between the key frames through key feature point matching, obtaining a graph topology structure of the key frames and inputting it into a graph neural network for posture optimization; sorting the key frames in a key frame database in positive order according to the size of timestamps, and retaining the second half of the key frames in the positive order for the next round of data collection and optimization; the method improves the key frame posture optimization accuracy and improves the performance of the SLAM system.

Description

Technical Field

[0001] The present invention relates to the technical field of simultaneous positioning and map construction, and in particular to a SLAM back-end pose optimization method based on graph neural network. Background Art

[0002] Positioning and navigation in mobile robots and autonomous driving primarily rely on simultaneous localization and mapping (SLAM) technology. Equipped with specific sensors, autonomous unmanned mobile devices can use SLAM systems to construct a global map of an unknown environment and achieve precise positioning and navigation. SLAM technology has important applications in areas such as autonomous driving and drones. In SLAM, finding the correct representation of the environment and using it to estimate the robot's pose trajectory plays a vital role in both positioning and mapping. With technological advancements, various SLAM technologies are gradually maturing. Pose graph optimization (PGO) is a key step in the backend of SLAM systems. However, existing pose optimization methods are typically modeled as a nonlinear least squares problem. When the graph is large, the entire pose optimization process consumes a significant amount of computational resources. Furthermore, due to the presence of noise and outliers in the environmental data collected by sensors, traditional optimization methods are highly susceptible to erroneous data and suffer from poor robustness. Summary of the Invention

[0003] In response to the above-mentioned deficiencies in the prior art, the present invention provides a SLAM backend pose optimization method based on graph neural network.

[0004] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is:

[0005] A SLAM backend pose optimization method based on graph neural network includes the following steps:

[0006] S1, using feature point detection method to extract key feature points of the image stream continuously input into the SLAM system, and record the feature descriptors of each key feature point;

[0007] S2. Project the key feature points into three-dimensional space, obtain corresponding points between the left camera image and the right camera image in the continuous image stream through stereo matching, obtain matching points of the key feature points, and triangulate the matching points to obtain the three-dimensional coordinates of the key feature points;

[0008] S3, set the first frame in the image stream continuously input to the SLAM system as a key frame, and determine whether the average parallax between the current frame and the previous key frame in the image stream continuously input to the SLAM system exceeds a set threshold. If so, use the current frame as the key frame, record the key feature points extracted from the key frame and the feature descriptors of the key feature points, and establish a key frame database. Otherwise, take the next frame and continue to compare it with the previous key frame;

[0009] S4, determining whether the number of key frames in the key frame database is greater than a preset threshold, if so, executing step S5, otherwise executing step S1;

[0010] S5. Define a graph structure, and perform key feature point matching based on the key frames, key frame feature points, and feature descriptors of the key frame feature points in the key frame database, determine the co-viewing relationship between the key frames, and obtain a graph topology structure of the key frames;

[0011] S6. Input the graph topology of the key frame into the graph neural network for pose optimization, and use the message passing mechanism to iteratively update the key frame pose to obtain the optimized key frame pose;

[0012] S7. After the pose optimization, the key frames in the key frame database are sorted in positive order according to the size of the timestamps. At the same time, the first half of the key frames in the positive order are eliminated, and the second half of the key frames are used for the next round of data collection and optimization, that is, step S1 is re-executed to perform a new round of key feature point extraction of the image stream.

[0013] Furthermore, step S1 specifically includes:

[0014] For each frame of image input to the SLAM system, the ORB feature point detection method is used to detect the corner points and edge features in each frame of image, namely the key feature points, and the positions of the key feature points are obtained, namely:

[0015] p i =(x i ,y i )

[0016] Among them, p i Indicates the position of the key feature point indexed by i, x i Indicates the horizontal coordinate of the key feature point with index i, y i Represents the vertical coordinate of the key feature point with index i;

[0017] At the same time, the feature descriptors of the key feature points are recorded.

[0018] Furthermore, step S2 specifically includes:

[0019] The key feature points extracted in step S1 are projected into the three-dimensional space. The corresponding points of the left camera image and the right camera image in the continuous image stream are obtained through stereo matching, that is, the matching points of the key feature points. The positions of the matching points are triangulated using the intrinsic parameter matrix of the camera to obtain the three-dimensional coordinates of the key feature points, namely:

[0020] P i =(X i ,Y i ,Z i )

[0021] Among them, P i Represents the three-dimensional coordinates of the key feature point indexed by i, X i Indicates the horizontal coordinate of the key feature point indexed by i in the three-dimensional coordinates, Y i Indicates the vertical coordinate of the key feature point indexed by i in the three-dimensional coordinates, Z i Represents the vertical coordinate of the key feature point with index i in three-dimensional coordinates.

[0022] Furthermore, step S5 specifically includes:

[0023] S51. Define a graph structure G = (V, E), where V represents a set of poses of key frames, and E represents a set of co-viewing relationships between key frames;

[0024] S52. According to the defined graph structure, key frames, key frame feature points, and feature descriptors of key frame feature points in the key frame database are obtained, and the key frame feature points of the key frames are matched to obtain a graph topology of the key frames, specifically:

[0025] If the i'th node v in the graph structure i' With the jth node v j Key frame K i' With K j There is a co-viewing relationship between the two key frames, that is, there are matching key feature points between the two key frames, then the i'th node v is constructed i' With the jth node v j Key frame K i' With K j The edge e between i'j , and at the same time, the edge e i'j The weight w i'j Set as keyframe K i' With K j The normalized value of the number of key feature point matches between them is used to obtain the graph topology of the key frame.

[0026] Furthermore, step S6 specifically includes:

[0027] S61. Input the graph topology structure of the key frame obtained in step S5 into the graph neural network for pose optimization. First, use the message passing mechanism to calculate the information transmitted from the adjacent nodes for each node in the graph topology structure of the key frame.

[0028] Among them, the information transmitted from the adjacent nodes is the impact of the neighboring nodes on the current node;

[0029] S62. Based on the information transmitted from the adjacent nodes, the aggregation function is used to summarize and obtain the hidden state of each node in the next layer of the graph neural network;

[0030] S63. According to the hidden state of each node in the next layer of the graph neural network, message passing and aggregation are performed in the next layer of the graph neural network. The iterative update continues until the last layer of the graph neural network to obtain the optimized key frame pose, namely:

[0031] T * =(R * ,t * )

[0032] Among them, T * represents the optimized key frame pose, R * represents the optimized rotation matrix, t * Represents the optimized translation vector.

[0033] Furthermore, the calculation formula of the information transmitted from the adjacent nodes in step S61 is:

[0034]

[0035] Among them, m i'j Represents the jth node v j Passed to the i'th node v i' News, f θ represents a message passing function, Indicates the i'th node v i' The k-th hidden state of Represents the jth node v j The k-th hidden state, e i'j Indicates the i'th node v i' With the jth node v j Key frame K i' With K j The edge between i'j Represents edge e i'j The weight of .

[0036] Furthermore, the formula for the hidden state of each node in the next layer of the graph neural network in step S62 is:

[0037]

[0038] in, Indicates the i'th node v i' The k+1th layer hidden state, AGG represents the aggregation function, σ represents the activation function, Indicates the i'th node v i' The set of neighbor nodes.

[0039] The present invention has the following beneficial effects:

[0040] 1. The proposed SLAM backend pose optimization method based on graph neural network is faster and uses less memory when constructing large scenes.

[0041] 2. During end-to-end learning, key feature points and key frames are directly extracted from the original observed image stream data, and the key frame poses are optimized, which reduces intermediate steps, simplifies the process, and makes the key frame poses more accurate, more consistent, and more robust. Therefore, the optimized key frame poses are used to update the map and path planning, which will improve the performance and accuracy of the entire SLAM system.

[0042] 3. Halving the number of keyframes and retaining some processed keyframes ensures smooth data collection and optimization for the next round, while balancing data continuity and efficient resource utilization, improving the accuracy and stability of positioning and mapping. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 This is a flow chart of a SLAM backend pose optimization method based on graph neural network proposed in the present invention;

[0044] Figure 2 Schematic diagram of the training process of graph neural network. DETAILED DESCRIPTION

[0045] The specific embodiments of the present invention are described below to facilitate understanding of the present invention by those skilled in the art. However, it should be clear that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the appended claims, these changes are obvious, and all inventions and creations utilizing the concepts of the present invention are protected.

[0046] like Figure 1 As shown, a SLAM backend pose optimization method based on graph neural network includes the following steps S1-S7:

[0047] S1. Use feature point detection method to extract key feature points of the image stream continuously input into the SLAM system and record the feature descriptors of each key feature point.

[0048] In this embodiment, the SLAM system is a simultaneous positioning and mapping system; wherein, the present invention adopts the ORB feature point detection method to extract key feature points, but in principle, any feature point detection method can be used.

[0049] Specifically, step S1 includes:

[0050] For each frame of image input to the SLAM system, the ORB feature point detection method is used to detect the corner points and edge features in each frame of image, namely the key feature points, and the positions of the key feature points are obtained, namely:

[0051] p i =(x i ,y i )

[0052] Among them, p i Indicates the position of the key feature point indexed by i, x i Indicates the horizontal coordinate of the key feature point with index i, y i Indicates the vertical coordinate of the key feature point with index i.

[0053] At the same time, the feature descriptors of the key feature points are recorded.

[0054] In this embodiment, the purpose of recording the feature descriptor of each key feature point is to facilitate the description of the key feature point so as to match the key feature points between different key frames in subsequent steps.

[0055] S2. Project the key feature points into three-dimensional space, obtain the corresponding points of the left camera image and the right camera image in the continuous image stream through stereo matching, obtain the matching points of the key feature points, and triangulate the matching points to obtain the three-dimensional coordinates of the key feature points.

[0056] Specifically, step S2 includes:

[0057] The key feature points extracted in step S1 are projected into the three-dimensional space. The corresponding points of the left camera image and the right camera image in the continuous image stream are obtained through stereo matching, that is, the matching points of the key feature points. The positions of the matching points are triangulated using the intrinsic parameter matrix of the camera to obtain the three-dimensional coordinates of the key feature points, namely:

[0058] P i =(X i ,Y i ,Z i )

[0059] Among them, P iRepresents the three-dimensional coordinates of the key feature point indexed by i, X i Indicates the horizontal coordinate of the key feature point indexed by i in the three-dimensional coordinates, Y i Indicates the vertical coordinate of the key feature point indexed by i in the three-dimensional coordinates, Z i Represents the vertical coordinate of the key feature point with index i in three-dimensional coordinates.

[0060] In this embodiment, the triangulation formula is: Where b is the camera baseline, d is the parallax, and f is the focal length. represents the horizontal coordinate of the left camera image, represents the horizontal coordinate of the right camera image, represents the vertical coordinate of the left camera image, Indicates the vertical coordinate of the right camera image.

[0061] S3. Set the first frame in the image stream continuously input to the SLAM system as a key frame, and determine whether the average disparity between the current frame and the previous key frame in the image stream continuously input to the SLAM system exceeds a set threshold. If so, set the current frame as the key frame, record the key feature points extracted from the key frame and the feature descriptors of the key feature points, and establish a key frame database. Otherwise, take the next frame and continue comparing it with the previous key frame.

[0062] In the present embodiment, the first frame in the image stream continuously input to the SLAM system is artificially set as a key frame, thereby judging whether there is a key frame in the subsequent frame; at the same time, the key feature points extracted from the identified key frames and the feature descriptors of the key feature points are recorded, thereby establishing a key frame database; the purpose of establishing a key frame database is: when the number of key frames in the key frame database reaches a preset threshold, the optimization process is triggered in the subsequent steps to reduce the amount of calculation, that is, each frame collected by the camera is not optimized, but only the key frames are optimized. Among them, the key frame can be selected based on the change in the number of key feature points and the amount of motion (such as the size of translation and rotation) between adjacent frames. In addition, the posture selected in the present embodiment is the key frame (not all image frames), the purpose of which is to reduce the amount of data, improve optimization efficiency, and meet real-time requirements.

[0063] S4. Determine whether the number of key frames in the key frame database is greater than a preset threshold. If so, execute step S5; otherwise, execute step S1.

[0064] In this embodiment, it is determined whether the number of key frames selected in step S3 is greater than a preset threshold. If so, the process proceeds to the next step for processing. Otherwise, step S1 is executed to continue extracting key feature points from the image stream and recording key frames. The purpose is to ensure that optimization is performed only when enough key frames are accumulated to ensure the stability and effect of the optimization.

[0065] S5. Define the graph structure, and perform key feature point matching based on the key frames, key frame feature points, and feature descriptors of the key frame feature points in the key frame database to determine the co-viewing relationship between the key frames and obtain the graph topology structure of the key frames.

[0066] In this embodiment, when the number of key frames meets the preset threshold condition, a graph structure is defined, that is, G = (V, E), and the edges and weights between key frames are determined by matching key feature points, thereby obtaining the graph topology structure of the key frames.

[0067] Specifically, step S5 includes S51-S52:

[0068] S51. Define a graph structure G = (V, E), where V represents a set of poses of key frames, and E represents a set of co-viewing relationships between key frames.

[0069] In this embodiment, the keyframe pose set V is composed of the poses T of each keyframe, where the pose T of the keyframe includes the translation vector t and the rotation matrix R, that is, T = (R, t). The co-visibility relationship set E between keyframes is composed of the co-visibility relationship edges e between keyframes.

[0070] S52. According to the defined graph structure, key frames, key frame feature points, and feature descriptors of key frame feature points in the key frame database are obtained, and the key frame feature points of the key frames are matched to obtain a graph topology of the key frames, specifically:

[0071] If the i'th node v in the graph structure i' With the jth node v j Key frame K i' With K j There is a co-viewing relationship between the two key frames, that is, there are matching key feature points between the two key frames, then the i'th node v is constructed i' With the jth node v j Key frame K i' With K j The edge e between i'j , and at the same time, the edge e i'j The weight w i'j Set as keyframe K i' With K j The normalized value of the number of key feature point matches between them is used to obtain the graph topology of the key frame.

[0072] In this embodiment, the graph topology of the keyframes obtained after key feature point matching includes all keyframes and the co-viewing relationships between them, providing a basis for subsequent pose optimization. Furthermore, this embodiment also rationally incorporates the matching information of the previous key feature points by setting edge weights. This incorporates the set constraints into the subsequent graph neural network training, improving the interpretability of the graph neural network and the accuracy of keyframe pose optimization.

[0073] S6. Input the graph topology structure of the key frame into the graph neural network for pose optimization, and use the message passing mechanism to iteratively update the key frame pose to obtain the optimized key frame pose.

[0074] The neural network model selected in this embodiment is a graph neural network (GNN), and the GNN is trained to obtain a trained GNN, which is then used to optimize the key frame poses in the constructed key frame graph topology structure. The training process of the GNN is as follows: Figure 2 As shown in the figure, specifically: take the same number of key frames as the preset threshold as the number of nodes for graph neural network training, and extract key feature points for each key frame, use these key feature point descriptors and pixel positions as the input of the node features of the graph neural network, and obtain the trained graph neural network through continuous iterative updates, and output the relative poses between the key frames.

[0075] In addition, the graph topology of the key frames has already determined the edge connectivity between the key frames before they are input into the graph neural network for pose optimization. That is, whether the edge relationship is established is determined by matching the key feature points of the two frames. Therefore, when the graph topology of the key frames is input into the graph neural network for pose optimization, the edge relationship remains unchanged. The purpose is to avoid eliminating erroneous edge structures in the graph neural network design, making the structure of the graph neural network simpler and less complex.

[0076] Specifically, step S6 includes S61-S63:

[0077] S61. Input the graph topology structure of the key frame obtained in step S5 into the graph neural network for pose optimization. First, use the message passing mechanism to calculate the information transmitted from the adjacent nodes for each node in the graph topology structure of the key frame; wherein, the information transmitted from the adjacent nodes is the influence of the neighboring nodes on the current node.

[0078] Specifically, the calculation formula of the information transmitted from the adjacent nodes in step S61 is:

[0079]

[0080] Among them, m i'jRepresents the jth node v j Passed to the i'th node v i' News, f θ represents a message passing function, Indicates the i'th node v i' The k-th hidden state of Represents the jth node v j The k-th hidden state, e i'j Indicates the i'th node v i' With the jth node v j Key frame K i' With K j The edge between i'j Represents edge e i'j The weight of .

[0081] S62. Based on the information transmitted from the adjacent nodes, the aggregation function is used to summarize and obtain the hidden state of each node in the next layer of the graph neural network.

[0082] Specifically, the formula for the hidden state of each node in the next layer of the graph neural network in step S62 is:

[0083]

[0084] in, Indicates the i'th node v i' The k+1th layer hidden state, AGG represents the aggregation function, σ represents the activation function, Indicates the i'th node v i' The set of neighbor nodes.

[0085] S63. According to the hidden state of each node in the next layer of the graph neural network, message passing and aggregation are performed in the next layer of the graph neural network. The iterative update continues until the last layer of the graph neural network to obtain the optimized key frame pose, namely:

[0086] T * =(R * ,t * )

[0087] Among them, T * represents the optimized key frame pose, R * represents the optimized rotation matrix, t * Represents the optimized translation vector.

[0088] In this embodiment, after processing by the graph neural network, the optimized positions of all key frames in the time period are output to obtain optimized key frame poses. These optimized key frame poses are not only more accurate, but also have stronger consistency and robustness. Therefore, using the optimized key frame poses for updating maps and path planning will improve the performance and accuracy of the entire SLAM system.

[0089] S7. After the pose optimization, the key frames in the key frame database are sorted in positive order according to the size of the timestamps. At the same time, the first half of the key frames in the positive order are eliminated, and the second half of the key frames are used for the next round of data collection and optimization, that is, step S1 is re-executed to perform a new round of key feature point extraction of the image stream.

[0090] In the present embodiment, after outputting the optimization result, the graph topology structure currently constructed is destroyed, and memory and resources are released. And in order to balance the continuity of data and the efficient utilization of resources, the number of key frames is halved, and the second half key frames in the key frames of the positive order are retained, so that the smooth progress of the next round of data collection and optimization can be guaranteed. Then return to step S1, restart and extract feature points, record key frames from the image stream, and progressively build a new graph structure, through such circulation, the SLAM system can continuously optimize posture, improve the precision and stability of positioning and mapping. In addition, steps S1-S7 fully illustrate that the present invention is segmented optimization, namely, according to the situation that new image stream is constantly inputted by real-time scene, after being designed to accumulate to certain image frame, the image frame of this period is started to be optimized, its purpose is more flexible, can handle the situation that image stream flow is always inputted, collects image stream while carrying out small-scale posture optimization.

[0091] Specific embodiments are used in the present invention to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core ideas. At the same time, for those skilled in the art, according to the ideas of the present invention, there may be changes in the specific implementation methods and application scopes. In summary, the contents of this specification should not be understood as limiting the present invention.

[0092] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and it should be understood that the scope of protection of the present invention is not limited to such specific descriptions and embodiments. Those skilled in the art can make various other specific variations and combinations based on the technical teachings disclosed in the present invention without departing from the essence of the present invention, and such variations and combinations are still within the scope of protection of the present invention.

Claims

1. A SLAM backend pose optimization method based on graph neural network, characterized in that: The following steps are involved: S1, using feature point detection method to extract key feature points of the image stream continuously input into the SLAM system, and record the feature descriptors of each key feature point; S2. Project the key feature points into three-dimensional space, obtain corresponding points between the left camera image and the right camera image in the continuous image stream through stereo matching, obtain matching points of the key feature points, and triangulate the matching points to obtain the three-dimensional coordinates of the key feature points; S3, set the first frame in the image stream continuously input to the SLAM system as a key frame, and determine whether the average parallax between the current frame and the previous key frame in the image stream continuously input to the SLAM system exceeds a set threshold. If so, use the current frame as the key frame, record the key feature points extracted from the key frame and the feature descriptors of the key feature points, and establish a key frame database. Otherwise, take the next frame and continue to compare it with the previous key frame; S4, determining whether the number of key frames in the key frame database is greater than a preset threshold, if so, executing step S5, otherwise executing step S1; S5. Define a graph structure, and perform key feature point matching based on the key frames, key frame feature points, and feature descriptors of the key frame feature points in the key frame database, determine the co-viewing relationship between the key frames, and obtain a graph topology structure of the key frames; S6. Input the graph topology of the key frame into the graph neural network for pose optimization, and use the message passing mechanism to iteratively update the key frame pose to obtain the optimized key frame pose; S7. After the pose optimization, the key frames in the key frame database are sorted in positive order according to the size of the timestamps. At the same time, the first half of the key frames in the positive order are eliminated, and the second half of the key frames are used for the next round of data collection and optimization, that is, step S1 is re-executed to perform a new round of key feature point extraction of the image stream.

2. The SLAM backend pose optimization method based on graph neural network according to claim 1, wherein Step S1 specifically includes: For each frame of image input to the SLAM system, the ORB feature point detection method is used to detect the corner points and edge features in each frame of image, namely the key feature points, and the position of the key feature points is obtained, namely: p i =(x i ,y i ) Among them, p i Indicates the position of the key feature point indexed by i, x i Indicates the horizontal coordinate of the key feature point with index i, y i Represents the vertical coordinate of the key feature point with index i; At the same time, the feature descriptors of the key feature points are recorded.

3. The SLAM back-end pose optimization method based on graph neural network according to claim 1, wherein Step S2 specifically includes: The key feature points extracted in step S1 are projected into the three-dimensional space. The corresponding points of the left camera image and the right camera image in the continuous image stream are obtained through stereo matching, that is, the matching points of the key feature points. The positions of the matching points are triangulated using the intrinsic parameter matrix of the camera to obtain the three-dimensional coordinates of the key feature points, namely: P i =(X i ,Y i ,Z i ) Among them, P i Represents the three-dimensional coordinates of the key feature point indexed by i, X i Indicates the horizontal coordinate of the key feature point indexed by i in the three-dimensional coordinates, Y i Indicates the vertical coordinate of the key feature point indexed by i in the three-dimensional coordinates, Z i Represents the vertical coordinate of the key feature point with index i in three-dimensional coordinates.

4. The SLAM back-end pose optimization method based on graph neural network according to claim 1, wherein Step S5 specifically includes: S51. Define a graph structure G = (V, E), where V represents a set of poses of key frames, and E represents a set of co-viewing relationships between key frames; S52. According to the defined graph structure, key frames, key frame feature points, and feature descriptors of key frame feature points in the key frame database are obtained, and the key frame feature points of the key frames are matched to obtain a graph topology of the key frames, specifically: If the i'th node v in the graph structure i' With the jth node v j Key frame K i' With K j There is a co-viewing relationship between the two key frames, that is, there are matching key feature points between the two key frames, then the i'th node v is constructed i' With the jth node v j Key frame K i' With K j The edge e between i'j , and at the same time, the edge e i'j The weight w i'j Set as keyframe K i' With K j The normalized value of the number of key feature point matches between them is used to obtain the graph topology of the key frame.

5. The SLAM back-end pose optimization method based on graph neural network according to claim 1, wherein Step S6 specifically includes: S61. Input the graph topology structure of the key frame obtained in step S5 into the graph neural network for pose optimization. First, use the message passing mechanism to calculate the information transmitted from the adjacent nodes for each node in the graph topology structure of the key frame. Among them, the information transmitted from the adjacent nodes is the impact of the neighboring nodes on the current node; S62. Based on the information transmitted from the adjacent nodes, the aggregation function is used to summarize and obtain the hidden state of each node in the next layer of the graph neural network; S63. According to the hidden state of each node in the next layer of the graph neural network, message passing and aggregation are performed in the next layer of the graph neural network. The iterative update continues until the last layer of the graph neural network to obtain the optimized key frame pose, namely: T * =(R * ,t * ) Among them, T * represents the optimized key frame pose, R * represents the optimized rotation matrix, t * Represents the optimized translation vector.

6. The SLAM backend pose optimization method based on graph neural network according to claim 5, wherein The calculation formula of the information transmitted from the adjacent nodes in step S61 is: Among them, m i'j Represents the jth node v j Passed to the i'th node v i' News, f θ represents a message passing function, Indicates the i'th node v i' The k-th hidden state of Represents the jth node v j The k-th hidden state, e i'j Indicates the i'th node v i' With the jth node v j Key frame K i' With K j The edge between i'j Represents edge e i'j The weight of .

7. The SLAM backend pose optimization method based on graph neural network according to claim 6, wherein The formula for the hidden state of each node in the next layer of the graph neural network in step S62 is: in, Indicates the i'th node v i' The k+1th layer hidden state, AGG represents the aggregation function, σ represents the activation function, Indicates the i'th node v i' The set of neighbor nodes.

Images