CN117387639B - Map updating system and method based on laser SALM - Google Patents
Map updating system and method based on laser SALM Download PDFInfo
- Publication number
- CN117387639B CN117387639B CN202311227115.XA CN202311227115A CN117387639B CN 117387639 B CN117387639 B CN 117387639B CN 202311227115 A CN202311227115 A CN 202311227115A CN 117387639 B CN117387639 B CN 117387639B
- Authority
- CN
- China
- Prior art keywords
- constraint
- map
- subgraph
- laser
- nodes
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000001514 detection method Methods 0.000 claims abstract description 25
- 238000012217 deletion Methods 0.000 claims abstract description 8
- 230000037430 deletion Effects 0.000 claims abstract description 8
- 230000000644 propagated effect Effects 0.000 claims abstract description 8
- 239000011159 matrix material Substances 0.000 claims description 14
- 230000009466 transformation Effects 0.000 claims description 11
- 230000001902 propagating effect Effects 0.000 claims description 7
- 238000005457 optimization Methods 0.000 claims description 6
- 230000000452 restraining effect Effects 0.000 claims description 6
- 230000002238 attenuated effect Effects 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 3
- 230000000717 retained effect Effects 0.000 claims 1
- 230000001131 transforming effect Effects 0.000 claims 1
- 238000010586 diagram Methods 0.000 description 5
- 230000007774 longterm Effects 0.000 description 3
- 238000009825 accumulation Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
- G01C21/28—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network with correlation of data from several navigational instruments
- G01C21/30—Map- or contour-matching
- G01C21/32—Structuring or formatting of map data
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/38—Electronic maps specially adapted for navigation; Updating thereof
- G01C21/3804—Creation or updating of map data
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/38—Electronic maps specially adapted for navigation; Updating thereof
- G01C21/3804—Creation or updating of map data
- G01C21/3807—Creation or updating of map data characterised by the type of data
- G01C21/3811—Point data, e.g. Point of Interest [POI]
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/88—Lidar systems specially adapted for specific applications
- G01S17/89—Lidar systems specially adapted for specific applications for mapping or imaging
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Computer Networks & Wireless Communication (AREA)
- Electromagnetism (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention discloses a map updating system and a method thereof based on laser SALM, wherein the system comprises a map building and positioning module, a key frame is generated, a map local map is built, and the position and the pose of a laser odometer are returned; the loop detection module is used for carrying out loop detection by using the generated key frames, acquiring corresponding loop constraint when loop information is detected, and optimizing the global map according to the constraint; the map updating module is used for map updating and constraint reconstruction and comprises a subgraph deletion detection unit, a constraint calculation unit and an updating module. The constraint propagation based on the grid nodes can be directly propagated to the node closest to the node, so that the information loss of constraint propagation is effectively reduced; while the constraint propagates, we can reduce its weight, and when there are multiple constraints on some two nodes, the constraint with high weight is reserved.
Description
Technical Field
The invention relates to the technical field of autonomous mapping and positioning, in particular to a map updating system and method based on laser SALM.
Background
Under the traditional SLAM (synchronous positioning and mapping) framework, SLAM is only used for constructing a global map, after an environment map is constructed, subsequent positioning adopts other modes such as particle filtering, and the positioning cannot be invalid when the surrounding environment is not changed greatly, but under the condition that the environment is changed greatly, the principle of a traditional laser matching map can be invalid, and the positioning is lost.
In the process of long-term graph construction and positioning, SLAM operates in the whole life cycle, positioning can be guaranteed to be carried out on the latest map in real time, in order to prevent unlimited increase of the number of subgraphs and the number of nodes, the map is updated usually with an edge step, and after deleting the nodes, the rest nodes are free from information loss. In the marginalization process, inversion of an information matrix is involved, but the inversion of the matrix is a complex operation, and besides time consumption, interference of factors such as matrix degradation, singular matrix and the like is faced, so that the marginalization process cannot be performed smoothly at all times. Still other methods are to perform positioning in the existing map, when the environment change is detected, perform a mapping program to update the map through the mapping program, which causes deviation between the newly built map and the original old map, and long-term updating can cause error accumulation, and meanwhile, the accumulated error cannot be removed timely, which is a huge disaster to positioning.
Disclosure of Invention
The invention aims to provide a map updating system and a map updating method based on laser SALM, and the method provides a constraint propagation method, so that the loss of information is reduced as much as possible while marginalization is avoided, and the robustness of the system is improved while the calculation force is reduced.
In order to achieve the above purpose, the invention adopts the following technical scheme:
A map updating method based on laser SALM comprises the following steps:
(1) Firstly, constructing a global map by using SLAM, wherein the global map is formed by a plurality of sub-maps, then acquiring laser key frames according to real-time scanning of a laser radar, forming a new sub-map by the plurality of laser key frames, and finally updating the sub-map in the global map;
(2) Adding the subgraph into the global map, simultaneously performing subgraph deletion detection, and if the coverage area of a certain subgraph is completely covered by a new subgraph, performing global map updating;
(3) Setting node grid resolution according to the size of the updated global map to generate a plurality of grid nodes;
(4) Loop detection, namely if the laser key frame is matched with the current track, reserving the current constraint; if the constraint is matched with the old track, the constraint is transmitted to the grid node; if the current constraint is matched with the original track, reserving the current constraint;
Constraint propagation to grid nodes refers to: the key frames and subgraphs in the track are restrained and propagated under grid nodes, and meanwhile, in the restraining and propagation process, restraining weights are attenuated;
(5) Judging the overlapping area, and judging whether the updated subgraph can completely cover the area of the old subgraph; if the overlay is possible, deleting the old subgraph; otherwise, the old subgraph is reserved;
(6) Extracting a Markov blanket, traversing all constraints and nodes, finding and deleting the nodes, deleting the nodes and the subgraphs with direct connection relation of the subgraphs, and adding the constraints into the Markov blanket;
(7) Propagation constraint, namely performing constraint propagation on variables which are not deleted in the Markov blanket;
if the deletion variable is not the current track, deleting directly; if the deleted variable is the current track, the classification processing is performed, specifically as follows:
if the timestamp at the other end of the constraint is smaller than the variable to be deleted, directly deleting the constraint;
if the time stamp at the other end of the constraint is larger than the variable to be deleted and the current subgraph does not have a loop, performing constraint propagation, and propagating to the previous subgraph;
if the time stamp at the other end of the constraint is larger than the variable to be deleted and the current subgraph has a loop, performing constraint propagation and propagating to the grid node;
(8) Deleting nodes, updating constraints, and deleting all the sub-graphs to be deleted and the corresponding constraints and nodes; then adding the new constraint obtained by constraint propagation into a constraint list;
(9) Global optimization, namely optimizing all laser key frames and subgraphs under the current track according to the constraint of the current track;
In the method, in the process of the invention, For the optimized laser key frame set,/>For the optimized sub-graph set, f is the cost function,For the current keyframe set,/>For the current sub-graph set,/>Is a grid node set;
the cost function f is
In the method, in the process of the invention,Constraint given by laser odometer,/> For the constraint of loop detection,/>For the constraint of laser key frames to grid nodes,/>For the constraint of subgraph to grid node,/> In order to restrict inter-sub-graph constraint generated by propagation in the propagation process, x is a laser key frame, s is a grid node, m is a sub-graph, z is a constraint relationship, Ω is an information matrix corresponding to the constraint,/>For homogeneous coordinate transformation
(10) Storing the updated map, and repeating the steps subsequently; and the real-time updating of the map is ensured.
Further, the constraint propagates as:
In the method, in the process of the invention, The pose of the y-th key frame x in the track j under the ith grid node S; /(I)For the coordinate transformation from the nth sub-graph m to the ith grid node S under the locus j-1, the coordinate transformation is obtained through the global pose of the ith grid node S and the nth sub-graph m, namely/> Constraint/>, by coordinate transformationAnd carrying out coordinate change to obtain the final result of constraint propagation.
Further, the weight attenuation in the constraint propagation process is related to the distance change before and after the constraint propagation,
Wherein Ω g is an information matrix of the corresponding constraint, Ω c is an information matrix of the loop detection constraint, where iitii d is a calculated weighted distance, T is a transformation relationship of two coordinate systems, including (x, y, θ), Is the constraint relation between the nth sub-image m under the track j-1 and the y laser key frame x under the track j,/>The coordinates of the nth sub-graph m to the ith grid node S under the track j-1 are transformed.
The invention also provides a map updating system based on the laser SALM, which comprises:
the map building and positioning module generates a key frame, builds a map local map, returns to the laser odometer pose, and calculates and obtains positioning output under the global map according to the global optimization result;
The loop detection module is used for carrying out loop detection by using the generated key frames, acquiring corresponding loop constraint when loop information is detected, and optimizing the global map according to the constraint;
The map updating module is used for map updating and constraint reconstruction, and comprises a sub-graph deleting detection unit for detecting whether a sub-graph needs to be deleted, a constraint calculating unit for recalculating constraint relations among other nodes in the Markov blanket after the sub-graph is deleted, and an updating unit for deleting the nodes and updating the constraints.
Further, the constraint calculating unit is based on constraint propagation of grid nodes, the constraint is propagated to the node closest to the grid nodes, the weight of the constraint is reduced while the constraint is propagated, and when a plurality of constraints exist in two nodes, the constraint with high weight is reserved.
The beneficial effects of the invention are as follows: the invention adds a map updating module after loop detection, thereby realizing long-term map building, positioning and map updating; constraint propagation based on grid nodes is designed, and the constraint is directly propagated to the node closest to the node, so that information loss of constraint propagation is effectively reduced. While the constraint propagates, we can reduce its weight, and when there are multiple constraints on some two nodes, the constraint with high weight is reserved.
Drawings
Fig. 1 is a block diagram of the system of the present invention.
Fig. 2 is a block diagram of a map updating module according to the present invention.
FIG. 3 is a schematic diagram of a grid node of the present invention.
FIG. 4 is a schematic representation of the propagation of constraints of the present invention.
FIG. 5 is a schematic diagram of a delete constraint of the present invention constraint propagation.
FIG. 6 is a schematic diagram of the propagation of the constraints of the present invention to a subgraph.
FIG. 7 is a schematic representation of the propagation of the constraints of the present invention to nodes.
FIG. 8 is a schematic flow chart of the method of the present invention.
Detailed Description
As shown in fig. 1, the map updating system based on the laser SALM provided in this embodiment includes a positioning map building module, a loop detection module, and a map updating module.
The positioning map building module is used for building a map and positioning according to the point cloud data of the laser radar; in the embodiment, a 2D laser radar is adopted to construct a local subgraph of the whole environment; and during positioning, a new map is built according to the laser radar real-time scanning.
The loop detection module is used for carrying out loop detection according to point cloud data acquired by the laser radar and the established map, and optimizing the global map according to loop constraint.
The two modules are common modules for updating and optimizing the existing map.
In this embodiment, a map updating module is added, as shown in fig. 2, and is used for map updating and global optimization, and the map updating module comprises a sub-graph deletion detection unit for detecting whether a sub-graph needs to be deleted, a constraint calculation unit for recalculating constraint relations among other nodes in the markov blanket after the sub-graph is deleted, and an updating unit for deleting the nodes and updating the constraints.
The constraint calculating unit is based on constraint propagation of grid nodes, and can directly propagate the constraint to the node closest to the grid nodes, so that information loss of constraint propagation is effectively reduced. While the constraint propagates, the weight of the constraint is reduced, and when a plurality of constraints exist on certain two nodes, the constraint with high weight is reserved.
Example 2
As shown in fig. 8, the map updating method based on the laser SALM provided in this embodiment includes the following steps:
(1) Firstly, constructing a global map by using SLAM, wherein the global map is formed by a plurality of sub-maps, then acquiring laser key frames according to real-time scanning of a laser radar, forming a new sub-map by the plurality of laser key frames, and finally updating the sub-map in the global map;
(2) Adding the subgraph into the global map, simultaneously performing subgraph deletion detection, and if the coverage area of a certain subgraph is completely covered by a new subgraph, performing global map updating;
(3) Setting node grid resolution according to the size of the updated global map to generate a plurality of grid nodes;
(4) Loop detection, namely if the laser key frame is matched with the current track, reserving the current constraint; if the constraint is matched with the old track, the constraint is transmitted to the grid node; if the current constraint is matched with the original track, reserving the current constraint;
Constraint propagation to grid nodes refers to: the key frames and subgraphs in the track are restrained and propagated under grid nodes, and meanwhile, in the restraining and propagation process, restraining weights are attenuated;
Taking node 3 in track 2, sub-graph 2 in track 1, and grid node 3 as examples, the constraint propagation specifically includes the following:
as shown in fig. 3, the constraint of node 3 in trace 2 and sub-graph 2 in trace 1 is propagated under grid node 3,
In the method, in the process of the invention,The pose of the node 3 in the lower track 2 of the grid node 3 is shown; /(I)For the coordinate transformation of sub-graph 2 to grid node 3 under track 1, this transformation is obtained by the global pose of s 3 and m 1,2, i.e./>Constraint/>, by the transformation pairAnd carrying out coordinate change to obtain a final result.
The weight attenuation in the constraint propagation process is related to the distance change before and after the constraint propagation,
Wherein the expression of II T d is a calculation weighted distance, T represents the transformation relation of two coordinate systems, and comprises (x, y, θ), specifically:
(5) Judging the overlapping area, and judging whether the updated subgraph can completely cover the area of the old subgraph; if the overlay is possible, deleting the old subgraph; otherwise, the old subgraph is reserved;
(6) Extracting a Markov blanket, traversing all constraints and nodes, finding and deleting the nodes, deleting the nodes and the subgraphs with direct connection relation of the subgraphs, and adding the constraints into the Markov blanket;
(7) Propagation constraint, namely performing constraint propagation on variables which are not deleted in the Markov blanket;
If the deletion variable is not the current track, deleting directly; if the deleted variable is the current track, classifying the deleted variable, wherein the specific processing is as follows:
If the other end time stamp of the constraint is smaller than the variable to be deleted, directly deleting the constraint, as shown in fig. 4;
If the time stamp at the other end of the constraint is larger than the variable to be deleted and the current subgraph does not have a loop, performing constraint propagation, and propagating to the previous subgraph, as shown in fig. 5; the sub-graph without the last sub-graph is not transmitted, and when the transmission is restrained, Meanwhile, the weight needs to be attenuated:
if the time stamp at the other end of the constraint is larger than the variable to be deleted and the current subgraph has a loop, performing constraint propagation and propagating to the grid node;
(8) Deleting nodes, updating constraints, and deleting all the sub-graphs to be deleted and the corresponding constraints and nodes; then adding the new constraint obtained by constraint propagation into a constraint list;
(9) Global optimization, namely optimizing all laser key frames and subgraphs under the current track according to the constraint of the current track;
In the method, in the process of the invention, For optimized keyframe set,/>For the optimized sub-graph set, f is the cost function,/>For the current keyframe set,/>For the current sub-graph set,/>Is a grid node set;
the cost function f is
In the method, in the process of the invention,Constraint given by laser odometer,/> For the constraint of loop detection,/>As a node-to-grid node constraint,For the constraint of subgraph to grid node,/> For inter-subgraph constraints generated by propagation in the constraint propagation process, z is a constraint relation, x is a key frame, s is a grid node, m is a subgraph, and Ω is an information matrix corresponding to the constraints;
(10) Storing the updated map, and repeating the steps subsequently; and the real-time updating of the map is ensured. Therefore, the map used every day is ensured to be updated in real time, and positioning loss and map building error accumulation can not occur.
The foregoing is merely a preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any modification and substitution based on the technical scheme and the inventive concept provided by the present invention should be covered in the scope of the present invention.
Claims (5)
1. The map updating method based on the laser SALM is characterized by comprising the following steps of:
(1) Firstly, constructing a global map by using SLAM, wherein the global map is formed by a plurality of sub-maps, then acquiring laser key frames according to real-time scanning of a laser radar, forming a new sub-map by the plurality of laser key frames, and finally updating the sub-map in the global map;
(2) Adding the subgraph into the global map, simultaneously performing subgraph deletion detection, and if the coverage area of a certain subgraph is completely covered by a new subgraph, performing global map updating;
(3) Setting node grid resolution according to the size of the updated global map to generate a plurality of grid nodes;
(4) Loop detection, namely if the laser key frame is matched with the current track, reserving the current constraint; if the constraint is matched with the old track, the constraint is transmitted to the grid node; if the current constraint is matched with the original track, reserving the current constraint;
Constraint propagation to grid nodes refers to: the key frames and subgraphs in the track are restrained and propagated under grid nodes, and meanwhile, in the restraining and propagation process, restraining weights are attenuated;
(5) Judging the overlapping area, and judging whether the updated subgraph can completely cover the area of the old subgraph; if the overlay is possible, deleting the old subgraph; otherwise, the old subgraph is reserved;
(6) Extracting a Markov blanket, traversing all constraints and nodes, finding and deleting the nodes, deleting the nodes and the subgraphs with direct connection relation of the subgraphs, and adding the constraints into the Markov blanket;
(7) Propagation constraint, namely performing constraint propagation on variables which are not deleted in the Markov blanket;
if the deletion variable is not the current track, deleting directly; if the deleted variable is the current track, the classification processing is performed, specifically as follows:
if the timestamp at the other end of the constraint is smaller than the variable to be deleted, directly deleting the constraint;
if the time stamp at the other end of the constraint is larger than the variable to be deleted and the current subgraph does not have a loop, performing constraint propagation, and propagating to the previous subgraph;
if the time stamp at the other end of the constraint is larger than the variable to be deleted and the current subgraph has a loop, performing constraint propagation and propagating to the grid node;
(8) Deleting nodes, updating constraints, and deleting all the sub-graphs to be deleted and the corresponding constraints and nodes; then adding the new constraint obtained by constraint propagation into a constraint list;
(9) Global optimization, namely optimizing all laser key frames and subgraphs under the current track according to the constraint of the current track;
In the method, in the process of the invention, For the optimized laser key frame set,/>For the optimized sub-graph set, f is the cost function,/>For the current keyframe set,/>For the current sub-graph set,/>Is a grid node set;
the cost function f is
In the method, in the process of the invention,Constraint given by laser odometer,/> For the constraint of loop detection,/>For the constraint of laser key frames to grid nodes,/>For the constraint of subgraph to grid node,/> To constrain inter-subgraph constraints generated by propagation in the propagation process, x i is the current keyframe set/>X j is the current keyframe set/>In j-th laser key frame,/>Is the constraint relation of laser key frames,/>An information matrix which is correspondingly constrained for the laser key frame; m j is the current sub-graph set/>In (j) th subgraph,/>Is the constraint relation between the laser key frame and the subgraph,/>An information matrix which is correspondingly constrained by the laser key frame and the subgraph; s j is the grid node set/>In j-th grid node,/>Is the constraint relation of grid nodes,/>An information matrix corresponding to the constraint for the grid node; m i Current subgraph set/>I-th subgraph,/>Is a constraint relation of subgraph,/>Information matrix of sub-graph corresponding constraint,/>Transforming homogeneous coordinates;
(10) Storing the updated map, and repeating the steps subsequently; and the real-time updating of the map is ensured.
2. The map updating method based on laser SALM according to claim 1, wherein the constraint propagation is:
In the method, in the process of the invention, The pose of the y laser key frame x in the track j under the ith grid node S; /(I)For the coordinate transformation from the nth sub-graph m to the ith grid node S under the track j-1, the coordinate transformation is obtained through the global pose of the ith grid node S and the nth sub-graph m; /(I)The constraint relation between the nth sub-image m under the track j-1 and the y laser key frame x under the track j is adopted; constraint/>, by coordinate transformationAnd carrying out coordinate change to obtain the final result of constraint propagation.
3. The map updating method based on laser SALM as claimed in claim 2, wherein the magnitude of the weight attenuation in the constraint propagation process is related to the distance change before and after the constraint propagation,
Wherein Ω g is an information matrix of the corresponding constraint, Ω c is an information matrix of the loop detection constraint, where T d is a calculated weighted distance, T is a transformation relationship of two coordinate systems, including (x, y, θ), W 1 and w 2 are weights; /(I)Is the constraint relation between the nth sub-image m under the track j-1 and the y laser key frame x under the track j,The coordinates of the nth sub-graph m to the ith grid node S under the track j-1 are transformed.
4. A system for implementing the laser SALM-based map updating method of any one of claims 1 to 3, comprising:
the map building and positioning module generates a key frame, builds a map local map, returns to the laser odometer pose, and calculates and obtains positioning output under the global map according to the global optimization result;
The loop detection module is used for carrying out loop detection by using the generated key frames, acquiring corresponding loop constraint when loop information is detected, and optimizing the global map according to the constraint;
The map updating module is used for map updating and constraint reconstruction, and comprises a sub-graph deleting detection unit for detecting whether a sub-graph needs to be deleted, a constraint calculating unit for recalculating constraint relations among other nodes in the Markov blanket after the sub-graph is deleted, and an updating unit for deleting the nodes and updating the constraints.
5. The map updating system based on laser SALM as claimed in claim 4, wherein said constraint calculating means is based on constraint propagation of grid nodes, and propagates the constraint directly to nodes closest to the grid nodes, and reduces the weight of the nodes while propagating the constraint, and when there are a plurality of constraints in any two nodes, the constraint with high weight is retained.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311227115.XA CN117387639B (en) | 2023-09-22 | 2023-09-22 | Map updating system and method based on laser SALM |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311227115.XA CN117387639B (en) | 2023-09-22 | 2023-09-22 | Map updating system and method based on laser SALM |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117387639A CN117387639A (en) | 2024-01-12 |
CN117387639B true CN117387639B (en) | 2024-06-14 |
Family
ID=89471073
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311227115.XA Active CN117387639B (en) | 2023-09-22 | 2023-09-22 | Map updating system and method based on laser SALM |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117387639B (en) |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112100298A (en) * | 2020-08-17 | 2020-12-18 | 深圳市优必选科技股份有限公司 | Drawing establishing method and device, computer readable storage medium and robot |
CN113670290A (en) * | 2021-06-30 | 2021-11-19 | 西南科技大学 | Mobile robot indoor map construction method based on multi-robot cooperation |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113624221B (en) * | 2021-06-30 | 2023-11-28 | 同济人工智能研究院(苏州)有限公司 | 2.5D map construction method integrating vision and laser |
CN115046540A (en) * | 2022-05-25 | 2022-09-13 | 新驱动重庆智能汽车有限公司 | Point cloud map construction method, system, equipment and storage medium |
CN115240047A (en) * | 2022-08-04 | 2022-10-25 | 中国矿业大学(北京) | Laser SLAM method and system fusing visual loopback detection |
CN115407352A (en) * | 2022-08-19 | 2022-11-29 | 杭州电子科技大学 | Robot positioning and mapping system and method based on particle filtering and mapping optimization |
-
2023
- 2023-09-22 CN CN202311227115.XA patent/CN117387639B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112100298A (en) * | 2020-08-17 | 2020-12-18 | 深圳市优必选科技股份有限公司 | Drawing establishing method and device, computer readable storage medium and robot |
CN113670290A (en) * | 2021-06-30 | 2021-11-19 | 西南科技大学 | Mobile robot indoor map construction method based on multi-robot cooperation |
Also Published As
Publication number | Publication date |
---|---|
CN117387639A (en) | 2024-01-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Ekanayake et al. | Computer vision-based interior construction progress monitoring: A literature review and future research directions | |
JP7166383B2 (en) | Method and apparatus for creating high-precision maps | |
CN113593017B (en) | Method, device, equipment and storage medium for constructing surface three-dimensional model of strip mine | |
CN113516664B (en) | Visual SLAM method based on semantic segmentation dynamic points | |
CN107515891A (en) | A kind of robot cartography method, apparatus and storage medium | |
CN112668696A (en) | Unmanned aerial vehicle power grid inspection method and system based on embedded deep learning | |
CN110260856A (en) | One kind building drawing method and device | |
CN113780270B (en) | Target detection method and device | |
WO2023206904A1 (en) | Pedestrian trajectory tracking method and system, and related apparatus | |
CN112917467B (en) | Robot positioning and map building method and device and terminal equipment | |
CN109657403A (en) | A kind of three-dimensional live bridge modeling optimization method based on unmanned plane oblique photograph | |
CN111091101A (en) | High-precision pedestrian detection method, system and device based on one-step method | |
Jiang et al. | Optimization of intelligent plant cultivation robot system in object detection | |
CN116166642A (en) | Spatio-temporal data filling method, system, equipment and medium based on guide information | |
CN115077540A (en) | Map construction method and device | |
CN117387639B (en) | Map updating system and method based on laser SALM | |
Zeng et al. | Low‐Shot Wall Defect Detection for Autonomous Decoration Robots Using Deep Reinforcement Learning | |
CN114115231B (en) | Space pose point cloud correction method and system for mobile robot | |
CN112926681B (en) | Target detection method and device based on deep convolutional neural network | |
De Geyter et al. | Automated training data creation for semantic segmentation of 3D point clouds | |
Liu et al. | Abnormal behavior analysis strategy of bus drivers based on deep learning | |
CN112069997B (en) | Unmanned aerial vehicle autonomous landing target extraction method and device based on DenseHR-Net | |
Dutta et al. | In situ climate modeling for analyzing extreme weather events | |
CN114964204A (en) | Map construction method, map using method, map constructing device, map using equipment and storage medium | |
CN116295507B (en) | Laser inertial odometer optimization method and system based on deep learning |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |