CN108520543B - Method, equipment and storage medium for optimizing relative precision map - Google Patents
Method, equipment and storage medium for optimizing relative precision map Download PDFInfo
- Publication number
- CN108520543B CN108520543B CN201810312477.1A CN201810312477A CN108520543B CN 108520543 B CN108520543 B CN 108520543B CN 201810312477 A CN201810312477 A CN 201810312477A CN 108520543 B CN108520543 B CN 108520543B
- Authority
- CN
- China
- Prior art keywords
- pose
- graph model
- optimization
- key landmark
- pose graph
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/20—Drawing from basic elements, e.g. lines or circles
- G06T11/206—Drawing of charts or graphs
-
- 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
-
- 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
- G01S19/00—Satellite radio beacon positioning systems; Determining position, velocity or attitude using signals transmitted by such systems
- G01S19/38—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system
- G01S19/39—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system the satellite radio beacon positioning system transmitting time-stamped messages, e.g. GPS [Global Positioning System], GLONASS [Global Orbiting Navigation Satellite System] or GALILEO
- G01S19/42—Determining position
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Theoretical Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Image Analysis (AREA)
- Processing Or Creating Images (AREA)
Abstract
The invention provides a method, equipment and computer storage medium for optimizing a relative precision map, wherein the method comprises the following steps: measuring the absolute accuracy of at least one key landmark in a specific environment area to obtain the pose of the at least one key landmark; constructing a relative precision map of the specific environment area, and establishing a pose graph model of the relative precision map, wherein the starting point and the end point of the pose graph model are not coincident; optimizing the pose graph model using the pose of the at least one key landmark.
Description
Technical Field
The present invention relates to a graph model optimization technique, and in particular, to a method, an apparatus, and a computer storage medium for optimizing a relative accuracy map.
Background
In the related art, there are various ways for constructing maps, such as maps established by Simultaneous Mapping And positioning (SLAM)/Motion computation Structure (SFM), And absolute accuracy maps obtained by absolute measurement techniques such as global satellite positioning And Mapping. Among them, since it is difficult to determine the absolute accuracy of a map created by SLAM/SFM without an input of external measurement such as Global Positioning System (GPS), it is called a relative accuracy map.
Fig. 1 is a diagram showing comparison effects before and after optimization of a graph model with a closed loop in the related art. The pose graph model optimization scheme with the closed loop in the related art can be optimized only when the robot returns to the place where the robot passes before (namely, the pose graph model has the closed loop). Specifically, as shown in fig. 1, before optimization, a robot makes a circular motion with a radius of 1m from a starting point a along a circle center O, and the robot goes through a series of poses a1- > a2- > A3 and so on, and after one circle, the robot should return to a, but the control input itself of the robot is noisy, so that it goes to B (if there is no noise, B should coincide with a); the graph model is optimized through the correlation technique, although the optimized track is approximate to an ellipse, edges formed by adjacent poses in the pose graph model are slightly adjusted, and A and B are overlapped.
FIG. 2 is a diagram illustrating a map model of relative accuracy without a closed loop in the related art; fig. 3 is a schematic diagram illustrating fusion of a plurality of sets of maps with relative accuracy in the environment of an intersection in the related art.
In combination with the map situations shown in fig. 2 and fig. 3, the existing technical solution for performing pose graph model optimization on a relative accuracy map shown in fig. 1 has the following defects:
first, as shown in fig. 2, there is no way to optimize the existing graph model in the absence of closed-loop map acquisition data.
Secondly, as shown in fig. 3, even if closed-loop map acquisition data exists, if multiple batches of maps intersect at the same intersection, it is difficult to determine the confidence of each batch of maps; to put it back, even if the confidence level can be determined well, the map at the intersection is still tangible, and great challenges are created for subsequent map merging and updating.
Disclosure of Invention
In order to overcome the defects of the prior technical scheme for optimizing a pose graph model of a relative precision map as shown in fig. 1, the invention provides a method, equipment and a computer storage medium for optimizing the relative precision map.
According to a first aspect of the present invention, there is provided a method of optimizing a relative accuracy map, comprising: measuring the absolute accuracy of at least one key landmark in a specific environment area to obtain the pose of the at least one key landmark; constructing a relative precision map of the specific environment area, and establishing a pose graph model of the relative precision map, wherein the starting point and the end point of the pose graph model are not coincident; optimizing the pose graph model using the pose of the at least one key landmark.
According to an embodiment of the invention, optimizing the pose graph model using the pose of the at least one key landmark comprises: detecting whether the intermediate nodes of the pose graph model form a closed loop or not; and if the intermediate node of the pose graph model is detected to form a closed loop, pre-optimizing the pose graph model of the relative precision map based on the intermediate node forming the closed loop.
According to an embodiment of the present invention, optimizing the pose graph model using the pose of the at least one key landmark comprises at least the following operations: a first operation of searching the pose graph model for the pose of a node corresponding to the pose of the at least one key landmark using a particular matching algorithm; a second operation of optimizing the pose graph model based on a transformation relationship between the pose of the at least one key landmark and the pose of the corresponding node.
According to an embodiment of the invention, wherein the specific matching algorithm comprises Iterative Closest Point (ICP) or projective transformation n points (PnP).
According to an embodiment of the present invention, optimizing the pose graph model based on the transformation relationship between the pose of the at least one key landmark and the pose of the corresponding node comprises: establishing a transformation relation between the pose of the at least one key landmark and the pose of the corresponding node, and taking the established transformation relation between the pose of the at least one key landmark and the pose of the corresponding node as a new edge; constructing a transformation relation between adjacent nodes in the pose graph model, and taking the transformation relation between the adjacent nodes as a primary side; and optimizing the pose graph model formed by combining the new edge and the primary edge through a specific graph model optimization algorithm.
According to an embodiment of the invention, wherein the specific graph model optimization algorithm comprises a least squares method or a Levenberg algorithm (LM).
According to an embodiment of the present invention, the optimizing the pose graph model using the pose of the at least one key landmark further comprises: after a second operation optimizes the pose graph model, increasing a resolution of the pose of the at least one key landmark and the pose of the corresponding node; and repeating the first operation and the second operation to perform iterative optimization on the optimized pose graph model by using the pose of the at least one key landmark.
According to an embodiment of the invention, the resolution of the pose of the at least one key landmark and the pose of the corresponding node is increased by a factor of two.
According to an embodiment of the invention, wherein the first operation further comprises: determining a search threshold; detecting whether the search threshold is greater than a predetermined threshold; if the searching threshold is detected to be larger than the preset threshold, searching the pose of the node corresponding to the pose of the at least one key landmark in the pose graph model according to the searching range corresponding to the searching threshold.
According to an embodiment of the invention, wherein the method further comprises: determining the optimization amplitude of the previous round in the iterative optimization process; dynamically adjusting the search threshold according to the magnitude of the optimized magnitude.
According to an embodiment of the present invention, wherein dynamically adjusting the search threshold according to the magnitude of the optimization magnitude comprises: if the optimization amplitude of the previous round is larger than the specific amplitude threshold value during the previous round of optimization, reducing the search threshold value during the next round of optimization by using the first adjustment amplitude; if the optimization amplitude of the previous round is not larger than the specific amplitude threshold value in the previous round of optimization, reducing or not adjusting the search threshold value in the next round of optimization by a second adjustment amplitude; and the value of the first adjustment amplitude is larger than that of the second adjustment amplitude.
According to an embodiment of the invention, wherein the method further comprises: and when the search threshold is smaller than a preset threshold, stopping iterative optimization and outputting the optimized pose graph model.
According to an embodiment of the present invention, while the first operation and the second operation are repeated to perform iterative optimization on the optimized pose graph model by using the pose of the at least one key landmark, the number of iterations is counted.
According to an embodiment of the invention, wherein the method further comprises: and when the counted iteration times reach an iteration time threshold value, stopping iterative optimization and outputting an optimized pose graph model.
According to a second aspect of the present invention, there is provided an apparatus for optimizing a relative accuracy map, the apparatus comprising: the absolute precision measuring device is used for measuring the absolute precision of at least one key landmark in a specific environment area to obtain the pose of the at least one key landmark; the relative precision model building device is used for building a relative precision map of the specific environment area and building a pose graph model of the relative precision map, and the starting point and the end point of the pose graph model are not coincident; and the pose graph model optimizing device is used for optimizing the pose graph model by using the pose of the at least one key landmark.
According to an embodiment of the invention, the pose graph model optimization device is further configured to detect whether an intermediate node of the pose graph model forms a closed loop; and if the intermediate node of the pose graph model is detected to form a closed loop, pre-optimizing the pose graph model of the relative precision map based on the intermediate node forming the closed loop.
According to an embodiment of the present invention, the pose graph model optimization device at least includes the following devices: first means for searching the pose graph model for the pose of a node corresponding to the pose of the at least one key landmark using a particular matching algorithm; second means for optimizing the pose graph model based on a transformation relationship between the pose of the at least one key landmark and the pose of the corresponding node.
According to an embodiment of the invention, wherein the specific matching algorithm comprises ICP or PnP.
According to an embodiment of the present invention, the second apparatus is further configured to establish a transformation relationship between the pose of the at least one key landmark and the pose of the corresponding node, and use the established transformation relationship between the pose of the at least one key landmark and the pose of the corresponding node as a new edge; constructing a transformation relation between adjacent nodes in the pose graph model, and taking the transformation relation between the adjacent nodes as a primary side; and optimizing a pose graph model formed by combining the new edge and the primary edge through a specific graph model optimization algorithm.
According to an embodiment of the invention, wherein the specific graph model optimization algorithm comprises least squares or LM.
According to an embodiment of the present invention, the pose graph model optimization apparatus further includes: resolution increasing means for increasing the resolution of the pose of the at least one key landmark and the pose of the corresponding node after the pose graph model is optimized by the second means; and the repeating device is used for repeating the first operation and the second operation so as to perform iterative optimization on the optimized pose graph model by using the pose of the at least one key landmark.
According to an embodiment of the invention, the resolution increasing means is further configured to increase the resolution of the pose of the at least one key landmark and the pose of the corresponding node by a double resolution.
According to an embodiment of the present invention, the first apparatus is further configured to determine a search threshold; detecting whether the search threshold is greater than a predetermined threshold; if the searching threshold is detected to be larger than the preset threshold, searching the pose of the node corresponding to the pose of the at least one key landmark in the pose graph model according to the searching range corresponding to the searching threshold.
According to an embodiment of the present invention, the first means is further configured to determine an optimization magnitude of a previous round in the iterative optimization process; dynamically adjusting the search threshold according to the magnitude of the optimized magnitude.
According to an embodiment of the present invention, the first means is further configured to decrease the search threshold in the next round of optimization by a first adjustment range if the optimization range in the previous round is greater than the specific range threshold in the previous round of optimization; if the optimization amplitude of the previous round is not larger than the specific amplitude threshold value in the previous round of optimization, reducing or not adjusting the search threshold value in the next round of optimization by a second adjustment amplitude; and the value of the first adjustment amplitude is larger than that of the second adjustment amplitude.
According to an embodiment of the present invention, the method further includes: and the output device is used for stopping iterative optimization and outputting the optimized pose graph model when the search threshold is detected to be smaller than a preset threshold.
According to an embodiment of the present invention, the method further includes: and the counting device is used for counting the iteration times while repeatedly performing the first operation and the second operation so as to perform iterative optimization on the optimized pose graph model by using the at least one key landmark.
According to an embodiment of the present invention, the method further includes: and the output device is used for stopping iterative optimization and outputting the optimized pose graph model when the counted iteration times reach the iteration time threshold.
According to a third aspect of the present invention, there is provided an apparatus for optimizing a relative accuracy map, comprising: one or more processors; a memory; a program stored in the memory, which when executed by the one or more processors, causes the processors to perform the method as described above.
According to a fourth aspect of the present invention, there is provided a computer readable storage medium storing a program which, when executed by a processor, causes the processor to perform the method as described above.
Compared with the traditional pose graph model optimization shown in fig. 1, the method for optimizing the relative precision map provided by the invention can optimize map acquisition data without closed loops.
Moreover, even if a plurality of batches of maps intersect at the same intersection in closed-loop map acquisition data and the confidence of each batch of maps cannot be determined, the invention obtains the pose of a key landmark (such as a real intersection) by an absolute precision measurement technology, thereby solving the problem of fusion deformation among the plurality of batches of maps.
In addition, aiming at the problem that the nodes of the map cannot search the key landmarks of the absolute accuracy map due to the large error of the relative accuracy map, the invention can respectively obtain edges with different information contents to carry out iterative pose map model optimization by using a coarse-to-fine matching mode, and finally the relative accuracy map is continuously close to the absolute accuracy map.
It is to be understood that the teachings of the present invention need not achieve all of the above-described benefits, but rather that specific embodiments may achieve specific technical results, and that other embodiments of the present invention may achieve benefits not mentioned above.
Drawings
The above and other objects, features and advantages of exemplary embodiments of the present invention will become readily apparent from the following detailed description read in conjunction with the accompanying drawings. Several embodiments of the invention are illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings and in which:
in the drawings, the same or corresponding reference numerals indicate the same or corresponding parts.
FIG. 1 is a diagram showing comparison effects before and after optimization of a graph model with a closed loop in the related art;
fig. 2 is a diagram showing a map model of relative accuracy in which a closed loop does not exist in the related art;
FIG. 3 is a schematic illustration of a plurality of sets of relative accuracy maps merged in the environment of an intersection in the related art;
FIG. 4 illustrates a flow diagram of a method of optimizing a relative accuracy map in accordance with an embodiment of the present invention;
FIG. 5 is a schematic diagram showing a three-dimensional coordinate system in the related art;
FIG. 6 is a state change display effect diagram illustrating attitude model optimization in an application example of the present invention;
FIG. 7 is a schematic diagram illustrating the components of an apparatus for optimizing a relative accuracy map according to an embodiment of the present invention;
FIG. 8 shows a schematic diagram of an apparatus for optimizing relative accuracy maps according to an embodiment of the present invention;
FIG. 9 illustrates a schematic diagram of a computer-readable storage medium for optimizing relative accuracy maps, according to an embodiment of the present invention.
Detailed Description
The principles and spirit of the present invention will be described with reference to a number of exemplary embodiments. It is understood that these embodiments are given solely for the purpose of enabling those skilled in the art to better understand and to practice the invention, and are not intended to limit the scope of the invention in any way. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
The following detailed description of embodiments of the invention refers to the accompanying drawings.
Before further detailed description of the present invention, names and terms related to embodiments of the present invention are described, and terms or terms designed in embodiments of the present invention are applicable to the following explanations.
1) Simultaneous mapping and localization (SLAM), placing a robot with sensors in an unknown environment, letting the robot attempt to incrementally build a continuous map and use the map for localization.
2) Motion computation Structures (SFM), a photogrammetric range imaging technique, estimates three-dimensional structures in a sequence of two-dimensional images, possibly combined with local motion signals.
3) An absolute precision map (absolute precision map) is a map obtained by absolute measurement technologies such as global positioning and mapping, and is called an absolute precision map.
4) Landmark (landmark), in particular an object of absolute position measured by absolute precision measurement techniques such as GPS, mapping, etc., should have a distinctive shape and characteristics, such as a geometry that can be easily distinguished and detected by means of relevant instruments on the vehicle, both natural and artificial. Some landmarks also contain additional information (e.g., bar codes, two-dimensional codes, etc.).
5) Pose (position), position and attitude (orientation), for example: in two dimensions (x, y, yaw), in three dimensions (x, y, z, yaw, pitch, roll), x, y, z respectively represent position coordinates, and ro l, yaw, pitch respectively represent rotation angles of the object around x, y, z axes, i.e. postures.
6) Pose graph model optimization (PGO), which builds a graph model for a series of poses, usually using least squares, LM, etc. algorithms for optimization.
7) Closed loop (loop closure), go back to where it was once walked.
8) Nodes (nodes), cells in the graph model, refer to poses in the PGO.
9) Edge (edge), a cell in the graph model, indicates the transformation relationship between two poses in the PGO.
10) Iterative Closest Point (ICP), an algorithm that iteratively gradually optimizes two sets of point clouds together. The ICP algorithm herein includes, but is not limited to, point-to-point distance (ICP), point-to-plane distance (ICP), point-to-Lmne distance (ICP), plane-to-plane distance (GICP), combined spatial distance and angular distance (normal ICP), and the like.
11) Projective transformation is carried out on n points (PnP), a set of 3D point clouds is given, and the poses of the calibrated cameras and the 2D pixel coordinates of the 3D point clouds in the image are estimated.
12) The shaft Angle (Axis-Angle), the Angle of rotation, represents the Angle at which the rotation is parameterized by two values, one Axis or line, and one that describes the amount of rotation about this Axis. It is also called the exponential coordinate of the rotation. For three-dimensional rotation, its rotation angle can be represented using norm (a) of three-dimensional vector a ═ (ax, ay, az), normalized three-dimensional vector
(ax/norm (A), ay/norm (A), az/norm (A)) represents a rotation axis. The counterclockwise rotation is defined herein as a positive angle rotation.
FIG. 4 illustrates a flow diagram of a method of optimizing a relative accuracy map in accordance with an embodiment of the present invention. As shown in fig. 4, the method includes: operation 401, performing absolute accuracy measurement on at least one key landmark in a specific environment area to obtain a pose of the at least one key landmark; operation 402, constructing a relative accuracy map of the specific environmental area, and establishing a pose graph model of the relative accuracy map, where a start point and an end point of the pose graph model do not coincide; at operation 403, the pose graph model is optimized with the pose of the at least one key landmark.
At operation 401, at least one key landmark L in a particular environmental region may be measured by an absolute accuracy measurement technique such as global positioning satellite, mapping, and the like m Measurements are made to obtain at least one key landmark L m I represents the number of key landmarks. It will be appreciated by those skilled in the art that in point cloud matching algorithms, the measured poses are typically represented by a three-dimensional coordinate system, as shown in fig. 5, a key landmark L m Can be expressed as (x, y, z, yaw, pitch, roll) by a three-dimensional rotating coordinate system.
At operation 402, a relative accuracy map of the specific environmental region may be constructed by SLAM/SFM, further establishing a pose graph model of the relative accuracy map. Of course, compared with the pose graph model in the prior art, similar to the comparison between fig. 2 and fig. 1 mentioned above, it is obvious that the starting point and the end point of the pose graph model aimed at by the present invention are not coincident from the view point of the graph model as a whole, that is, the pose graph model established as a whole is a non-closed loop model. The whole pose graph model is a non-closed loop model, but a certain segment or a plurality of segments may still exist in the middle segment of the pose graph model and are closed loop models.
Before performing operation 403 to optimize the pose graph model with the poses of the at least one key landmark, a pre-optimization may be performed for the case where a closed loop exists in the pose graph model established in operation 402.
According to an embodiment of the invention, optimizing the pose graph model using the pose of the at least one key landmark comprises: detecting whether the intermediate nodes of the pose graph model form a closed loop or not; and if the intermediate node of the pose graph model is detected to form a closed loop, pre-optimizing the pose graph model of the relative precision map based on the intermediate node forming the closed loop. It will be understood by those skilled in the art that the process of pre-optimizing the pose graph model of the relative accuracy map is similar to the above-mentioned optimization process of fig. 1, and therefore, the detailed description is not repeated here.
After pre-optimizing the built pose graph model, operation 403 optimizes the pose graph model with the poses of the at least one key landmark comprises at least the following: a first operation of searching the pose graph model for the pose of a node corresponding to the pose of the at least one key landmark using a particular matching algorithm; a second operation of optimizing the pose graph model based on a transformation relationship between the pose of the at least one key landmark and the pose of the corresponding node.
In a first operation, ICP or PnP may be utilized to find and associate a key landmarks L m Corresponding b nodes N n Where n represents a node number. Taking the ICP matching algorithm as an example, the following calculation formula 1 may be adopted to perform pose matching of the point cloud, so as to search for the pose of the node corresponding to the pose of the at least one key landmark in the bit pose graph model.
Wherein R is a three-dimensional rotation matrix, T is a three-dimensional translation matrix, p i And q is i For the corresponding points (e.g., matching by a certain threshold euclidean distance), | | | is a euclidean distance calculation symbol, and the pose (position and attitude, here denoted by R, T) of the point cloud to be aligned is optimized by minimizing the euclidean distances of all the corresponding points.
At a second operation, optimizing the pose graph model based on a transformed relationship between the pose of the at least one key landmark and the pose of the corresponding node comprises: establishing a transformation relation between the pose of the at least one key landmark and the pose of the corresponding node, and taking the established transformation relation between the pose of the at least one key landmark and the pose of the corresponding node as a new edge; constructing a transformation relation between adjacent nodes in the pose graph model, and taking the transformation relation between the adjacent nodes as a primary side; and optimizing a pose graph model formed by combining the new edge and the primary edge through a specific graph model optimization algorithm. Wherein the particular graph model optimization algorithm comprises a least squares method or a Levenberg algorithm (LM).
Through the explanation of the specific implementation of the operations 401 to 403, the primary optimization of the pose graph model of the relative precision map is completed; in order to further achieve an improvement in the relative accuracy of the relative accuracy map so that the relative accuracy thereof is continuously close to the absolute accuracy of the absolute accuracy map, a repeated iterative process of the pose map model on which the optimization has been performed once will be described in detail next. For example, in the state change display effect graph optimized by the pose graph model shown in fig. 6, the key landmark and the node close to the key landmark are matched through an ICP algorithm to obtain a conversion relationship between the two, that is, to establish an edge, in order to more easily realize matching between the key landmark and the node, the resolution of (1) - (2) in fig. 6 is used in the first round of graph model process, and the search threshold is a numerical value with the resolution as the minimum unit, so when the resolution is smaller, the physical search range is larger, but the information amount contained in the edge obtained by matching is smaller, and therefore, the state of (1) in fig. 6 is optimized to the state of (2); in order to further enhance the optimization and increase the information content included in the edge obtained by matching, it is necessary to further increase the resolution of the pose of the at least one key landmark and the pose of the corresponding node, so as to enter the iterative optimization process of (2) - (3) in fig. 6.
According to an embodiment of the present invention, optimizing the pose graph model using the pose of the at least one key landmark further comprises: after the pose graph model is optimized by the second operation, increasing the resolution of the poses of the at least one key landmark and the corresponding node; and repeating the first operation and the second operation to perform iterative optimization on the optimized pose graph model by using the pose of the at least one key landmark.
Wherein the resolution of the pose of the at least one key landmark and the pose of the corresponding node may be increased by a doubling of the resolution.
According to an embodiment of the invention, the first operations further comprise: determining a search threshold; detecting whether the search threshold is greater than a predetermined threshold; if the searching threshold is detected to be larger than the preset threshold, searching the pose of the node corresponding to the pose of the at least one key landmark in the pose graph model according to the searching range corresponding to the searching threshold.
Further, the method further comprises: determining the optimization amplitude of the previous round in the iterative optimization process; dynamically adjusting the search threshold according to the magnitude of the optimized magnitude. Specifically, dynamically adjusting the search threshold according to the magnitude of the optimized magnitude includes: if the optimization amplitude of the previous round is larger than the specific amplitude threshold value during the previous round of optimization, reducing the search threshold value during the next round of optimization by using the first adjustment amplitude; if the optimization amplitude of the previous round is not larger than the specific amplitude threshold value in the previous round of optimization, reducing or not adjusting the search threshold value in the next round of optimization by a second adjustment amplitude; and the value of the first adjustment amplitude is larger than that of the second adjustment amplitude. For example, a 64-line radar is also used, the preset threshold is preset, and is generally set to be 3-9 units (the unit here refers to a space cube, refer to fig. 6), and a certain range may also be set, and the dynamic adjustment is performed according to the optimization condition, for example, the edge optimization amplitude of the previous round is large, so that the search threshold in the next round may be reduced more, and the edge optimization amplitude of the previous round is small, so that the search threshold in the next round of optimization may be reduced little or no.
The process of iterative optimization for repeating the first and second operations has been described in detail above. It is known that the iterative optimization process is necessarily accompanied by a cutoff condition for ending the iterative optimization process.
According to an embodiment of the present invention, while the first operation and the second operation are repeated to perform iterative optimization on the optimized pose graph model by using the pose of the at least one key landmark, the number of iterations is counted. And further, when the counted iteration times reach an iteration time threshold, stopping iteration optimization and outputting the optimized pose graph model.
According to an embodiment of the invention, the method further comprises: and when the search threshold is smaller than a preset threshold, stopping iterative optimization and outputting the optimized pose graph model.
And stopping iterative optimization and outputting an optimized pose graph model when the counted iteration times reach an iteration time threshold or the search threshold is smaller than a preset threshold through the two listed cut-off judgment conditions, thereby realizing the graph model optimization of the relative precision map.
Compared with the traditional pose graph model optimization shown in fig. 1, the method for optimizing the relative precision map provided by the invention can optimize the map acquisition data of the whole non-closed loop model.
Moreover, even if a plurality of batches of maps intersect at the same intersection in closed-loop map acquisition data and the confidence of each batch of maps cannot be determined, the invention obtains the pose of a key landmark (a real intersection) by an absolute precision measurement technology, thereby solving the problem of fusion deformation among the plurality of batches of maps.
In addition, aiming at the problem that the nodes of the map cannot search the key landmarks of the absolute accuracy map due to the large error of the relative accuracy map, the invention can respectively obtain edges with different information contents to carry out iterative pose map model optimization by using a coarse-to-fine matching mode, and finally the relative accuracy map is continuously close to the absolute accuracy map.
Fig. 7 is a schematic diagram illustrating a configuration of an apparatus for optimizing a relative-accuracy map according to an embodiment of the present invention. As shown in fig. 7, the apparatus 70 for optimizing a relative-accuracy map includes: an absolute accuracy measurement device 701, configured to perform absolute accuracy measurement on at least one key landmark in a specific environment region, to obtain a pose of the at least one key landmark; a relative accuracy model building device 702, configured to build a relative accuracy map of the specific environment area, and build a pose graph model of the relative accuracy map, where a start point and an end point of the pose graph model do not coincide; a pose graph model optimizing device 703 configured to optimize the pose graph model using the pose of the at least one key landmark.
According to an embodiment of the present invention, the pose graph model optimization apparatus 703 is further configured to detect whether an intermediate node of the pose graph model forms a closed loop; and if the intermediate node of the pose graph model is detected to form a closed loop, pre-optimizing the pose graph model of the relative precision map based on the intermediate node forming the closed loop.
According to an embodiment of the present invention, the pose graph model optimization apparatus 703 at least includes the following apparatuses: a first means 7031 for searching the pose graph model for the pose of the node corresponding to the pose of the at least one key landmark using a particular matching algorithm; a second means 7032 for optimizing the pose graph model based on a transformation relationship between the pose of the at least one key landmark and the pose of the corresponding node.
According to an embodiment of the invention, the specific matching algorithm comprises ICP or PnP.
According to an embodiment of the present invention, the second device 7032 is further configured to establish a transformation relationship between the pose of the at least one key landmark and the pose of the corresponding node, and use the established transformation relationship between the pose of the at least one key landmark and the pose of the corresponding node as a new edge; constructing a transformation relation between adjacent nodes in the pose graph model, and taking the transformation relation between the adjacent nodes as a primary side; and optimizing a pose graph model formed by combining the new edge and the primary edge through a specific graph model optimization algorithm.
According to an embodiment of the invention, the specific graph model optimization algorithm comprises least squares or LM.
According to an embodiment of the present invention, the pose graph model optimization apparatus 703 further includes: resolution increasing means 7033 for increasing the resolution of the pose of the at least one key landmark and the pose of the corresponding node after the pose graph model is optimized by the second means; a repeating unit 7034 configured to repeat the first operation and the second operation to perform iterative optimization on the optimized pose graph model by using the pose of the at least one key landmark.
According to an embodiment of the present invention, the resolution increasing means 7033 is further configured to increase the resolution of the pose of the at least one key landmark and the pose of the corresponding node by a double resolution.
According to an embodiment of the present invention, the first device 7031 is further configured to determine a search threshold; detecting whether the search threshold is greater than a predetermined threshold; if the searching threshold is detected to be larger than the preset threshold, searching the pose of the node corresponding to the pose of the at least one key landmark in the pose graph model according to the searching range corresponding to the searching threshold.
According to an embodiment of the present invention, the first means 7031 is further configured to determine an optimization magnitude of a previous round in the iterative optimization process; dynamically adjusting the search threshold according to the magnitude of the optimized magnitude.
According to an embodiment of the present invention, the first means 7031 is further configured to decrease the search threshold in the next round of optimization by a first adjustment range if the optimization range in the previous round is greater than the specific range threshold in the previous round of optimization; if the optimization amplitude of the previous round is not larger than the specific amplitude threshold value in the previous round of optimization, reducing or not adjusting the search threshold value in the next round of optimization by a second adjustment amplitude; and the value of the first adjustment amplitude is larger than that of the second adjustment amplitude.
According to an embodiment of the present invention, the apparatus 70 further includes an output device 704, configured to stop the iterative optimization and output the optimized pose graph model when the search threshold is detected to be smaller than the predetermined threshold.
According to an embodiment of the present invention, the apparatus 70 further includes a statistics device 705 for counting the number of iterations while repeating the first operation and the second operation to iteratively optimize the optimized pose graph model by using the at least one key landmark.
According to an embodiment of the present invention, the output device 704 is further configured to stop the iterative optimization and output the optimized pose graph model when the counted number of iterations reaches an iteration number threshold.
Here, it should be noted that: the description of the above device embodiment is similar to the description of the above method, and the description of the beneficial effects of the method is not repeated. For technical details not disclosed in the embodiments of the apparatus of the present invention, reference is made to the description of the embodiments of the method of the present invention.
Exemplary device
Having described the method and apparatus of an exemplary embodiment of the present invention, next, an artificial intelligence based article generating apparatus according to another exemplary embodiment of the present invention is described.
As will be appreciated by one skilled in the art, aspects of the present invention may be embodied as a system, method or program product. Thus, various aspects of the invention may be embodied in the form of: an entirely hardware embodiment, an entirely software embodiment (including firmware, microcode, etc.) or an embodiment combining hardware and software aspects that may all generally be referred to herein as a "circuit," module "or" system.
In some possible embodiments, an artificial intelligence based article generation apparatus of the present invention can include at least one or more processors, and at least one memory. Wherein the memory stores a program that, when executed by the processor, causes the processor to perform the steps described herein, e.g., the processor may perform operation 401 as shown in fig. 4 to perform an absolute accuracy measurement of at least one key landmark in a particular environmental region to obtain a pose of the at least one key landmark; operation 402, constructing a relative accuracy map of the specific environmental area, and establishing a pose graph model of the relative accuracy map, where a start point and an end point of the pose graph model do not coincide; at operation 403, the pose graph model is optimized with the pose of the at least one key landmark.
Fig. 8 shows a schematic diagram of an apparatus for optimizing a relative accuracy map according to an embodiment of the present invention.
An apparatus for optimizing a relative-accuracy map according to this embodiment of the present invention is described below with reference to fig. 8. The device 800 shown in fig. 8 is only an example and should not bring any limitations to the functionality or scope of use of the embodiments of the present invention.
As shown in FIG. 8, device 800 is in the form of a general purpose computing device, including but not limited to: the at least one processor 810, the at least one memory 820, and a bus 860 connecting the various system components (including the memory 820 and the processor 810).
The bus 860 includes an address bus, a control bus, and a data bus.
The memory 820 may include volatile memory, such as Random Access Memory (RAM)821 and/or cache memory 822, and may further include Read Only Memory (ROM) 823.
Memory 820 may also include a set (at least one) of program modules 824, such program modules 824 including, but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which, or some combination thereof, may comprise an implementation of a network environment.
The device 800 may also communicate with one or more external devices 80 (e.g., keyboard, pointing device, bluetooth device, etc.). Such communication may occur via input/output (I/O) interface 840 and displayed on display unit 830. Also, device 800 may communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN) and/or a public network, such as the Internet) via network adapter 850. As shown, the network adapter 850 communicates with the other modules in the device 800 via a bus 860. It should be understood that although not shown in the figures, other hardware and/or software modules may be used in conjunction with the device 800, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, among others.
Exemplary program product
In some possible embodiments, aspects of the present invention may also be embodied in a computer-readable storage medium comprising program code to, when executed by a processor, cause the processor to perform the steps of the method described above, e.g., the processor may perform operation 401, as shown in fig. 4, of performing an absolute accuracy measurement of at least one key landmark in a particular environmental area, resulting in a pose of the at least one key landmark; operation 402, constructing a relative accuracy map of the specific environmental area, and establishing a pose graph model of the relative accuracy map, where a start point and an end point of the pose graph model do not coincide; at operation 403, the pose graph model is optimized with the pose of the at least one key landmark.
The computer-readable storage medium may take any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
FIG. 9 illustrates a schematic diagram of a computer-readable storage medium for optimizing relative accuracy maps, according to an embodiment of the present invention.
As shown in fig. 9, a computer-readable storage medium 3 according to an embodiment of the present invention is described, which can employ a portable compact disc read only memory (CD-ROM) and include program codes, and can be run on a terminal device, such as a personal computer. However, the computer-readable storage medium of the present invention is not limited thereto, and in this document, a readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
Program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device over any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computing device (e.g., over the internet using an internet service provider).
Moreover, while the operations of the method of the invention are depicted in the drawings in a particular order, this does not require or imply that the operations must be performed in this particular order, or that all of the illustrated operations must be performed, to achieve desirable results. Additionally or alternatively, certain steps may be omitted, multiple steps combined into one step execution, and/or one step broken down into multiple step executions.
While the spirit and principles of the invention have been described with reference to several particular embodiments, it is to be understood that the invention is not limited to the disclosed embodiments, nor is the division of aspects, which is for convenience only as the features in such aspects may not be combined to benefit. The invention is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.
Claims (26)
1. A method of optimizing a relative accuracy map, the method comprising:
measuring the absolute accuracy of at least one key landmark in a specific environment area to obtain the pose of the at least one key landmark;
constructing a relative precision map of the specific environment area, and establishing a pose graph model of the relative precision map, wherein the starting point and the end point of the pose graph model are not coincident;
optimizing the pose graph model with the pose of the at least one key landmark;
wherein optimizing the pose graph model using the pose of the at least one key landmark comprises at least the following: a first operation of searching the pose graph model for the pose of a node corresponding to the pose of the at least one key landmark using a particular matching algorithm; a second operation of optimizing the pose graph model based on a transformation relationship between the pose of the at least one key landmark and the pose of the corresponding node;
wherein optimizing the pose graph model using the pose of the at least one key landmark further comprises: after a second operation optimizes the pose graph model, increasing a resolution of the pose of the at least one key landmark and the pose of the corresponding node; and repeating the first operation and the second operation to perform iterative optimization on the optimized pose graph model by using the pose of the at least one key landmark.
2. The method of claim 1, wherein optimizing the pose graph model using the pose of the at least one key landmark comprises:
detecting whether the intermediate nodes of the pose graph model form a closed loop or not;
and if the intermediate node of the pose graph model is detected to form a closed loop, pre-optimizing the pose graph model of the relative precision map based on the intermediate node forming the closed loop.
3. The method of claim 1, wherein the particular matching algorithm comprises Iterative Closest Point (ICP) or projective transformation n points (PnP).
4. The method of claim 1, wherein optimizing the pose graph model based on a transformation relationship between the pose of the at least one key landmark and the pose of the corresponding node comprises:
establishing a transformation relation between the pose of the at least one key landmark and the pose of the corresponding node, and taking the established transformation relation between the pose of the at least one key landmark and the pose of the corresponding node as a new edge;
constructing a transformation relation between adjacent nodes in the pose graph model, and taking the transformation relation between the adjacent nodes as a primary side;
and optimizing a pose graph model formed by combining the new edge and the primary edge through a specific graph model optimization algorithm.
5. The method of claim 4, wherein the particular graph model optimization algorithm comprises a least squares method or a Levenberg algorithm (LM).
6. The method of claim 1, wherein the resolution of the pose of the at least one key landmark and the pose of the corresponding node are increased by a factor of two.
7. The method of claim 1, wherein the first operations further comprise: determining a search threshold; detecting whether the search threshold is greater than a predetermined threshold; if the searching threshold is detected to be larger than the preset threshold, searching the pose of the node corresponding to the pose of the at least one key landmark in the pose graph model according to the searching range corresponding to the searching threshold.
8. The method of claim 7, wherein the method further comprises: determining the optimization amplitude of the previous round in the iterative optimization process; dynamically adjusting the search threshold according to the magnitude of the optimized magnitude.
9. The method of claim 8, wherein dynamically adjusting the search threshold as a function of the magnitude of the optimization magnitude comprises: if the optimization amplitude of the previous round is larger than the specific amplitude threshold value during the previous round of optimization, reducing the search threshold value during the next round of optimization by using the first adjustment amplitude; if the optimization amplitude of the previous round is not larger than the specific amplitude threshold value in the previous round of optimization, reducing or not adjusting the search threshold value in the next round of optimization by a second adjustment amplitude; and the value of the first adjustment amplitude is larger than that of the second adjustment amplitude.
10. The method of claim 7, wherein the method further comprises: and when the search threshold is smaller than a preset threshold, stopping iterative optimization and outputting the optimized pose graph model.
11. The method of claim 1, wherein the number of iterations is counted while the first and second operations are repeated to iteratively optimize the optimized pose graph model with the pose of the at least one key landmark.
12. The method of claim 11, wherein the method further comprises: and when the counted iteration times reach an iteration time threshold value, stopping iterative optimization and outputting an optimized pose graph model.
13. An apparatus for optimizing a relative accuracy map, the apparatus comprising:
the absolute precision measuring device is used for measuring the absolute precision of at least one key landmark in a specific environment area to obtain the pose of the at least one key landmark;
the relative precision model building device is used for building a relative precision map of the specific environment area and building a pose graph model of the relative precision map, and the starting point and the end point of the pose graph model are not coincident;
a pose graph model optimization device for optimizing the pose graph model using the pose of the at least one key landmark;
the pose graph model optimization device at least comprises the following devices: first means for searching the pose graph model for a pose of a node corresponding to the pose of the at least one key landmark using a particular matching algorithm; second means for optimizing the pose graph model based on a transformation relationship between the pose of the at least one key landmark and the pose of the corresponding node;
wherein, the pose graph model optimizing device further comprises: resolution increasing means for increasing the resolution of the pose of the at least one key landmark and the pose of the corresponding node after the pose graph model is optimized by the second means; and the repeating device is used for repeating the first operation and the second operation so as to perform iterative optimization on the optimized pose graph model by using the pose of the at least one key landmark.
14. The apparatus of claim 13, wherein the pose graph model optimization means is further configured to detect whether intermediate nodes of the pose graph model form a closed loop; and if the intermediate node of the pose graph model is detected to form a closed loop, pre-optimizing the pose graph model of the relative precision map based on the intermediate node forming the closed loop.
15. The apparatus of claim 13, wherein the particular matching algorithm comprises ICP or PnP.
16. The apparatus of claim 13, wherein the second means is further for establishing a transformation relationship between the pose of the at least one key landmark and the pose of the corresponding node, and treating the established transformation relationship between the pose of the at least one key landmark and the pose of the corresponding node as a new edge; constructing a transformation relation between adjacent nodes in the pose graph model, and taking the transformation relation between the adjacent nodes as a primary side; and optimizing a pose graph model formed by combining the new edge and the primary edge through a specific graph model optimization algorithm.
17. The apparatus of claim 16, wherein the particular graph model optimization algorithm comprises a least squares or LM.
18. The apparatus of claim 13, wherein the resolution increasing means is further configured to increase the resolution of the pose of the at least one key landmark and the pose of the corresponding node by a doubling of the resolution.
19. The apparatus of claim 13, wherein the first means is further for, determining a search threshold; detecting whether the search threshold is greater than a predetermined threshold; if the searching threshold is detected to be larger than the preset threshold, searching the pose of the node corresponding to the pose of the at least one key landmark in the pose graph model according to the searching range corresponding to the searching threshold.
20. The apparatus of claim 19, wherein the first means is further for determining an optimization magnitude for a previous round in the iterative optimization process; dynamically adjusting the search threshold according to the magnitude of the optimized magnitude.
21. The apparatus of claim 20, wherein the first means is further configured to decrease the search threshold for the next round of optimization by a first adjustment amount if the optimization amount of the previous round is greater than the specific amount threshold for the previous round of optimization; if the optimization amplitude of the previous round is not larger than the specific amplitude threshold value in the previous round of optimization, reducing or not adjusting the search threshold value in the next round of optimization by a second adjustment amplitude; and the value of the first adjustment amplitude is larger than that of the second adjustment amplitude.
22. The apparatus of claim 19, further comprising: and the output device is used for stopping iterative optimization and outputting the optimized pose graph model when the search threshold is detected to be smaller than a preset threshold.
23. The apparatus of claim 13, further comprising: and the counting device is used for counting the iteration times while repeatedly performing the first operation and the second operation so as to perform iterative optimization on the optimized pose graph model by using the at least one key landmark.
24. The apparatus of claim 23, further comprising: and the output device is used for stopping iterative optimization and outputting the optimized pose graph model when the counted iteration times reach the iteration time threshold.
25. An apparatus for optimizing a relative accuracy map, comprising:
one or more processors;
a memory;
a program stored in the memory, which when executed by the one or more processors, causes the processors to perform the method of any one of claims 1-12.
26. A computer-readable storage medium storing a program which, when executed by a processor, causes the processor to perform the method of any one of claims 1-12.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810312477.1A CN108520543B (en) | 2018-04-09 | 2018-04-09 | Method, equipment and storage medium for optimizing relative precision map |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810312477.1A CN108520543B (en) | 2018-04-09 | 2018-04-09 | Method, equipment and storage medium for optimizing relative precision map |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108520543A CN108520543A (en) | 2018-09-11 |
| CN108520543B true CN108520543B (en) | 2022-08-09 |
Family
ID=63432097
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810312477.1A Active CN108520543B (en) | 2018-04-09 | 2018-04-09 | Method, equipment and storage medium for optimizing relative precision map |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108520543B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109461208B (en) * | 2018-11-15 | 2022-12-16 | 杭州易现先进科技有限公司 | Three-dimensional map processing method, device, medium and computing equipment |
| CN109813319B (en) * | 2019-03-07 | 2021-09-28 | 山东大学 | Open loop optimization method and system based on SLAM (Simultaneous localization and mapping) mapping |
| CN110196044A (en) * | 2019-05-28 | 2019-09-03 | 广东亿嘉和科技有限公司 | It is a kind of based on GPS closed loop detection Intelligent Mobile Robot build drawing method |
| CN111856499B (en) * | 2020-07-30 | 2021-06-18 | 浙江华睿科技有限公司 | Map construction method and device based on laser radar |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8452535B2 (en) * | 2010-12-13 | 2013-05-28 | GM Global Technology Operations LLC | Systems and methods for precise sub-lane vehicle positioning |
| CN106296812A (en) * | 2016-08-18 | 2017-01-04 | 宁波傲视智绘光电科技有限公司 | Synchronize location and build drawing method |
| CN106643720A (en) * | 2016-09-28 | 2017-05-10 | 深圳市普渡科技有限公司 | Method for map construction based on UWB indoor locating technology and laser radar |
| CN107025661A (en) * | 2016-01-29 | 2017-08-08 | 成都理想境界科技有限公司 | A kind of method for realizing augmented reality, server, terminal and system |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104834309B (en) * | 2015-04-10 | 2018-08-21 | 浙江工业大学 | The optimal touring control method of single mobile robot based on target following control strategy |
| US10203209B2 (en) * | 2016-05-25 | 2019-02-12 | Regents Of The University Of Minnesota | Resource-aware large-scale cooperative 3D mapping using multiple mobile devices |
-
2018
- 2018-04-09 CN CN201810312477.1A patent/CN108520543B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8452535B2 (en) * | 2010-12-13 | 2013-05-28 | GM Global Technology Operations LLC | Systems and methods for precise sub-lane vehicle positioning |
| CN107025661A (en) * | 2016-01-29 | 2017-08-08 | 成都理想境界科技有限公司 | A kind of method for realizing augmented reality, server, terminal and system |
| CN106296812A (en) * | 2016-08-18 | 2017-01-04 | 宁波傲视智绘光电科技有限公司 | Synchronize location and build drawing method |
| CN106643720A (en) * | 2016-09-28 | 2017-05-10 | 深圳市普渡科技有限公司 | Method for map construction based on UWB indoor locating technology and laser radar |
Non-Patent Citations (1)
| Title |
|---|
| 面向机器人室内建图的RGB-D图像对齐算法;王峰等;《机器人 》;20150315;第37卷(第3期);第129-135页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108520543A (en) | 2018-09-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107909612B (en) | Method and system for visual instant positioning and mapping based on 3D point cloud | |
| CN108520543B (en) | Method, equipment and storage medium for optimizing relative precision map | |
| CN109521403B (en) | Parameter calibration method, device and equipment of multi-line laser radar and readable medium | |
| CN109270545B (en) | Positioning true value verification method, device, equipment and storage medium | |
| KR102548282B1 (en) | High-precision mapping method and device | |
| CN113376650B (en) | Mobile robot positioning method and device, electronic equipment and storage medium | |
| CN107784671B (en) | Method and system for visual instant positioning and drawing | |
| CN111094895B (en) | System and method for robust self-repositioning in pre-constructed visual maps | |
| Aider et al. | A model-based method for indoor mobile robot localization using monocular vision and straight-line correspondences | |
| Bülow et al. | Spectral 6dof registration of noisy 3d range data with partial overlap | |
| US20140142891A1 (en) | Generaton of map data | |
| CN112444246B (en) | Laser fusion positioning method in high-precision digital twin scene | |
| US9513130B1 (en) | Variable environment high integrity registration transformation system and related method | |
| CN111680596A (en) | Positioning truth value verification method, device, equipment and medium based on deep learning | |
| Ghosh et al. | Multi sensor data fusion for 6D pose estimation and 3D underground mine mapping using autonomous mobile robot | |
| Ratter et al. | GPU accelerated graph SLAM and occupancy voxel based ICP for encoder-free mobile robots | |
| TW202238449A (en) | Indoor positioning system and indoor positioning method | |
| Tseng et al. | A new architecture for simultaneous localization and mapping: an application of a planetary rover | |
| CN113822996A (en) | Pose estimation method and device for robot, electronic device and storage medium | |
| CN115239899B (en) | Pose map generation method, high-precision map generation method and device | |
| Lei et al. | Pose estimation of a noncooperative target based on monocular visual slam | |
| Shmatko et al. | Estimation of rotation measurement error of objects using computer simulation | |
| CN113516772B (en) | Three-dimensional scene reconstruction method and device, BIM model optimization method and device | |
| Arzberger et al. | Towards spherical robots for mobile mapping in human made environments | |
| Jiang et al. | 3D reconstruction of spherical images based on incremental structure from motion |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20190618 Address after: 311215 Room 102, 6 Blocks, C District, Qianjiang Century Park, Xiaoshan District, Hangzhou City, Zhejiang Province Applicant after: Hangzhou Yixian Advanced Technology Co.,Ltd. Address before: 310052 Building No. 599, Changhe Street Network Business Road, Binjiang District, Hangzhou City, Zhejiang Province, 4, 7 stories Applicant before: NETEASE (HANGZHOU) NETWORK Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |