CN113724385A - Robot topology map generation system, method, computer device and storage medium - Google Patents
Robot topology map generation system, method, computer device and storage medium Download PDFInfo
- Publication number
- CN113724385A CN113724385A CN202110874179.3A CN202110874179A CN113724385A CN 113724385 A CN113724385 A CN 113724385A CN 202110874179 A CN202110874179 A CN 202110874179A CN 113724385 A CN113724385 A CN 113724385A
- Authority
- CN
- China
- Prior art keywords
- track
- map
- scene
- generate
- image
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000013507 mapping Methods 0.000 claims abstract description 78
- 230000009471 action Effects 0.000 claims abstract description 39
- 238000010586 diagram Methods 0.000 claims abstract description 32
- 238000001914 filtration Methods 0.000 claims abstract description 21
- 238000005457 optimization Methods 0.000 claims abstract description 13
- 238000007781 pre-processing Methods 0.000 claims description 11
- 230000008569 process Effects 0.000 claims description 10
- 238000009499 grossing Methods 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 3
- 238000000265 homogenisation Methods 0.000 description 8
- 230000000877 morphologic effect Effects 0.000 description 6
- 241000282414 Homo sapiens Species 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000004590 computer program Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/05—Geographic models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10028—Range image; Depth image; 3D point clouds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
- G06T2207/20032—Median filtering
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Remote Sensing (AREA)
- Computer Graphics (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention relates to the field of robot navigation, and discloses a robot topological map generation system, a robot topological map generation method, a robot topological map generation computer device and a storage medium, wherein the robot topological map generation system comprises a memory, a processor and computer readable instructions which are stored in the memory and can run on the processor, and the following steps are realized when the processor executes the computer readable instructions: acquiring an action track of the robot in the mapping scene and a scene map of the mapping scene; filtering the action track to generate a uniform point track; processing a scene map to generate a scene skeleton map; mapping the homogenized point track to a scene skeleton diagram, and generating a track skeleton diagram through image optimization processing; and carrying out linearization processing on the track skeleton diagram to generate a topological map of the mapping scene. The invention greatly improves the drawing efficiency of the topological map and reduces the artificial errors of the topological map.
Description
Technical Field
The invention relates to the field of robot navigation, in particular to a robot topological map generation system, a robot topological map generation method, computer equipment and a storage medium.
Background
In the automatic navigation process of the robot, a pre-drawn topological map is usually relied on. Topological maps are typically drawn by human beings. However, for a complex scene, the process of drawing the topological map is complicated, time and labor are wasted, and the situation that the topological map is not matched with the actual scene exists, which affects the normal operation of the robot.
Disclosure of Invention
Therefore, in order to solve the above technical problems, it is necessary to provide a robot topology map generation system, a robot topology map generation method, a computer device, and a storage medium, so as to improve the drawing efficiency of the topology map and reduce the human error of the topology map.
A robotic topology map generation system comprising a memory, a processor, and computer readable instructions stored in the memory and executable on the processor, the processor when executing the computer readable instructions implementing the steps of:
acquiring an action track of a robot in an image establishing scene and a scene map of the image establishing scene;
filtering the action track to generate a uniform point track;
processing the scene map to generate a scene skeleton map;
mapping the homogenized point track to the scene skeleton map, and performing image optimization processing to generate a track skeleton map;
and carrying out linearization processing on the track skeleton diagram to generate a topological map of the mapping scene.
A robot topological map generation method comprises the following steps:
acquiring an action track of a robot in an image establishing scene and a scene map of the image establishing scene;
filtering the action track to generate a uniform point track;
processing the scene map to generate a scene skeleton map;
mapping the homogenized point track to the scene skeleton map, and performing image optimization processing to generate a track skeleton map;
and carrying out linearization processing on the track skeleton diagram to generate a topological map of the mapping scene.
A computer device comprising a memory, a processor and computer readable instructions stored in the memory and executable on the processor, the processor being configured to execute the computer readable instructions to implement the robot topology map generation method described above.
One or more readable storage media storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the robot topology map generation method as described above.
According to the method, the topological map can be automatically generated only by acquiring the action track and the scene map, so that the drawing efficiency of the topological map is greatly improved, and the human errors of the topological map are reduced.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments of the present invention will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without inventive labor.
FIG. 1 is a flow chart of a robot topology map generation system according to an embodiment of the present invention;
FIG. 2 is an image including a trajectory of uniformized points after filtering processing according to an embodiment of the present invention;
FIG. 3 is a diagram of a scene skeleton after inverse processing according to an embodiment of the present invention;
FIG. 4 is a skeletal diagram of a trajectory after inversion processing in accordance with an embodiment of the present invention;
FIG. 5 is a topological map after inversion processing in accordance with an embodiment of the present invention;
FIG. 6 is a schematic diagram of a computer device according to an embodiment of the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1, a robot topology map generation system includes a memory, a processor, and computer program computer readable instructions stored in the memory and executable on the processor, wherein the processor is configured to execute the computer readable instructions to implement the following steps:
s10, acquiring the action track of the robot in the mapping scene and the scene map of the mapping scene.
Understandably, the mapping scenario may be a work scenario of the robot, such as a restaurant, hotel, hospital, or other indoor or outdoor scenario. The action track may be a walking track of the robot in a mapping scenario. The action trajectory comprises a point trajectory of several robots. In an example, the point trajectories may be acquired at preset time intervals. The preset time interval can be set according to actual needs. For example, the preset time interval may be 0.1s to 1s, and specifically may be 0.1s, 0.5s, or 1s, which is not particularly limited herein. In other examples, the point trajectories may be collected according to a preset displacement interval and/or an angle interval, for example, the point trajectories are collected once every 0.2m of movement, or once every 20 degrees of change of the angle, and the like, and the specific numerical values are not limited herein.
The scene map may be a planar map previously drawn by creating a scene, or a map drawn by depth image data. Typically, the scene map is a two-dimensional map.
In an example, the mapping scenario may be a restaurant. When the robot enters the restaurant, a worker can push the robot to travel along the pedestrian path of the restaurant to form a movement track, and meanwhile, the laser radar or the depth camera on the robot body can acquire depth image data along the pedestrian path. A two-dimensional map can be constructed from these depth image data and the action trajectory.
In alternative embodiments, the robot may be pushed to walk along the restaurant once, or multiple times. And optionally, after the pushing robot finishes walking to the path of the restaurant to be mapped from the starting point, the pushing robot needs to be pushed to return to the starting point when the map is built, so that the map pushing is finished.
In other embodiments, the robot may be guided to follow the user to go round the restaurant by way of a guidance control.
In an alternative embodiment, the depth camera may be a multi-view camera, an RGBD camera, or a monocular camera, and the depth information is acquired by tracking consecutive image frames of the monocular camera.
S20, filtering the action track to generate a uniform point track; and processing the scene map to generate a scene skeleton map.
Understandably, in the action track, the time intervals of the acquisition of the adjacent track points are equal, and the distances between the adjacent track points have certain difference. The spacing is related to the speed of movement of the robot over the time interval. The motion trajectory needs to be filtered, so that the difference of the distances is reduced, and a uniform point trajectory is obtained. The uniform point track means that the distance between adjacent track points is within a distance range, so that the uniform distance is ensured, and the condition of overlarge or undersize does not occur. As shown in fig. 2, fig. 2 is an image including a homogenized point trajectory after being filtered. Wherein the linear curve in fig. 2 is a uniform point trajectory.
And the scene map is filtered, so that noise in the scene map can be filtered. And carrying out binarization processing on the filtered image to obtain a binarized image, so that information in a scene map can be conveniently extracted, and the identification efficiency is improved. In an example, the feasible area without obstacles in the filtered scene map may be set to 1, and the area with obstacles may be set to 0. The binary image can be processed by adopting a morphological image processing method, the framework of the binary image is extracted, and a scene framework image is generated. In an example, in a scene skeleton map, the pixel value of the skeleton position is set to 255. As shown in fig. 3, fig. 3 is a skeleton diagram of a scene after being processed in an inverted manner. Here, after the inversion processing, the pixel value of the skeleton position is set to 0.
And S30, mapping the homogenization point track to the scene skeleton map, and generating a track skeleton map through image optimization processing.
Understandably, the homogenization point trajectory can be mapped to the scene skeleton map according to a preset mapping rule (such as a scale factor). After the uniformization point tracks are added on the scene skeleton diagram, appropriate optimization measures can be selected to optimize the scene skeleton diagram added with the uniformization point tracks, so that all track points in the uniformization point tracks are connected together and are subjected to thinning treatment to form the track skeleton diagram. As shown in fig. 4, fig. 4 is a diagram illustrating a skeleton of a trajectory after inversion processing. The track skeleton in the track skeleton diagram is composed of a plurality of sections of skeletons, and each section of skeleton is an irregular curve line segment.
S40, carrying out linearization processing on the track skeleton diagram to generate a topological map of the mapping scene.
Understandably, each segment of the skeleton in the trajectory skeleton map may be subjected to a linearization process (i.e., straight line fitting), and the straight line segment after the linearization process is output as a path. Meanwhile, the point where the more than three straight line segments intersect is taken as a node to be output. The topological map includes the output paths and nodes.
In the process of carrying out linearization, a point set of the segmented skeleton can be used as the input of the straight-line segment, and the straight-line segment can be obtained through least square processing. The least squares method uses the following formula:
wherein e is an error term, (x)i，yi) Coordinates of the ith point in the point set of the segmented skeleton are obtained, and a and b are parameters to be solved. When the value of e is minimum, the values of a and b can be solved.
As shown in fig. 5, fig. 5 is a topology map processed by inversion.
In the embodiment, the topological map can be automatically generated only by acquiring the action track and the scene map, so that the drawing efficiency of the topological map is greatly improved, and the human errors of the topological map are reduced.
Optionally, in step S10, the acquiring an action trajectory of the robot in the mapping scene and a scene map of the mapping scene includes:
s101, acquiring depth image data acquired when the robot passes through the action track, wherein the depth image data is acquired through a depth camera and/or a laser radar;
and S102, splicing the depth image data according to the position of the robot in the action track to generate the scene map.
Understandably, the depth image data may be depth information data acquired by a depth camera and/or a lidar mounted on the robot body. The depth camera or the laser radar can acquire depth image data of the mapping scene on different track points. And splicing the depth image data based on the positions of the track points to generate a scene map of the mapping scene.
For example, the depth image data acquired by the laser radar may be specifically a three-dimensional point cloud, and then the three-dimensional point cloud is subjected to two-dimensional projection to generate a two-dimensional scene map. Here, a two-dimensional scene map may be generated by projecting the three-dimensional point cloud onto a ground plane (which may be a horizontal plane) of the scene under construction.
Optionally, in step S20, the filtering the action trajectory to generate a homogenized point trajectory includes:
s201, interpolating the action track, and then performing median filtering and mean value smoothing to generate the homogenization point track; in the uniformization point track, the distance between adjacent track points is larger than a first threshold value and smaller than a second threshold value.
Understandably, the motion track can be interpolated, the number of track points is increased, and the distance between all adjacent track points is smaller than a first threshold value. And then removing track points with undersized intervals through median filtering, and then performing mean value smoothing treatment to enable the intervals of the adjacent track points to be approximately equal, thereby obtaining a uniform point track. In the uniformization point track, the distance between the adjacent track points is larger than a first threshold value and smaller than a second threshold value.
Optionally, the first threshold includes 0.25m, and the second threshold includes 1 m.
Understandably, the first threshold and the second threshold can be set according to actual needs. In an example, the first threshold comprises 0.25m and the second threshold comprises 1 m.
Optionally, in step S20, that is, the processing the scene map to generate the scene skeleton map includes:
s202, carrying out image preprocessing on the scene map to generate an image preprocessing image;
s203, carrying out binarization processing on the image preprocessing image to generate a binarized image;
and S204, performing skeleton extraction on the binary image to generate the scene skeleton map.
Understandably, the scene map may be image preprocessed to generate an image preprocessed image. In an example, the image pre-processing may be image median filtering and image mean filtering. The image median filtering and the image mean filtering can filter noise in the scene map and optimize the image quality of the scene map. The image preprocessing image is subjected to binarization processing to generate a binarization image, so that information in a scene map can be conveniently extracted, and the identification efficiency of the obstacle information is improved. And performing morphological processing on the binary image, extracting a skeleton, and generating a scene skeleton map. In the scene skeleton map, the pixel value of the skeleton position is 255.
The binarization processing is carried out on the filtered image, specifically, a pixel/grid with an obstacle is recorded as 0, and a pixel/grid without the obstacle is recorded as 1; alternatively, the pixel/grid of the obstacle is recorded as 1, and the pixel/grid without the obstacle is recorded as 0, which is not limited herein. The binarized image is then morphologically processed, i.e., the skeleton of the continuous, non-obstacle-present pixel/grid region is extracted. Here, the skeleton may refer to a morphological skeleton calculated using morphological operators. The morphological framework may include two forms, one defined by the morphological opening, which may reconstruct the original shape of the opening; the other is computed by hit-or-miss transformation (a transformation mechanism) which preserves the topology of the shape.
Optionally, in step S30, the mapping the homogenized point trajectory to the scene skeleton map, and generating a trajectory skeleton map through image optimization processing, includes:
s301, mapping the homogenization point track to the scene skeleton map to generate a track mapping image;
s302, performing expansion operation and filling operation on the track mapping image to generate a closed area image;
s303, thinning the edge of the closed region image to generate the track skeleton diagram.
Understandably, in step S301, the trajectory of the uniformization point may be mapped to the scene skeleton map, and a trajectory mapping image may be generated. When mapping is carried out, the coordinates of track points in the uniformization point track can be amplified or reduced in an equal proportion. For example, the scale factor is 0.05, the coordinates of the track point are (100 ), and after mapping to the scene skeleton map, the track mapping coordinates corresponding to the track point are (5, 5).
In step S302, an expansion operation and a filling operation may be performed on the trajectory mapping image to generate a closed region image. And performing expansion operation on the track mapping image, so that all track points in the original homogenized point track can be connected together, and then performing filling operation to form a plurality of closed areas, namely closed area images. For example, the original track points have a diameter of 4px and a track point interval of 4 px; if the track points expand by one time and the diameter is changed to 8px, the adjacent track points are partially overlapped, and the originally dispersed track points are connected with each other to form a continuous line segment. Wherein the partial line segments may enclose a closed area.
In step S303, the edge of the closed region image may be refined to generate a skeleton map of the trajectory. As shown in fig. 4, fig. 4 is a skeleton diagram of a trajectory obtained after thinning processing and inversion processing.
Optionally, in step S301, the mapping the homogenized point trajectory to the scene skeleton map to generate a trajectory mapping image includes:
s3011, obtaining a scale factor of the homogenization point track;
s3012, converting the point coordinates of the homogenized point track into mapping coordinates according to the scale factors;
and S3013, adding track points in the scene skeleton map according to the mapping coordinates, and generating the track mapping image.
Understandably, the scale factor corresponds to the scale of the map, and the coordinates of the track points can be enlarged or reduced. The scale factor changes with the change of the scale size between the uniformization point track and the scene skeleton diagram. Step S3011 is executed to obtain a scale factor for homogenizing the dot trajectory. In one example, the scale factor has a value of 0.05.
In step S3012, the point coordinates of the homogenized point trajectory may be converted into mapping coordinates according to a scale factor. In one example, the mapping coordinates are the product of the point coordinates of the uniformized point trajectory and the inverse of the scale factor, formulated as:
wherein, P1(X1，Y1) To map the coordinates, P (X, Y) is the point coordinates of the homogenized point trajectory, scale is the scale factor.
In step S3012, a track point may be added to the scene skeleton map according to the mapping coordinates, and a track mapping image may be generated. The trajectory mapping image includes obstacle information in the original scene skeleton diagram and trajectory data of the robot.
Optionally, in step S40, performing linearization processing on the trajectory skeleton diagram to generate a topological map of the mapping scene, where the step includes:
s401, grouping track points in the track skeleton diagram to obtain a plurality of track point sets, wherein one track point set corresponds to one section of skeleton;
s402, performing linearization processing on the track point set through a least square method to generate a straight line segment;
s403, selecting intersection points of straight line segments with more than a specified number, and setting the intersection points as nodes;
s404, generating the topological map according to the straight line segment and the node.
Understandably, the track skeleton map comprises a plurality of skeleton sections. Each section of framework is composed of a plurality of track points, and the track points can form a track point set. That is, one set of track points corresponds to one section of skeleton.
The trajectory point set can be subjected to linearization processing by a least square method to generate a straight line segment. A set of trajectory points may generate a straight line segment. In one example, the least squares method uses the following equation:
wherein e is an error term, (x)i，yi) And (4) coordinates of the ith point in the track point set, wherein a and b are parameters to be solved. When the value of e is minimum, the values of a and b can be solved.
After the linearization treatment, a plurality of straight line segments can be obtained. Intersection points of straight line segments with more than a specified number can be selected and set as nodes. Here, the specified number of pieces may be 3, and the node represents a branching intersection in the map. In other embodiments, the specified number of bars may also be 2. For example, there is one straight line segment AB (points a and B are the two endpoints of the straight line segment AB, respectively), and another straight line segment CD (points C and D are the two endpoints of the straight line segment CD, respectively), and point C is on the straight line segment (i.e., between point a and point B). At this time, point C is a node.
All the straight line segments and nodes can be combined to generate a topological map of the mapping scene. The straight line segments represent paths of the map, the nodes represent intersection points of the paths, namely, branched intersections, namely, the straight line segments form nodes of the topological map, and the nodes form nodes of the topological map.
The modules in the robot topology map generation system can be wholly or partially realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In an embodiment, a robot topological map generation method is provided, and the robot topological map generation method corresponds to the robot topological map generation system in the embodiment one to one. As shown in fig. 1, the robot topology map generation method provided in this embodiment includes the following steps:
s10, acquiring an action track of the robot in the mapping scene and a scene map of the mapping scene;
s20, filtering the action track to generate a uniform point track; processing the scene map to generate a scene skeleton map;
s30, mapping the homogenization point track to the scene skeleton diagram, and generating a track skeleton diagram through image optimization processing;
and S40, carrying out linearization processing on the track skeleton diagram to generate a topological map of the mapping scene.
Optionally, in step S10, the acquiring an action trajectory of the robot in the mapping scene and a scene map of the mapping scene includes:
s101, acquiring depth image data acquired when the robot passes through the action track, wherein the depth image data is acquired through a depth camera and/or a laser radar;
s102, splicing the depth image data according to the position of the robot in the action track to generate three-dimensional point cloud of the mapping scene;
s103, performing two-dimensional projection on the three-dimensional point cloud to generate the scene map.
Optionally, in step S20, the filtering the action trajectory to generate a homogenized point trajectory includes:
s201, interpolating the action track, and then performing median filtering and mean value smoothing to generate the uniform point track, wherein in the uniform point track, the distance between adjacent track points is greater than a first threshold value and smaller than a second threshold value.
Optionally, the first threshold includes 0.25m, and the second threshold includes 1 m.
Optionally, in step S20, that is, the processing the scene map to generate the scene skeleton map includes:
carrying out image preprocessing on the scene map to generate an image preprocessing image;
carrying out binarization processing on the image preprocessing image to generate a binarization image;
and performing skeleton extraction on the binary image to generate the scene skeleton map.
Optionally, in step S30, the mapping the homogenized point trajectory to the scene skeleton map, and generating a trajectory skeleton map through image optimization processing, includes:
s301, mapping the homogenization point track to the scene skeleton map to generate a track mapping image;
s302, performing expansion operation and filling operation on the track mapping image to generate a closed area image;
s303, thinning the edge of the closed region image to generate the track skeleton diagram.
Optionally, in step S301, the mapping the homogenized point trajectory to the scene skeleton map to generate a trajectory mapping image includes:
s3011, obtaining a scale factor of the homogenization point track;
s3012, converting the point coordinates of the homogenized point track into mapping coordinates according to the scale factors;
and S3013, adding track points in the scene skeleton map according to the mapping coordinates, and generating the track mapping image.
For specific limitations of the robot topology map generation method, reference may be made to the above limitations of the robot topology map generation system, which are not described herein again. It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.
In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in fig. 6. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a readable storage medium and an internal memory. The readable storage medium stores an operating system, computer readable instructions, and a database. The internal memory provides an environment for the operating system and execution of computer-readable instructions in the readable storage medium. The database of the computer device is used for storing data related to the robot topology map generation method. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer readable instructions, when executed by a processor, implement a robot topology map generation method. The readable storage media provided by the present embodiment include nonvolatile readable storage media and volatile readable storage media.
In one embodiment, a computer device is provided, comprising a memory, a processor, and computer readable instructions stored on the memory and executable on the processor, the processor when executing the computer readable instructions implementing the steps of:
acquiring an action track of a robot in an image establishing scene and a scene map of the image establishing scene;
filtering the action track to generate a uniform point track;
processing the scene map to generate a scene skeleton map;
mapping the homogenized point track to the scene skeleton map, and generating a track skeleton map through image optimization processing;
and carrying out linearization processing on the track skeleton diagram to generate a topological map of the mapping scene.
In one embodiment, one or more computer-readable storage media storing computer-readable instructions are provided, the readable storage media provided by the embodiments including non-volatile readable storage media and volatile readable storage media. The readable storage medium has stored thereon computer readable instructions which, when executed by one or more processors, perform the steps of:
acquiring an action track of a robot in an image establishing scene and a scene map of the image establishing scene;
filtering the action track to generate a uniform point track;
processing the scene map to generate a scene skeleton map;
mapping the homogenized point track to the scene skeleton map, and generating a track skeleton map through image optimization processing;
and carrying out linearization processing on the track skeleton diagram to generate a topological map of the mapping scene.
It will be understood by those of ordinary skill in the art that all or part of the processes of the methods of the above embodiments may be implemented by hardware related to computer readable instructions, which may be stored in a non-volatile readable storage medium or a volatile readable storage medium, and when executed, the computer readable instructions may include processes of the above embodiments of the methods. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of the functional units and modules is illustrated, and in practical applications, the above-mentioned function distribution may be performed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-mentioned functions.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present invention, and are intended to be included within the scope of the present invention.
Claims (11)
1. A robotic topology map generation system comprising a memory, a processor, and computer readable instructions stored in the memory and executable on the processor, wherein the processor is configured to execute the computer readable instructions to implement the steps of:
acquiring an action track of a robot in an image establishing scene and a scene map of the image establishing scene;
filtering the action track to generate a uniform point track;
processing the scene map to generate a scene skeleton map;
mapping the homogenized point track to the scene skeleton map, and performing image optimization processing to generate a track skeleton map;
and carrying out linearization processing on the track skeleton diagram to generate a topological map of the mapping scene.
2. The robot topology map generation system of claim 1, wherein said obtaining an action trajectory of a robot in a mapping scenario and a scenario map of the mapping scenario comprises:
acquiring depth image data acquired when the robot passes through the action track;
and splicing the depth image data according to the position of the robot in the action track to generate the scene map.
3. The robot topology map generating system according to claim 1, wherein the filtering the action trajectory to generate a homogenized point trajectory includes:
and interpolating the action track, and then performing median filtering and mean value smoothing to generate the homogenized point track, wherein in the homogenized point track, the distance between adjacent track points is greater than a first threshold value and smaller than a second threshold value.
4. The robotic topological map generation system of claim 3, wherein said first threshold comprises 0.25m and said second threshold comprises 1 m.
5. The robot topology map generation system of claim 1, said processing said scene map to generate a scene skeleton map, comprising:
carrying out image preprocessing on the scene map to generate an image preprocessing image;
carrying out binarization processing on the image preprocessing image to generate a binarization image;
and performing skeleton extraction on the binary image to generate the scene skeleton map.
6. The robot topology map generating system according to claim 1, wherein the mapping the homogenized point trajectory to the scene skeleton map and generating a trajectory skeleton map through image optimization processing includes:
mapping the homogenized point track to the scene skeleton map to generate a track mapping image;
performing expansion operation and filling operation on the track mapping image to generate a closed area image;
and thinning the edge of the closed region image to generate the track skeleton diagram.
7. The robot topology map generating system of claim 6, said mapping said homogenized point trajectory to said scene skeleton map, generating a trajectory mapping image, comprising:
obtaining a scale factor of the uniformization point track;
converting the point coordinates of the homogenized point track into mapping coordinates according to the scale factors;
and adding track points in the scene skeleton map according to the mapping coordinates to generate the track mapping image.
8. The robot topology map generation system of claim 1, wherein the performing the linearization process on the trajectory skeleton map to generate the topology map of the mapping scene comprises:
grouping the track points in the track skeleton diagram to obtain a plurality of track point sets, wherein one track point set corresponds to one section of skeleton;
performing linearization processing on the track point set by a least square method to generate a straight line segment;
selecting intersection points of straight line segments with more than a specified number, and setting the intersection points as nodes;
and generating the topological map according to the straight line segments and the nodes.
9. A robot topological map generation method is characterized by comprising the following steps:
acquiring an action track of a robot in an image establishing scene and a scene map of the image establishing scene;
filtering the action track to generate a uniform point track;
processing the scene map to generate a scene skeleton map;
mapping the homogenized point track to the scene skeleton map, and performing image optimization processing to generate a track skeleton map;
and carrying out linearization processing on the track skeleton diagram to generate a topological map of the mapping scene.
10. A computer device, wherein the processor is configured to execute the computer readable instructions to implement the robot topology map generation method of claim 9.
11. One or more readable storage media storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the robot topology map generation method of claim 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110874179.3A CN113724385A (en) | 2021-07-30 | 2021-07-30 | Robot topology map generation system, method, computer device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110874179.3A CN113724385A (en) | 2021-07-30 | 2021-07-30 | Robot topology map generation system, method, computer device and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113724385A true CN113724385A (en) | 2021-11-30 |
Family
ID=78674531
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110874179.3A Pending CN113724385A (en) | 2021-07-30 | 2021-07-30 | Robot topology map generation system, method, computer device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113724385A (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107704837A (en) * | 2017-10-19 | 2018-02-16 | 千寻位置网络有限公司 | The extracting method of road network topological sum geological information |
| CN110335282A (en) * | 2018-12-25 | 2019-10-15 | 广州启明星机器人有限公司 | A kind of profile line segment feature extraction algorithm based on grid |
| CN111158384A (en) * | 2020-04-08 | 2020-05-15 | 炬星科技（深圳）有限公司 | Robot mapping method, device and storage medium |
| CN111768489A (en) * | 2019-04-01 | 2020-10-13 | 顺丰科技有限公司 | Indoor navigation map construction method and system |
| CN112269848A (en) * | 2020-10-31 | 2021-01-26 | 武汉中海庭数据技术有限公司 | Crowdsourcing trajectory data fusion method and device |
| CN112328730A (en) * | 2021-01-04 | 2021-02-05 | 腾讯科技（深圳）有限公司 | Map data updating method, related device, equipment and storage medium |
| CN112683286A (en) * | 2021-03-16 | 2021-04-20 | 季华实验室 | Method and system for establishing topological road network map, storage medium and electronic equipment |
| CN113065594A (en) * | 2021-04-01 | 2021-07-02 | 中科星图空间技术有限公司 | Road network extraction method and device based on Beidou data and remote sensing image fusion |
-
2021
- 2021-07-30 CN CN202110874179.3A patent/CN113724385A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107704837A (en) * | 2017-10-19 | 2018-02-16 | 千寻位置网络有限公司 | The extracting method of road network topological sum geological information |
| CN110335282A (en) * | 2018-12-25 | 2019-10-15 | 广州启明星机器人有限公司 | A kind of profile line segment feature extraction algorithm based on grid |
| CN111768489A (en) * | 2019-04-01 | 2020-10-13 | 顺丰科技有限公司 | Indoor navigation map construction method and system |
| CN111158384A (en) * | 2020-04-08 | 2020-05-15 | 炬星科技（深圳）有限公司 | Robot mapping method, device and storage medium |
| CN112269848A (en) * | 2020-10-31 | 2021-01-26 | 武汉中海庭数据技术有限公司 | Crowdsourcing trajectory data fusion method and device |
| CN112328730A (en) * | 2021-01-04 | 2021-02-05 | 腾讯科技（深圳）有限公司 | Map data updating method, related device, equipment and storage medium |
| CN112683286A (en) * | 2021-03-16 | 2021-04-20 | 季华实验室 | Method and system for establishing topological road network map, storage medium and electronic equipment |
| CN113065594A (en) * | 2021-04-01 | 2021-07-02 | 中科星图空间技术有限公司 | Road network extraction method and device based on Beidou data and remote sensing image fusion |
Non-Patent Citations (2)
| Title |
|---|
| 王娜等: "基于细化算法的移动机器人拓扑地图创建", 计算机技术与发展, vol. 19, no. 10, pages 11 - 14 * |
| 蒋林;李峻;马先重;聂文康;朱建阳;雷斌;: "一种改进骨架提取的Voronoi路径规划", 机械工程学报, no. 13 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20220028163A1 (en) | Computer Vision Systems and Methods for Detecting and Modeling Features of Structures in Images | |
| CN109541634B (en) | Path planning method and device and mobile device | |
| CN109993780B (en) | Three-dimensional high-precision map generation method and device | |
| CN107610212B (en) | Scene reconstruction method and device, computer equipment and computer storage medium | |
| CN105160702B (en) | The stereopsis dense Stereo Matching method and system aided in based on LiDAR point cloud | |
| CN111968229A (en) | High-precision map making method and device | |
| CN111198378B (en) | Boundary-based autonomous exploration method and device | |
| CN112904369B (en) | Robot repositioning method, apparatus, robot, and computer-readable storage medium | |
| CN112306050B (en) | Autonomous robot, walking path planning method and device thereof and storage medium | |
| WO2021134285A1 (en) | Image tracking processing method and apparatus, and computer device and storage medium | |
| CN112434707B (en) | Traffic intersection RPP point automatic extraction method and device | |
| CN115267796B (en) | Positioning method, positioning device, robot and storage medium | |
| CN108564604B (en) | Binocular vision stereo matching method and device based on plane constraint and triangulation | |
| CN116540726A (en) | Intelligent obstacle avoidance method, system and medium for patrol robot | |
| CN114663598A (en) | Three-dimensional modeling method, device and storage medium | |
| CN113724383A (en) | Robot topology map generation system, method, computer device and storage medium | |
| CN113607161B (en) | Robot navigation path width acquisition system, method, robot and storage medium | |
| CN113724385A (en) | Robot topology map generation system, method, computer device and storage medium | |
| CN114459483B (en) | Landmark navigation map construction and application method and system based on robot navigation | |
| CN113724384A (en) | Robot topology map generation system, method, computer device and storage medium | |
| CN113920020A (en) | Human point cloud real-time repairing method based on depth generation model | |
| Kurka et al. | Automatic estimation of camera parameters from a solid calibration box | |
| Martinez et al. | Map-based lane identification and prediction for autonomous vehicles | |
| CN112433193A (en) | Multi-sensor-based mold position positioning method and system | |
| CN113741422B (en) | Robot topology map generation system, method, computer device and storage medium |
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 |