CN114659530A - Grid model map construction method for intelligent robot path planning - Google Patents
Grid model map construction method for intelligent robot path planning Download PDFInfo
- Publication number
- CN114659530A CN114659530A CN202210242363.0A CN202210242363A CN114659530A CN 114659530 A CN114659530 A CN 114659530A CN 202210242363 A CN202210242363 A CN 202210242363A CN 114659530 A CN114659530 A CN 114659530A
- Authority
- CN
- China
- Prior art keywords
- grid
- basic
- model
- robot
- map
- 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
- 238000013439 planning Methods 0.000 title claims abstract description 29
- 238000010276 construction Methods 0.000 title claims abstract description 16
- 238000000034 method Methods 0.000 claims abstract description 42
- 238000013507 mapping Methods 0.000 claims abstract description 16
- 230000000007 visual effect Effects 0.000 claims abstract description 7
- 238000006243 chemical reaction Methods 0.000 claims description 4
- 238000012512 characterization method Methods 0.000 claims 2
- 238000012876 topography Methods 0.000 claims 1
- 238000005070 sampling Methods 0.000 description 27
- 230000008569 process Effects 0.000 description 13
- 238000005259 measurement Methods 0.000 description 11
- 238000010586 diagram Methods 0.000 description 10
- 238000004088 simulation Methods 0.000 description 10
- 239000011159 matrix material Substances 0.000 description 9
- 238000004458 analytical method Methods 0.000 description 8
- 238000001228 spectrum Methods 0.000 description 8
- 238000011160 research Methods 0.000 description 7
- 230000033001 locomotion Effects 0.000 description 6
- 230000007613 environmental effect Effects 0.000 description 5
- 238000007726 management method Methods 0.000 description 5
- 239000013598 vector Substances 0.000 description 5
- 230000014509 gene expression Effects 0.000 description 4
- 230000009466 transformation Effects 0.000 description 4
- 230000007246 mechanism Effects 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 239000002689 soil Substances 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000000903 blocking effect Effects 0.000 description 2
- 210000002318 cardia Anatomy 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000013499 data model Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000008520 organization Effects 0.000 description 2
- QNRATNLHPGXHMA-XZHTYLCXSA-N (r)-(6-ethoxyquinolin-4-yl)-[(2s,4s,5r)-5-ethyl-1-azabicyclo[2.2.2]octan-2-yl]methanol;hydrochloride Chemical compound Cl.C([C@H]([C@H](C1)CC)C2)CN1[C@@H]2[C@H](O)C1=CC=NC2=CC=C(OCC)C=C21 QNRATNLHPGXHMA-XZHTYLCXSA-N 0.000 description 1
- 101100136092 Drosophila melanogaster peng gene Proteins 0.000 description 1
- 241000931888 Pyxis Species 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000019771 cognition Effects 0.000 description 1
- 238000013523 data management Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 229910052751 metal Inorganic materials 0.000 description 1
- 235000020030 perry Nutrition 0.000 description 1
- 238000010845 search algorithm Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
- G01C21/28—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network with correlation of data from several navigational instruments
- G01C21/30—Map- or contour-matching
- G01C21/32—Structuring or formatting of map data
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Automation & Control Theory (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention provides a regular hexagon grid model map construction method for intelligent robot path planning, which comprises the following steps: dividing a plane area in a visual range into a plurality of seamless-connected non-overlapping regular hexagonal blank basic grids according to the size of a preset grid and map data; acquiring point cloud data of each blank basic grid, endowing each blank basic grid with topographic features, grid element attributes and grid edge attributes by combining map data to obtain a basic grid, forming a grid model covering a plane area, determining coordinates of a robot and a laser point in the grid model, calculating a probability estimation value of the basic grid where the laser point is located, and determining whether the grid is in an occupied state or an idle state; updating the states of other basic grids between the basic grid where the robot is located and the basic grid where the laser point is located; and carrying out incremental mapping on the grid model of the environment according to the state of each basic grid. The map obtained by the method can be applied to map construction and path planning of the intelligent robot.
Description
Technical Field
The invention relates to a robot-oriented map construction method, in particular to a regular hexagon grid model map construction method for intelligent robot path planning.
Background
In the field of computer science, iran-foster (1998) proposed the concept of grids (Grid) and Grid computing. The system integrates high-speed internet, high-new energy computers, large databases, sensors and remote equipment into a whole, and provides more resources, functions and interactivity for users. Essentially, such a grid refers to a system of grids that are built such that information resources are on-demand as electrical resources. However, grid computing is too idealized, and the information resource composition is very complex and far from comparable to power resources; resource sharing and problem solving in a complex heterogeneous environment of cross-platform, cross-organization and cross-trust domain have great technical difficulty, so that grid computing in the general field is gradually replaced by cloud computing, and the concept of the grid is gradually weakened (wanggy et al, 2016).
However, after the grid concept is proposed, the concept of "spatial information grid" is proposed in the field of geographic information science, which is the fusion and integration of grid technology and spatial information technology. The grid model described herein is just a model that uses the concept of grids in the field of geographic information science. The intelligent robot-oriented grid model recognizes the spatial environment of the intelligent robot by taking the robot as a main body, and the fact is essentially in the same way as recognizing the spatial environment of the intelligent robot by taking the human as the main body. With the development of artificial intelligence, the robot actively recognizes the spatial environment, which is a necessary product of technical development.
Traditionally, grids in the field of geographic information science are another form of data organization corresponding to a vector data structure, called a grid data model. The space environment is divided into regular grids, and each grid is endowed with a corresponding attribute value to represent space entity data and finally used for describing the space environment. Though the grid map based on the grid data model is a relatively simple map type, the wording peng et al (2002) considers that the grid map divides a drawing area into grids according to plane coordinates or earth longitude and latitude, and describes or expresses attribute classification, statistical classification and variation parameters in the grids by taking the grids as units, which is equivalent to expressing the rule of dynamic space-time variation on a two-dimensional space. Therefore, the method has strong adaptability and diversity.
On the other hand, li derren et al (2003) propose a concept of a spatial information multilevel grid from the viewpoint of city management application. It divides the world and the whole country into grids with different thickness levels according to different grid sizes. Each grid of the system determines the geographic position by the longitude and latitude coordinates of the central point of the grid, and simultaneously records basic data items closely related to the grid. Yuanxian et al (2005), Li De ren et al (2006) and Yongzhi (2014) have intensively studied the basic principle and the application mode of the spatial information multilevel grid from the aspects of precision analysis, grid subdivision, coding design, data retrieval, scheduling strategy and the like. Because the concept of the spatial information multilevel grid is derived from the connotative extension of the city management application, the adopted grid is mostly irregular polygons corresponding to administrative regions, and the limitation is large.
Meanwhile, the theoretical research and application practice of global-level geographic information are becoming more intensive, and more effective management mechanisms for multi-scale data are needed to realize multi-level management and invocation of massive global data (Zhao Sheng, 2004). Therefore, goodcold et al (2000) propose a spherical Discrete Grid (Discrete Global Grid) that "divides the earth's large block area into progressively finer, mutually nested small areas based on the concept of loop segmentation". The spherical discrete grid abstracts the earth surface into a spherical surface, divides the spherical surface into a series of grid units according to a certain rule, describes spatial positions, forms and distributions by utilizing the grid units, organizes and manages spatial data, and realizes spatial object modeling, analysis and expression (Dutton, 1998; Sahr et al, 2003; cardia, 2006; Yuan Wen, 2004; Zhao Chong et al, 2007).
The spherical discrete grid meets the requirement of a computer on data discretization processing, gets rid of the constraint of map projection, and is expected to fundamentally solve the problems of data fracture, geometric deformation, topology inconsistency and the like of the traditional plane model in global space data management and multi-scale operation. However, from a technical point of view, spherical discrete meshes belong more to spatial localization schemes and spatial retrieval mechanisms (litderren et al, 2003).
The research result of the spatial information grid provides more subdivision methods for the research content of the text. In contrast, in the field of robotics, researches on probabilistic positioning, path planning, automatic control and the like are mostly concerned, and a map model used by a robot is rarely studied. Quijano and Garrido (2007) simulated the Exploration environment of a Robot Using Hexagonal grids instead of quadrangular grids, and analyzed the influence of the two grids on the search efficiency Using a variety of different path search algorithms (Quijano H J, Garrido L.Impropriation Cooperative Robot expansion Using an Hexagonal World retrieval [ C ]// Electronics, Robotics & automatic Mechanics conference. IEEE Computer Society, 2007.). Li et al (2017) analyzed the path Planning efficiency of unmanned surface vehicles (LI, Teng, et al. A. hexagonal grid-based sampling plan for an aqueous environmental monitoring system. in:2017IEEE International conference Systems, Man, and Cybernetics (SMC) IEEE,2017.p.3683-3688.) during unmanned surface vehicle environmental monitoring using hexagons to segment the sampling area and plan the path based on a Spanning Tree-based Planning method. Aiming at the defects of the traditional quadrilateral grids in environment description and path planning, the pottery philosophy and the like (2020) fully compare the construction processes of the traditional grids, rhombic grids, hexagonal grids and triangular grids, research the arrangement codes of the hexagonal grids, analyze motion indexes corresponding to different grids based on an octree strategy, and verify the advantages of the hexagonal grids in environment description, path planning and iteration times through an A algorithm, a Dijkstra algorithm and a BFS algorithm (pottery philosophy, high jump, Zhengtianjiang, and the like; path planning research in a honeycomb grid map based on the A algorithm, the university of China and North, the Nature edition, 2020(4): 310-. The contents of the above studies have a certain reference significance to the present text. However, they focus more on the path planning itself, rather than on the mesh model, again with certain limitations.
Disclosure of Invention
In order to solve the technical problems, the invention aims to provide a regular hexagon grid model map construction method for path planning of an intelligent robot.
Based on the above purpose, the invention provides a regular hexagon mesh model map construction method for intelligent robot path planning, which comprises the following steps:
dividing a plane area in a visual range into a plurality of seamless-connected non-overlapping regular hexagonal blank basic grids according to the size of a preset grid and map data;
acquiring point cloud data of each blank basic grid, and endowing each blank basic grid with topographic features, grid element attributes and grid edge attributes by combining map data to obtain a basic grid so as to form a grid model covering the plane area, wherein the topographic features are used for describing fluctuation changes of a space environment; the cell attributes are used for describing the geographical elements which are dominant in the spatial environment of the basic grid; the grid edge attribute is used for describing geographic elements at the boundary of the space environment where the basic grid is located;
determining the coordinates of the robot and the laser point in the grid model, and calculating the probability estimation value of the basic grid where the laser point is located, so as to determine that the basic grid is in an occupied state or an idle state;
updating the states of other basic grids between the basic grid where the robot is located and the basic grid where the laser point is located;
and carrying out incremental mapping on the grid model of the environment according to the state of each basic grid.
Preferably, the terrain features describe the terrain relief form within the occupation range of each basic grid mesh by using terrain feature factors, wherein the terrain feature factors comprise micro factors and macro factors, the micro factors comprise slopes, plane curvatures and section curvatures, and the micro factors comprise average slopes, surface cutting depths, terrain relief degrees and elevation variation coefficients.
Preferably, the cell attributes include a dot element, a planar element and a partial linear element in the geographic environment, and the cell edge attributes include a linear element located at a boundary of the basic cell in the geographic environment.
Preferably, the cell attributes of the basic grid include superposition of a position-occupied cell and an indication cell, wherein the position-occupied cell is a space element occupying the largest area or occupying the dominant position or occupying the important position of the current basic grid; the indicative cells are spatial elements that occupy extreme importance within the current base grid area.
Preferably, the edge attribute of the basic grid is the superposition of an obstacle type edge and an occupation type edge, the obstacle type edge is used for describing the on and off relations between two adjacent basic grids, and the occupation type edge exists depending on the obstacle type edge and is used for representing additional information on the edge.
Preferably, the specific method for subdividing the planar area in the visual range into a plurality of seamlessly connected and non-overlapping regular hexagonal blank basic grids according to the preset grid size and map data comprises the following steps:
determining a grid starting point and a grid orientation;
determining coordinate values of each point and each edge of the grid, wherein the edge is identified as A, B, C, D, E, F, and the points are identified as 1, 2, 3, 4, 5 and 6;
and generating all point coordinates of the whole blank grid model by using the grid size and the initial point coordinates according to a grid generation algorithm, thereby realizing grid generation of the blank grid model of the whole plane area.
Preferably, the specific method for determining the coordinates of the robot and the laser point in the grid model is as follows:
and determining the positions of the robot and the laser point in the plane rectangular coordinate system, and determining the coordinates of the robot and the laser point in the grid model by adopting a grid direct coding method according to the mutual conversion relation between the grid codes and the plane rectangular coordinate in the blank basic grid model.
Preferably, the probability estimation value of the basic grid where the laser point is located is calculated by using a hidden Markov model.
Preferably, the Bresenham algorithm is used for updating the states of other basic grids between the basic grid where the robot is located and the basic grid where the laser points are located.
Preferably, the incremental mapping of the grid model of the environment according to the states of the basic grids specifically includes: and performing incremental mapping on the grid model of the environment by using a map matching algorithm.
Compared with the prior art, the invention has the following beneficial effects:
firstly, the grid structure is utilized to expand multiple expressions of spatial environment attributes, a grid model is perfected from three aspects of topographic features, grid element attributes and grid edge attributes, the grid model is changed from a subdivision algorithm to a novel data structure for describing the spatial environment, a map model facing robot cognition is innovatively provided, and the method can be well applied to the map construction and path planning process of an intelligent robot.
And secondly, a more advantageous regular hexagonal grid model is used for replacing an occupation grid map model, so that the description of the space environment is more reasonable.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this application, illustrate embodiments of the application and, together with the description, serve to explain the application and are not intended to limit the application.
FIG. 1 is a schematic diagram illustrating center-to-center spacing analysis of a grid in an embodiment of the present invention;
FIGS. 2(a) - (b) are frequency spectrums of continuous band-limited signals after passing through a rectangular lattice sampling grid according to an embodiment of the present invention;
FIGS. 3(a) - (b) are schematic diagrams of frequency spectrum and spatial sampling direction vectors of continuous signals after passing through regular hexagonal sampling grids in the embodiment of the present invention
FIG. 4 is a diagram illustrating exemplary hexagonal grid attributes in accordance with an embodiment of the present invention;
FIG. 5 is a schematic diagram of "forest" cells in an embodiment of the present invention;
FIG. 6 is a schematic diagram of a "city" cell in an embodiment of the invention;
FIG. 7 is a schematic diagram of a "railway" cell in an embodiment of the present invention;
FIG. 8 is a schematic view of a "river" cell in an embodiment of the invention;
FIG. 9 is a flow chart of a lidar-based regular hexagonal mesh model construction in an embodiment of the present invention;
FIG. 10 is a schematic diagram of grid size, starting point and orientation in an embodiment of the invention;
FIG. 11 is a schematic diagram of a direct coding scheme of a regular hexagonal mesh model according to an embodiment of the present invention;
FIG. 12 is a schematic diagram illustrating an equally spaced rectangular coordinate system according to an embodiment of the present invention;
FIG. 13 is a hidden Markov model in an embodiment of the invention;
fig. 14 is a flowchart of a regular hexagonal mesh model mapping method for intelligent robot path planning in the embodiment of the present invention.
Detailed Description
The invention is further described with reference to the following figures and examples.
It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the disclosure. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments according to the present application. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
The embodiment provides a regular hexagon mesh model map building method for intelligent robot path planning, as shown in fig. 14, the method includes:
dividing a plane area in a visual range into a plurality of seamless-connected non-overlapping regular hexagonal blank basic grids according to the size of a preset grid and map data;
acquiring point cloud data of each blank basic grid, and endowing each blank basic grid with topographic features, grid element attributes and grid edge attributes by combining map data to obtain a basic grid so as to form a grid model covering the plane area, wherein the topographic features are used for describing fluctuation changes of a space environment; the cell attributes are used for describing the geographical elements which are dominant in the spatial environment of the basic grid; the grid edge attribute is used for describing geographic elements at the boundary of the space environment where the basic grid is located;
determining the coordinates of the robot and the laser point in the grid model, and calculating the probability estimation value of the basic grid where the laser point is located, so as to determine that the basic grid is in an occupied state or an idle state;
updating the states of other basic grids between the basic grid where the robot is located and the basic grid where the laser point is located;
and carrying out incremental mapping on the grid model of the environment according to the state of each basic grid.
As a preferred embodiment, the terrain features describe the terrain relief form in each of the occupation ranges of the basic grids by using terrain feature factors, wherein the terrain feature factors include micro factors and macro factors, the micro factors include gradient, slope direction, plane curvature, section curvature and the like, and the micro factors include average gradient, surface cutting depth, terrain relief degree, elevation variation coefficient and the like.
As a preferred embodiment, the cell attributes include a point element, a planar element and a partial linear element in the geographic environment, and the cell edge attributes include a linear element located at a boundary of a basic cell in the geographic environment.
As a preferred embodiment, the cell attributes of the basic grid include a superposition of a occupation cell and an indication cell, wherein the occupation cell is a space element occupying the largest area or the dominant position of the current cell, or occupying an important position, such as a water system, a residential area, vegetation and the like (G in fig. 4); the indicative cells are space elements, such as railways, highways, etc., that occupy extreme importance within the area of the current cell.
Specifically, a placeholder cell refers to a cell that occupies a dominant element in the area of the grid, i.e., the base grid occupies a spatial region within which the largest, or dominant, or significant position-occupying element serves as the cell attribute of the grid. The occupation type cells inevitably highlight the important elements, neglect the secondary elements, and further generalize the expression space environment. For example, the "forest" cells (FIG. 5) and the "city" cells (FIG. 6).
An indicator type cell is a special cell, which is itself a type of spatial element that is insufficient to constitute a placeholder type cell, but which needs to be differentiated into a placeholder type cell in a specific application due to its extreme importance. To distinguish from placeholder cells, such cells are named indicator cells herein. For example, a traffic element having an important influence on robot motion, simulation model deduction, although it appears as a linear element, it is given a lattice attribute such as "railway" (fig. 7); even if a certain mesh has been given an occupancy attribute, in order to highlight traffic information, the traffic information may be superimposed on the original attribute, which is defined as an indicative type cell with an obvious traffic feature, for example, "main roads in a city".
As a preferred embodiment, the edge attribute of the base grid is a superposition of an obstacle type edge and a position-occupied edge, and the obstacle type is used for describing an "on" and "off" relationship between two adjacent grids, for example, the edge attribute is a non-traversable river, which indicates that the current grid cannot cross the edge to reach the adjacent grid; the occupancy-type lattice edge exists in a manner to adhere to the barrier-type lattice edge, and represents additional information on the lattice edge, for example, a river (fig. 8) is a main source of the barrier-type lattice edge as a main element for obstructing traffic. However, in the case of a river, attachments such as bridges existing above the river exist depending on the river, and the river is not an obstacle-type lattice edge due to the bridge. Such as A, B, C, D, E and F shown in fig. 4. The attribute information of the grid model mainly comes from map data, and the terrain features, the grid element attributes and the grid edge attributes can be used as a data basis for real-time modeling of the intelligent robot, so that the intelligent robot can better serve path planning application.
For spatial environment description, a topographic feature describes fluctuation change of a spatial environment, a cell attribute describes a point element and a plane element in the spatial environment, and a partial line element, a cell edge attribute describes a line element in the spatial environment, which jointly determine a basic frame of a conceptual model of the spatial environment, and a large number of basic grids which are seamless, continuous and non-overlapping, and accompanying attribute information jointly form a basic appearance of the spatial environment. In path planning for an intelligent robot or a simulation model, a terrain feature, a cell attribute and a cell edge attribute play different roles respectively. Intelligent robots or simulation models are affected by terrain slopes, and as the complexity of the terrain increases, they pay more money through the grid, thereby affecting the farthest points they can reach. The cell attributes determine the different costs that they need to consume when passing cells of a particular attribute, similar to grids of different soil qualities, hard soil cells can always run longer distances at a lower cost than soft soil cells. The grid edge attributes determine whether the grid edges can cross the grid to reach the adjacent grid, and the grid edge attributes are similar to those of rivers, steep ridges, gullies and the like, have certain blocking effects on the motion of the robot and the simulation model deduction, and only when the performance of the robot and the simulation model exceeds the limit of the rivers, the steep ridges or the gullies, the blocking effects cannot be generated on the robot and the simulation model, otherwise, the motion of the robot and the simulation model can be blocked.
The terrain features, cell attributes and cell edge attributes of the mesh model are all embodied in the path planning application through cost values.
As a preferred embodiment, the specific method for subdividing the planar area in the visual range into a plurality of seamlessly connected and non-overlapping regular hexagonal blank basic grids according to the preset grid size and map data includes:
determining a grid starting point and a grid orientation;
determining coordinate values of each point and each edge of the grid, wherein the edge is identified as A, B, C, D, E, F, and the points are identified as 1, 2, 3, 4, 5 and 6;
and generating all point coordinates of the whole blank grid model by using the grid size and the initial point coordinates according to a grid generation algorithm, thereby realizing the grid generation of the blank grid model of the whole plane area.
As a preferred embodiment, the specific method for determining the coordinates of the robot and the laser point in the mesh model is as follows:
and determining the positions of the robot and the laser point in the plane rectangular coordinate system, and determining the coordinates of the robot and the laser point in the grid model by adopting a grid direct coding method according to the mutual conversion relation between the grid codes and the plane rectangular coordinate in the blank basic grid model.
As a preferred embodiment, a hidden markov model is used to calculate the probability estimation value of the base grid where the laser spot is located.
As a better implementation mode, the Bresenham algorithm is used for updating the states of other basic grids between the basic grid where the robot is located and the basic grid where the laser points are located.
As a preferred embodiment, the incremental mapping of the grid model of the environment according to the states of the basic grids specifically includes: and performing incremental mapping on the grid model of the environment by using a map matching algorithm.
In the practical application process, map data and laser sensors are two main sources of modeling of a regular hexagonal grid model facing an intelligent robot. The map data is used as a main mode for describing the space environment, and a basic grid is provided for the regular hexagonal grid model. The laser sensor can provide point cloud data with strong current situation for the regular hexagon grid model, and is suitable for real-time modeling and rapid updating of the grid model. In practical significance, when the laser sensor is used for building the regular hexagon grid model, the laser sensor can be fused to obtain relative point cloud data to realize real-time modeling and quick updating of the grid model according to the basic grid built by the map data.
The grid model is a grid unit which divides a region into a series of regular polygons according to a certain rule, and describes spatial position, element attribute and spatial distribution by utilizing the grid unit to realize organization, management, modeling, analysis and expression of spatial data, and the following differences of the regular quadrilateral model and the regular hexagonal model in the aspects of environmental description, path planning and the like prove the rationality and feasibility of the application of the regular hexagonal grid model to the robot map:
the geometric structure refers to a series of regular polygon mesh units which are covered on a plane or a spherical surface, continuously, seamlessly and non-overlapping, divide a space environment, determine a space position and describe space attributes. Considering the particularity of the meshes covering the study area, these regular polygon meshes should satisfy at least two conditions: one is that the distance from the center point of the mesh to each edge is the same in all directions. Ideally, the distances from the center point of the grid to the 360 ° directional side are all equal, i.e., circles. However, circles cannot be continuously spliced on a region, and therefore, an inscribed regular polygon of the circle needs to be selected to perform approximation processing instead of the circle. The second is that the grid must be a pattern that can be stitched continuously and without overlap over the area.
The geometry of a suitable mesh model can be determined from the two conditions described above.
Assuming a positive n-polygon is present, each internal angle Δ Α is:
if the pattern can form continuous, seamless and non-overlapping splicing on a plane, m regular n-polygons are required to be spliced at one point by one internal angle respectively, and the plane is exactly and completely covered. Therefore, the temperature of the molten metal is controlled,
the following can be obtained from formula (1) and formula (2):
wherein m and n are integers, and n can only take 3, 4 and 6. Therefore, there are only three types of regular polygons conforming to the geometry of the mesh model, namely regular trilateral, regular quadrilateral and regular hexagon. In theory, all three regular polygons can be used as the geometry of the mesh model.
(1) Grid center distance analysis
The grid center distance refers to the distance from the grid center point to the grid edge. The geometry of the mesh model is usually approximated by selecting a regular polygon inscribed in a circle. Obviously, the distance from the center point of the mesh to the edge of the mesh will vary unequally. Wherein, the vertical distance from the center point of the mesh to one side of the regular polygon is the nearest center distance (D)min) It is also called equidistant direction; the distance from the center point of the mesh to the vertex of the regular polygon is the farthest center distance (D)max). The ratio of the nearest center-to-center distance to the farthest center-to-center distance is referred to as the grid center-to-center distance error (FIG. 1). In general, a near-ideal mesh should have more equidistant directions and less center-to-center error of the mesh.
And respectively carrying out grid center distance analysis on the regular triangle, the regular quadrangle and the regular hexagon to obtain analysis results shown in the table 1.
TABLE 1 error table of center-to-center distances of grids
Categories | Equidistance direction (/ one) | Error of the measurement |
|
3 | 2.000 |
|
4 | 1.414 |
|
5 | 1.150 |
It can be found that: the equidistant direction of the regular hexagon is the most when viewed from the equidistant direction. The more equidistant directions indicate that the more directions can be selected from the center point of the grid, so that more reasonable experimental results can be obtained. And secondly, from the view of the error of the center distance of the grid, the error of the regular quadrangle and the regular hexagon is smaller than that of the regular triangle, and the smaller the error is, the more accurate approximation result can be obtained. Therefore, the experimental requirements can be better met by both the regular quadrangle and the regular hexagon, and the advantages of the regular hexagon are more prominent.
(2) Grid sampling density analysis
Overlaying the appropriate grid over the study area and then determining the attribute information for each grid is similar to creating a "digital image," referred to as a "digital image" sample. The grid density within a certain range is called the grid sampling density.
It is assumed that the result of fourier transform of two-dimensional continuous image information is shown by a hatched portion in fig. 2(a), i.e., the frequency band of the signal is within a circular region of a radius W in the frequency domain. The spectrum of the signal after the continuous band-limited image passes through a regular quadrilateral sampling grid is shown in FIG. 2(b), and the sampling interval in the horizontal direction is T1Sampling interval in vertical direction of T2The sampling interval satisfies the Nyquist sampling theory, i.e.
The arrangement of the frequency spectrum of the sampled signal in the frequency domain space is determined by the sampling matrix. The sampling matrix of the regular quadrilateral mesh is a diagonal matrix V, i.e.:
as can be seen from fig. 2(a), the space area occupied by the regular hexagon circumscribed by the circle is smaller than that occupied by the regular quadrangle. After the continuous image signals pass through a regular hexagonal sampling grid satisfying the Nyquist sampling theorem, the frequency spectrums of the formed signals are as shown in fig. 3(a), and it can be seen that the frequency spectrums of the sampled signals are much more compact in arrangement manner of frequency domain space than the frequency spectrums of the signals sampled by the regular quadrilateral sampling grid. The spatial sampling interval is defined by 2 direction vectors v1、v2It is decided that, as shown in FIG. 3(b), the sampling matrix V is composed of 2 direction vectors V1、v2The composition is as follows:
the sampling grid of FIG. 3 is a regular hexagon, and thus existsThe relationship (2) of (c). Since the frequency spectrum of the continuous signal lies within a circular area of radius W, in order to satisfy the Nyquist sampling theorem, i.e.The sampling matrix of the regular hexagon is:
the sampling density is the reciprocal of the determinant of the sampling matrix, so the comparison of the formula (4) and the formula (6) can prove that the minimum sampling density of the regular hexagonal structure is reduced by 13.4 percent compared with the minimum sampling density of the regular quadrilateral structure, and therefore, under the theoretical condition that the grids have the same sampling density, the approximation quality of the regular hexagonal grid is always superior to that of the regular triangular or regular quadrilateral grid.
The basic process of building a regular hexagonal grid model based on laser sensors is shown in fig. 9, in combination with a base grid built from map data, the relative position coordinates of the robot, and the planar rectangular coordinates of the laser spot. Firstly, constructing a basic grid model according to map data and set grid dimensions, wherein the basic grid model comprises the steps of establishing a geometric structure of the grid model, and endowing topographic features, grid element attributes and grid edge attributes of the grid model according to map elements, and the process is called grid modeling based on the map data; however, it is not intended to describe fully how to populate a mesh model with map data, and therefore only a brief description of the geometric modeling of the mesh is given, i.e. the underlying mesh model generated at this step is a blank mesh model. Secondly, determining the coordinates of the robot and the laser point in a grid model according to the coordinates of the robot and the laser point, calculating a probability estimation value of a grid where the laser point is located by applying a hidden Markov model, and determining an occupation state or an idle state of the grid; thirdly, updating the states of other grids between the two grids by using a Bresenham algorithm; and finally, incremental map building of the environment grid model is realized by applying a map matching algorithm.
Specifically, geometric modeling of a mesh model requires explicit mesh size, mesh starting point, mesh orientation, and the like. As shown in fig. 10, the mesh size refers to a distance H between opposite sides of a regular hexagonal mesh, a starting point of the mesh is a lower left corner point, and the mesh faces north of the edge;
each regular hexagon grid of the grid model can be regarded as an independent unit, uniformly coded attribute information is stored, and interaction with the robot or the simulation model independently occurs. Each grid comprises two parts, namely grid cells and grid edges. Thus, the geometric modeling of the mesh model essentially determines the coordinate values of each edge at each point of the mesh, where the edge is identified as A, B, C, D, E, F and the points are identified as 1, 2, 3, 4, 5, 6, as shown in FIG. 4.
Assuming center of lower left mesh of mesh modelThe coordinate of the center point is O (X)0,Y0) And the grid size is H, then the coordinates of the center point of the grid in the first row and the second column in the grid can be calculated by equation (7), and the coordinates of each point of the grid can be calculated by equation (8).
And generating all point coordinates of the whole grid model by utilizing information such as grid size, initial point coordinates and the like according to a grid generation algorithm, thereby realizing grid generation of the grid model of the whole plane area.
In the interaction process of the robot or the simulation model and the grid model, the interconversion between the grid code and the plane rectangular coordinate is usually required, as shown in fig. 11, so that the rapid and effective spatial operation of the robot or the simulation model is realized. Therefore, network coding needs to satisfy both the coordinate transformation and the spatial operation. More typical trellis coding mechanisms include Generalized Balanced Triplets (GBT) (Gibson et al, 1982; Sahr et al, 2003), "membership graph" structures (cardia, 2005; Zhang Yongsheng et al, 2007; child Xiaochong, 2010), PYXIS structures (Perry, 2006; Vince et al, 2009; YONG et al, 2003), and the like. The grid direct coding method effectively meets the requirements (Figure 11) in the two aspects, the two-dimensional array is used for managing the attribute information of the grid, and the direct coding method can save nearly half of the computer storage space, which has very important significance for a grid model with a large area range.
The conversion formula from plane rectangular coordinates to grid codes is shown in formula (9).
The transformation formula from grid coding to plane rectangular coordinates is shown in equation (10).
Wherein, Ihex、JhexRespectively, the horizontal and vertical codes of the grid, and X, Y the horizontal and vertical coordinates of the rectangular coordinate system. (X)min,Ymin) Is the lower left corner of the study area in the rectangular coordinate system, (X)max,Ymax) Is the upper right corner of the research area in the rectangular coordinate system, O is the origin of the rectangular coordinate system, XdisIs a distance in the X direction, YdisIs the separation distance in the Y direction (see fig. 12).
The regular hexagon mesh model building diagram facing the intelligent robot essentially belongs to the simultaneous positioning and mapping process of the robot, and generally speaking, the probability method is used for representing the possibility of the state of the robot at a certain moment. Suppose that the state information of the robot at that time is xtControl information is utThe sensor measurement information is ztThen the robotic system mainly solves two probabilities: probability of state transition p (x)t|xt-1,u1:t) And the measurement probability p (z)t|xt)。
The hidden markov model describes the basic process of robot movement and measurement (fig. 13), which shows the state of the robot at a moment in time only with the state x at the previous moment in timet-1And control information u of the current timetAbout measurement information z at time t of the robottOnly the status information x associated with the current timetIt is relevant.
The odometer detects control quantities such as linear speed, angular speed and the like of the robot in real time, and simultaneously determines pose information of the robot at the current moment according to state information at the previous moment; the laser sensor scans the surrounding environment at a specific frequency to acquire measurement information of the environment, and the measurement information is used for constructing a grid model.
Assuming that the pose of the robot is known, the construction of the grid model solves the problem of how to determine the noise and the uncertaintyQualitative measurement data creates a consistency map. The mesh model represents the map using a series of binary random variables whose values represent whether the current position is free or occupied by an obstacle. Therefore, the map construction process is to calculate the posterior probability p (m | x) of the whole map according to the given pose information and the measurement information1:t,z1:t) The process of (1).
The mesh model discretizes the two-dimensional space into individual cells of equal size, each cell representing the area encompassed by the grid. Since each cell state in the mesh model is either occupied or idle, p (m) can be usedi) Representing the probability of a cell being occupied by an obstacle, p (m)i) 1 indicates that the unit is in an occupied state, p (m)i) 0 means that it is in an idle state. Assuming that each cell in the grid map is independent of each other, the posterior probability p (m | x) of the global map m1:t,z1:t) Comprises the following steps:
wherein x is1:tAs a sequence of states (poses) of the robot, z1:tIs a measurement sequence of the robot.
For the posterior probability of any independent unit, the Bayesian criterion can be used to obtain:
when x is known from the hidden Markov modeltWhen known, ztAndx1:t-1and z1:t-1Is irrelevant, therefore
p(zt|mi,x1:t,z1:t-1)=p(zt|mi,xt),p(zt|x1:t,z1:t-1)=p(zt|xt)。p(zt|mi,xt) Simplified by Bayesian criterion as follows:
if the information z is measuredtWhen unknown, then the robot state x at time ttDoes not contain independent area m in grid mapiInformation of (2), i.e. xtTo miNo information is provided, p (m)i|xt)=p(mi),
p(mi|x1:t,z1:t-1)=p(mi|x1:t-1,z1:t-1). Rearranging p (m)i|x1:t,z1:t) The following can be obtained:
the state of each cell is only occupied and idle, so that the probability of the corresponding independent cell being idleExpressed as:
approximately eliminating the irrelevant terms in equations (15) and (16), then:
taking the logarithm of equation (17) yields:
if the log probability representing the posterior probability is used, equation (16) is expressed as:
wherein lt-1,iLogarithmic geometry of the posterior probability of the robot at time t-1, l0,iAs an initial value before sensor measurement, equation (19) satisfies a structure of recursively calculating the unit occupation probability. That is to say, the pose state and the observation information of the robot at the moment t can realize the update of the occupation probability of any unit on the basis of the existing grid model probability at the previous moment.
Thus, the construction algorithm 1 of the mesh model is shown.
And in the process of updating the grid model, the grid through which the connecting line passes is determined according to the connecting line of the grid where the robot is located and any grid in a determined state, and the state of the corresponding grid is estimated. For the regular quadrilateral mesh model, the process is updated by using Bresenham algorithm, and for the regular hexagonal mesh model, a specific processing algorithm is needed.
Suppose the position of the robot is (x)R,yR) The coordinates returned by the sensor scan are (x)S,yS) According to the regular hexagon grid model, the grid serial number of the coordinate scanned and returned by the sensor is (i)S,jS) Then the probability value that it is occupied by an obstacle is determined by algorithm 1. When the grid (i)S,jS) After the probability value occupied by the obstacle is determined, the occupied (or idle) state of other grids passed by the scanning line can be determined according to the position of the robot, the coordinates returned by the sensor scanning and the grid serial number. This process is shown in algorithm 2.
Probability estimation and map updating solve the problem of building a robot map when the robot is at a certain fixed position. Along with the continuous movement of the robot, the probability values of all grids are continuously updated, so that different grid models with the same resolution need to be continuously spliced in an incremental mapping mode until a complete grid model is formed.
The computer stores each grid of the grid model in a matrix array, each matrix element corresponding to a grid, such that the complete environmental grid model corresponds to a pixel of the matrix array. Suppose that the robot constructs a mesh model Map at different timesGAnd mesh model MapWThere are overlapping regions, the problem then is how to stitch to achieve incremental mapping of the environmental mesh model. Firstly, the edge extraction algorithm is utilized to respectively extract the mesh models MapGAnd mesh model MapWExtracting edge pixels to form a set of edge pixel points, i.e.Andwherein g isiAnd wiIs a vector of NgAnd NwThe numbers of elements of the edge pixel point sets G and W are respectively indicated. Mesh model MapGAnd mesh model MapWG for the edge pixel point set of the overlap regionεIt is expressed as a subset of G and W, and the percentage of overlap is expressed as ε. Then, the net at different timeConverting the incremental mapping problem of the grid model into an image registration problem, calculating rigid body transformation T of the robot to be { R, T }, so that the point set T (G) after transformation can be well matched with the point set W, and further expressing the incremental mapping problem of the grid model as a minimization problem, namely:
wherein for the control parameter, | · | represents the minimum overlap percentage of the number of elements in the set, and ε is usedminAnd (4) showing. And the grid model of the robot is built in a gradual incremental manner by continuously updating and matching the grid model, and finally, a complete grid model is formed.
According to the invention, by expanding the attribute information of the grid model, the grid model is perfected from three aspects of topographic features, grid element attributes and grid edge attributes, so that the grid model is converted into a novel data structure for describing a space environment from a subdivision algorithm, and the data structure can be well applied to map construction and path planning application of an intelligent robot.
Although the embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and not to be construed as limiting the present invention, and those skilled in the art can make changes, modifications, substitutions and alterations to the above embodiments without departing from the principle and spirit of the present invention, and any simple modification, equivalent change and modification made to the above embodiments according to the technical spirit of the present invention still fall within the technical scope of the present invention.
Claims (10)
1. A regular hexagon mesh model map construction method for intelligent robot path planning is characterized by comprising the following steps:
dividing a plane area in a visual range into a plurality of seamless-connected non-overlapping regular hexagonal blank basic grids according to the size of a preset grid and map data;
acquiring point cloud data of each blank basic grid, and endowing each blank basic grid with topographic features, grid element attributes and grid edge attributes by combining map data to obtain a basic grid so as to form a grid model covering the plane area, wherein the topographic features are used for describing fluctuation changes of a space environment; the cell attributes are used for describing the geographical elements which are dominant in the spatial environment of the basic grid; the grid edge attribute is used for describing geographic elements at the boundary of the space environment where the basic grid is located;
determining coordinates of the robot and the laser points in the grid model, and calculating a probability estimation value of a basic grid where the laser points are located, so as to determine that the basic grid is in an occupied state or an idle state;
updating the states of other basic grids between the basic grid where the robot is located and the basic grid where the laser point is located;
and carrying out incremental mapping on the grid model of the environment according to the state of each basic grid.
2. The method of claim 1, wherein the terrain features describe the topography of each base mesh within its footprint using terrain characterization factors, the terrain characterization factors including micro factors including slope, plane curvature and section curvature and macro factors including mean slope, surface cut depth, terrain relief and elevation coefficient of variation.
3. The method as claimed in claim 1, wherein the cell attributes include point elements, planar elements and partial line elements in the geographic environment, and the cell edge attributes include line elements at the boundary of the basic cell in the geographic environment.
4. The method for constructing the regular hexagon grid model map for the intelligent robot path planning as claimed in claim 3, wherein the cell attributes of the base grid comprise the superposition of a position-occupying cell and an indication cell, wherein the position-occupying cell is a space element occupying the largest area or the dominant position or the important position of the current base grid; the indicative cells are spatial elements that occupy extreme importance within the current base grid area.
5. The method as claimed in claim 3, wherein the edge attributes of the basic grids are the superposition of barrier-type edges and position-occupied edges, the barrier-type edges are used for describing the on and off relations between two adjacent basic grids, and the position-occupied edges exist depending on the barrier-type edges and are used for representing additional information on the edges.
6. The method for constructing a regular hexagonal grid model map for intelligent robot path planning as claimed in claim 1, wherein the specific method for subdividing the planar area in the visual range into a plurality of seamlessly connected non-overlapping regular hexagonal blank basic grids according to the preset grid size in combination with map data is as follows:
determining a grid starting point and a grid orientation;
determining coordinate values of each point and each edge of the grid, wherein the edge is identified asA、B、C、D、E、FPoint labels are 1, 2, 3, 4, 5, 6;
and generating all point coordinates of the whole blank grid model by using the grid size and the initial point coordinates according to a grid generation algorithm, thereby realizing grid generation of the blank grid model of the whole plane area.
7. The method for constructing a regular hexagonal grid model map for path planning of an intelligent robot according to claim 1, wherein the specific method for determining the coordinates of the robot and the laser point in the grid model is as follows:
and determining the positions of the robot and the laser point in the plane rectangular coordinate system, and determining the coordinates of the robot and the laser point in the grid model by adopting a grid direct coding method according to the mutual conversion relation between the grid codes and the plane rectangular coordinates in the blank basic grid model.
8. The method for constructing a regular hexagonal grid model map for intelligent robot path planning as claimed in claim 1, wherein a hidden markov model is used to calculate the probability estimation value of the basic grid on which the laser spot is located.
9. The method for constructing a regular hexagonal grid model map for intelligent robot path planning according to claim 1, wherein the Bresenham algorithm is used to update the states of other basic grids between the basic grid where the robot is located and the basic grid where the laser points are located.
10. The method for constructing a regular hexagonal grid model map for path planning of an intelligent robot according to claim 1, wherein the incremental mapping of the grid model of the environment according to the states of the basic grids specifically comprises: and performing incremental mapping on the grid model of the environment by using a map matching algorithm.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210242363.0A CN114659530A (en) | 2022-03-11 | 2022-03-11 | Grid model map construction method for intelligent robot path planning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210242363.0A CN114659530A (en) | 2022-03-11 | 2022-03-11 | Grid model map construction method for intelligent robot path planning |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114659530A true CN114659530A (en) | 2022-06-24 |
Family
ID=82030017
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210242363.0A Pending CN114659530A (en) | 2022-03-11 | 2022-03-11 | Grid model map construction method for intelligent robot path planning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114659530A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20210354690A1 (en) * | 2020-05-12 | 2021-11-18 | Motional Ad Llc | Vehicle operation using a dynamic occupancy grid |
CN115587460A (en) * | 2022-11-24 | 2023-01-10 | 河北纬坤电子科技有限公司 | Digital simulation method and device for road traffic condition |
CN116734879A (en) * | 2023-06-13 | 2023-09-12 | 南京信息工程大学 | Time optimal path planning method based on hexagonal grid |
-
2022
- 2022-03-11 CN CN202210242363.0A patent/CN114659530A/en active Pending
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20210354690A1 (en) * | 2020-05-12 | 2021-11-18 | Motional Ad Llc | Vehicle operation using a dynamic occupancy grid |
US11814039B2 (en) * | 2020-05-12 | 2023-11-14 | Motional Ad Llc | Vehicle operation using a dynamic occupancy grid |
CN115587460A (en) * | 2022-11-24 | 2023-01-10 | 河北纬坤电子科技有限公司 | Digital simulation method and device for road traffic condition |
CN116734879A (en) * | 2023-06-13 | 2023-09-12 | 南京信息工程大学 | Time optimal path planning method based on hexagonal grid |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Xia et al. | Geometric primitives in LiDAR point clouds: A review | |
CN114659530A (en) | Grid model map construction method for intelligent robot path planning | |
WO2020253171A1 (en) | Terrain modeling method and system amalgamating geometric characteristics and mechanical characteristics | |
CN108805350B (en) | Search and rescue range prediction method based on multi-dimensional Monte Carlo theory | |
CN109256028B (en) | Method for automatically generating unmanned high-precision road network | |
CN106199558A (en) | Barrier method for quick | |
CN112184736B (en) | Multi-plane extraction method based on European clustering | |
CN115564926B (en) | Three-dimensional patch model construction method based on image building structure learning | |
CN116843845A (en) | Spatial data integration method and system applied to digital twin city | |
CN112348867B (en) | Urban high-precision three-dimensional terrain construction method and system based on LiDAR point cloud data | |
CN109101743B (en) | Method for constructing high-precision road network model | |
CN110889198A (en) | Multi-factor joint learning-based dead reckoning probability distribution prediction method and system | |
CN111476286A (en) | Map construction method for mobile robot | |
Lu et al. | Graph-embedded lane detection | |
CN111881919B (en) | Line element intelligent simplification method and device based on tracking type grid subdivision | |
Hardouin et al. | Surface-driven Next-Best-View planning for exploration of large-scale 3D environments | |
He | Automated 3D building modelling from airborne LiDAR data | |
CN116993750A (en) | Laser radar SLAM method based on multi-mode structure semantic features | |
CN115619953A (en) | Rugged terrain-oriented mobile robot terrain mapping method and system | |
CN116299313A (en) | Laser radar-based intelligent vehicle passable area detection method | |
Smith et al. | Autonomous passage planning for a polar vessel | |
CN118171498B (en) | Grid technology-based computable airspace modeling method | |
CN115435793B (en) | Robot path planning method and device, storage medium and electronic equipment | |
Carneiro et al. | Incorporation of morphological properties of buildings’ descriptors computed from GIS and LIDAR data on an urban multi-agent vector based geo-simulator | |
Ma et al. | An Accuracy Evaluation Method for Multi-source Data Based on Hexagonal Global Discrete Grids |
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 |