Robot path planning method based on topology awareness

Abstract

The application relates to the technical field of path planning, in particular to a robot path planning method based on topological perception, which comprises the following steps: constructing a grid and counting four-way shielding frequency, marking sparse transition dense levels according to a frequency threshold, identifying sparse grid clusters and calculating a boundary ratio, marking a traffic priority or constraint section, analyzing path section direction stability to screen continuous priority edges, calculating edge connection fluctuation to identify structure sensitive sections, and counting execution failure proportions to mark potential non-passable path sections. According to the application, the grid shielding frequency in the obstacle distribution map is quantitatively calculated, the consistency of the planned path in the global navigation direction is improved, the critical structure area is accurately identified by combining path section boundary topological connection density analysis, the adaptability of the path in the structure change area is effectively improved, the potential non-passable state of the path section is dynamically marked, and the response efficiency and execution stability of the path planning to complex environment changes are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of path planning, in particular to a robot path planning method based on topology perception. BACKGROUND

[0002] The technical field of path planning mainly studies how to generate an optimal or feasible path for a mobile body from a starting position to a target position in a known or unknown environment. This field covers techniques such as geometric path planning, sampling space planning, graph search algorithms, numerical optimization methods, and reinforcement learning path generation. The core goals include obstacle avoidance, path smoothness, computational efficiency, and path length optimization, and it is widely used in intelligent manufacturing, warehouse logistics, service robots, surgical navigation systems, and mobile platform autonomous navigation scenarios.

[0003] Among them, the robot path planning method is a technical solution for realizing autonomous motion control of mobile robots. Its purpose is to automatically calculate the path that the robot needs to take from the current position to the target position according to environmental information and task goals, ensuring that the path is spatially feasible and meets the requirements of obstacle avoidance, efficiency, and path constraints. The method can be applied to navigation robots, AGV cars, inspection robots, and other systems to realize indoor and outdoor autonomous navigation, task path planning and execution, and improve system intelligence and operational efficiency.

[0004] Traditional planning methods using graph search or sampling space methods have difficulty accurately reflecting the spatial passability state in obstacle-dense or locally complex regions, lack a hierarchical expression mechanism for local topological structure characteristics, and cause path passability misjudgment when traversing structure boundary ambiguous regions during path generation. In global path direction control, only the shortest distance or cost function is considered, and the direction continuity of path segments is ignored, resulting in direction jump phenomenon in the planned path, affecting the smoothness and execution efficiency of the actual motion of the robot. In addition, the processing granularity is rough in the utilization of path execution history data, and the path validity label is not dynamically adjusted according to the path call record, resulting in weak adaptability of the path planning to environmental changes in long-term operation. SUMMARY

[0005] The purpose of the present application is to solve the shortcomings in the prior art, and a robot path planning method based on topology perception is proposed.

[0006] In order to achieve the purpose, the present application adopts the following technical scheme: a robot path planning method based on topology perception, comprising the following steps:

[0007] S1: Obtain a two-dimensional obstacle distribution map in a robot working area, establish a fixed-size grid unit set in the area, calculate the four-direction average blocking frequency of each grid, and perform interval judgment on the frequency and a preset blocking frequency threshold to generate grid-level obstacle blocking level label information;

[0008] S2: Call the grid-level obstacle blocking level label information, calculate the boundary grid number of each cluster area, mark the cluster area as a passing priority section or a constraint section, and generate a topological map passable area initial mapping structure;

[0009] S3: Based on the topological map passable area initial mapping structure, obtain the line direction vector in each pair of path sections, extract the maximum deviation value and the median deviation value in the angle sequence and perform difference operation, select the path section satisfying the stable direction feature, and generate a direction stable path section mapping result;

[0010] S4: Call the direction stable path section mapping result, calculate the ratio difference value of the average value of the number of path section ends and the total number of neighborhood edges according to the path section start and end nodes, if the difference value is greater than the boundary fluctuation coefficient threshold, mark the corresponding section path as a structure critical section, and generate a structure sensitive path section identification result.

[0011] The grid-level obstacle blocking level label information includes sparse level region annotation, transition level region annotation, dense level region annotation, direction blocking statistical layer and grid number index, the topological map passable area initial mapping structure is specifically a passing priority region set, a constraint region set, a space connected cluster label, a boundary identification bitmap and a passable section space coordinate set, the direction stable path section mapping result includes a continuity priority path section set, a path section direction deviation rate label, a path section target alignment angle information, a path section direction stability score and a path section sequence number table, and the structure sensitive path section identification result specifically refers to a structure critical path section set, a path node connection edge fluctuation label, a three-order adjacent node relationship graph, a boundary jump section identification set and a topological overload flag.

[0012] The grid-level obstacle blocking level label information acquisition step specifically includes:

[0013] S111: Obtain a two-dimensional obstacle distribution map in a robot working area, establish a fixed-size grid unit set in the area, collect the obstacle blocking count value of each grid in four orthogonal directions, and perform average value calculation on the count value of each direction, establish an average blocking frequency data set based on the four-direction blocking of the grid, and generate grid average blocking frequency information;

[0014] S112: Based on the grid average occlusion frequency information, the average frequency value of each grid is called and interval judgment is performed with the set occlusion frequency threshold value, the section recognition operation of the frequency value interval is performed, the corresponding section is judged to be a low frequency section, a medium frequency section or a high frequency section, and a label attribute is set, and a frequency interval label list is generated;

[0015] S113: According to the frequency interval label list, each grid is classified and labeled according to the label attribute, the grid number index and the label corresponding relationship are called, the space labels of the three types of sparse, transition and dense labels are implemented, and the grid level obstacle occlusion level label information is obtained.

[0016] The application improves that the obtaining step of the topological graph passable region initial mapping structure is specifically:

[0017] S211: The grid level obstacle occlusion level label information is called, the four-adjacent grid connectivity identification is performed based on the spatial adjacency relationship of the sparse labeled grid according to the sparse level label unit, the grid groups in spatial continuous distribution are aggregated, the closed region set composed of each group of adjacent grids is extracted and identified, and the sparse region connectivity cluster number index is generated;

[0018] S212: The sparse region connectivity cluster number index is called, the boundary grid number and the core region grid point number of each cluster region are obtained according to the grid number information of the cluster region, the ratio of the boundary grid number to the core grid number is calculated, the structure compactness difference value is obtained by operation, and the structure boundary ratio difference sequence is generated;

[0019] S213: According to the structure boundary ratio difference sequence, the structure compactness ratio threshold value is judged, whether the structure compactness difference value is less than the structure compactness ratio threshold value is judged, if it is true, the corresponding cluster region is marked as a pass priority section, if it is not true, it is marked as a constraint section, the cluster region is classified and labeled, and the topological graph passable region initial mapping structure is obtained.

[0020] The application improves that the obtaining step of the direction stable path section mapping result is specifically:

[0021] S311: Based on the topological graph passable region initial mapping structure, the passable node set included in the pass priority section is obtained, the double node pairs are combined in the node set according to the spatial coordinate sequence, the start and end coordinate points of each path section are extracted and the path section center line vector is constructed, the direction angle calculation is performed combined with the current preset global motion direction vector of the robot, and the path section direction offset angle sequence is generated;

[0022] S312: call the path segment direction offset angle sequence, calculate the maximum value and the median value of the path segment direction offset angle, extract the difference value as the direction fluctuation intensity factor, collect the path length value and the angle change rate of the path segment, perform joint offset factor calculation, obtain the direction offset joint intensity value, compare the direction offset joint intensity value of each path segment with the direction offset stability threshold value, screen the path segments meeting the direction consistency, and obtain a direction consistency path segment sequence;

[0023] S313: according to the direction consistency path segment sequence, performing marking operation on the screened path segments in the original topology graph structure, assigning the path segments with the edge weight identification attribute of the continuity priority edge, and establishing a direction stable path segment mapping result.

[0024] The application improves that the obtaining step of the structure sensitive path segment recognition result is specifically:

[0025] S411: call the direction stable path segment mapping result, according to the start and end nodes of the path segment, count the number of topology connection edges of each node, record the arithmetic mean value of the number of edges of the two end nodes of the path segment, and establish the average edge number information of the two end nodes of the path segment;

[0026] S412: according to the average edge number information of the two end nodes of the path segment, obtain the third-order adjacent node set of the two end nodes of each path segment, and count the total number of connection edges of each node in the set, combine the network density background and the node distribution center of gravity index of the path segment in the topology graph, and obtain the structure boundary deviation intensity value, to obtain a structure boundary deviation intensity value sequence;

[0027] S413: call the structure boundary deviation intensity value sequence, judge whether the structure boundary deviation intensity value is greater than the set boundary fluctuation coefficient threshold value, if the condition is met, mark the corresponding path segment as a structure critical segment, and establish a structure sensitive path segment recognition result.

[0028] The application improves that the method further comprises the following steps:

[0029] S5: obtaining the path segment in the structure sensitive path segment recognition result, calculating the path call failure ratio, comparing the failure ratio with the passable state benchmark ratio, if higher than the set proportion threshold value, performing potential non-passable label calibration operation on the path segment, and generating dynamic passable topology path annotation information;

[0030] The dynamic passable topology path annotation information includes potential non-passable path segment index, path segment history failure comparison table, dynamic path label graph, passable state level label and path segment state change record.

[0031] The application improves that the obtaining step of the dynamic passable topology path annotation information is specifically:

[0032] S511: Obtain the structure-sensitive path segment identification result, read the call number of each path segment in the robot continuous path call record and the actual successful record number, difference calculate the call two data, corresponding processing with the path segment index, generate the path segment call failure number sequence;

[0033] S512: Based on the path segment call failure number sequence, according to the failure number and the total number of calls of the path segment, execute the failure frequency ratio calculation, compare with the set passable state benchmark ratio item by item, extract the path segment index higher than the threshold value and summarize, generate the path segment passable failure ratio value set;

[0034] S513: Call the path segment passable failure ratio value set, according to the path segment index meeting the threshold condition, label the potential non-passable attribute label of the path segment, and complete the state writing in the topological graph structure, establish dynamic passable topological path labeling information.

[0035] Compared with the prior art, the advantages and positive effects of the present application are:

[0036] In the present application, the grid occlusion frequency in the obstacle distribution map is quantitatively calculated, and the sparse, transition, dense levels are divided according to the interval attribution, the hierarchical expression of the passability of the working area is realized, the boundary compactness of the connected cluster is evaluated, the fine description of the path accessibility structure is enhanced, the direction stability of the path segment is identified by using the direction vector angle difference operation, the consistency of the planned path in the global navigation direction is improved, the critical structure area is accurately identified by combining the path segment boundary topological connection density analysis, the adaptability of the path in the structure change area is effectively improved, and the potential non-passable state of the path segment is dynamically labeled by introducing the path call frequency and the execution failure ratio linkage mechanism, the response efficiency and the execution stability of the path planning to the complex environment change are improved, and the environmental adaptability and the dynamic robustness of the overall path are strengthened. BRIEF DESCRIPTION OF DRAWINGS

[0037] Figure 1 The method flowchart of the present application;

[0038] Figure 2 The flowchart of the present application for obtaining grid-level obstacle occlusion level label information;

[0039] Figure 3 The flowchart of the present application for obtaining the initial mapping structure of the topological graph passable area;

[0040] Figure 4 The flowchart of the present application for obtaining the direction stable path segment mapping result;

[0041] Figure 5A flowchart for obtaining a structure-sensitive path segment identification result of the present application;

[0042] Figure 6 A flowchart for obtaining dynamic passable topology path labeling information of the present application. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical scheme and advantages of the present application clearer, the present application will be further described in detail below in combination with the drawings and examples. It should be understood that the specific examples described herein are only used to explain the present application and do not limit the present application.

[0044] In the description of the present application, it should be understood that the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and therefore cannot be understood as indicating or implying that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application. In addition, in the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise explicitly and specifically limited.

[0045] Please refer to Figure 1 The present application provides a technical scheme: a robot path planning method based on topology perception, comprising the following steps:

[0046] S1: obtaining a two-dimensional obstacle distribution map in a robot work area, establishing a fixed-size grid cell set in the area, collecting the number of obstacle projections in four orthogonal directions of the grid, calculating the average frequency of four-directional obstacle blocking of each grid, and performing interval judgment on the frequency and a preset blocking frequency threshold, calling interval attribution labels to generate three levels of grid label types of sparse, transition and dense, and generating grid-level obstacle blocking level label information;

[0047] The blocking frequency threshold can be defined as the average value of the number of obstacle projections in each direction in a unit area grid, which is usually obtained according to the laser scanning blocking count of each direction, and is used to distinguish the complexity level of the environment structure;

[0048] S2: calling the grid-level obstacle blocking level label information, identifying the connected cluster area of the corresponding type of grid in space according to the sparse level label unit, calculating the number of boundary grids for each cluster area, extracting the ratio of the number of boundary grids to the number of core area grid points, and judging the ratio with the compactness ratio threshold, marking the cluster area as a pass priority segment or a constraint segment according to the judgment result, and generating an initial mapping structure of the passable area of the topology map;

[0049] Compactness ratio threshold can refer to shape factor, defined as the ratio of boundary length to area, commonly used to identify whether the region is edge or concentrated, reference source such as compactness measurement method in urban space modeling;

[0050] S3: Based on the initial mapping structure of the passable region in the topology graph, according to the passable node set included in the passable segment, extract the path segment of double nodes, get the direction vector of each pair of path segment, calculate the included angle between the current robot global expected motion direction vector, extract the maximum deviation value and the median deviation value in the included angle sequence and perform difference operation, call the difference result and compare it with the direction deviation stability threshold value, filter the path segment that meets the stable direction feature, and mark the path segment in the original topology graph structure as continuous priority edge, and generate the direction stable path segment mapping result;

[0051] The center line direction vector is the unit vector between the start point and the end point of the path segment, and the included angle can be calculated by inverse cosine; The direction deviation stability threshold value is an important parameter to control the smoothness of the navigation path, which can be determined by simulation debugging or according to the minimum turning angle path rule;

[0052] S4: Call the direction stable path segment mapping result, according to the path segment start and end node, count the number of topology connection edges of each node and the total number of adjacent three order nodes, calculate the ratio difference value of the average value of the number of edges at both ends of the path segment and the total number of adjacent edges, if the difference value is greater than the boundary fluctuation coefficient threshold value, mark the corresponding path segment as structure critical segment, and generate the structure sensitive path segment identification result;

[0053] The boundary fluctuation coefficient threshold value can be set as the standard deviation ratio of the number of connection edges, which is used to identify the edge segment and bottleneck segment in the topology graph. The related concept can be referred to the edge density change judgment criterion in graph neural network;

[0054] S5: Get the path segment in the structure sensitive path segment identification result, and read the difference value between the frequency of path segment in robot continuous path call record and the number of path execution success record, calculate the path call failure ratio, compare the failure ratio with the passable state benchmark ratio, if higher than the set proportion threshold value, execute the potential unpassable label marking operation on the path segment, and generate the dynamic passable topology path labeling information;

[0055] The passable state benchmark ratio refers to the experience threshold of path historical call failure ratio, which can be set in combination with the task failure rate record in the robot path planning system;

[0056] The grid-level obstacle blocking level label information includes sparse level region annotation, transition level region annotation, dense level region annotation, direction blocking statistical layer and grid number index, the topological graph passable region initial mapping structure specifically includes passable priority region set, constraint region set, space connected cluster label, boundary identification bitmap and passable segment space coordinate set, the direction stable path segment mapping result includes continuity priority path segment set, path segment direction offset rate label, path segment target alignment angle information, path segment direction stability score and path segment sequence number table, the structure sensitive path segment identification result specifically includes structure critical path segment set, path node connection edge fluctuation label, three-order adjacent node relationship graph, boundary jump segment identification set and topological overload flag, and the dynamic passable topological path label information includes potential impassable path segment index, path segment history failure comparison table, dynamic path label graph, passable state level label and path segment state change record.

[0057] Referring to Figure 2 , the acquisition step of the grid-level obstacle blocking level label information is specifically as follows:

[0058] S111: acquiring a two-dimensional obstacle distribution map in a robot working area, establishing a fixed grid cell set in the area, collecting obstacle blocking count values of each grid in four orthogonal directions, and performing average value calculation on the count values of each direction to establish an average blocking frequency data set of the grid in four directions, and generating grid average blocking frequency information;

[0059] When establishing a fixed grid cell set in the robot working area, first, the known space boundary coordinate point set is read from the environment map file to construct a grid system with a size of 1 meter x 1 meter, the unique number information of each cell is obtained, and a number matrix is generated in the two-dimensional plane, then the obstacle blocking count of each grid cell in the north, east, south and west directions is collected, the blocked times in the four directions are counted and recorded as an integer array, taking grid G1 as an example, if the north is 2 times, the east is 1 time, the south is 3 times, and the west is 0 times, then the direction blocking array is {2, 1, 3, 0}, then the average operation is performed on the four direction values of each grid to obtain the average blocking frequency of the grid, for example, the average frequency of G1 is (2+1+3+0) / 4=1.5, the average frequency data sequence of all grids is formed, and then a two-dimensional array is established as the average blocking frequency data set, and the result of this step is the grid average blocking frequency information;

[0060] Table 1: Grid direction blocking frequency calculation table

[0061]

[0062] As shown in Table 1, the table shows the number of occlusions in four directions of the partial grid and its average occlusion frequency, and the frequency value is used for subsequent interval identification and label determination.

[0063] S112: Based on the average occlusion frequency information of the grid, the average frequency value of each grid is called and interval judgment is performed with the set occlusion frequency threshold value. The segment recognition operation in the interval where the frequency value is located is executed, and the corresponding segment is judged to be a low frequency segment, a medium frequency segment or a high frequency segment, and a label attribute is set. A frequency interval label list is generated.

[0064] Based on the average occlusion frequency information of the grid, the average frequency value of each grid is called and interval judgment is performed with the set occlusion frequency threshold value. The occlusion frequency threshold value is two interval breakpoints, which are 1 and 2. The setting is based on the statistical boundary of the average value μ and the standard deviation σ of the occlusion frequency of the overall grid in the region. The first threshold value is set as μ-0.5σ, and the second threshold value is set as μ+0.5σ. When μ=1.5 and σ=0.5, the first threshold value is calculated as 1.25, and the second threshold value is calculated as 1.75. Therefore, 1 and 2 are respectively rounded and set as the interval limit values. The low frequency interval is set as [0, 1), the medium frequency interval is set as [1, 2), and the high frequency interval is set as [2, ∞). The average frequency value of each grid is brought into the set interval judgment formula to execute the segment attribution recognition operation. For example, if the average frequency value of G1 is 1.5, it is attributed to the medium frequency segment. If the average frequency value of a certain grid is 2.25, it is attributed to the high frequency segment. If the average frequency value is 0.8, it is attributed to the low frequency segment. Interval judgment is determined by inequality joint expression. Finally, each grid is attributed to its corresponding frequency label category and its label information is recorded to form a structured table composed of grid number and segment label. The result is a frequency interval label list.

[0065] S113: According to the frequency interval label list, each grid is classified and labeled according to the label attribute. The grid number index and the label corresponding relationship are called to implement the spatial marking of three types of labels: sparse, transition, and dense. The grid-level obstacle occlusion level label information is obtained.

[0066] According to the frequency interval label list, all grids are classified and processed according to their assigned label attributes. First, the grid with a low frequency label is classified as a sparse area grid. The medium frequency is classified as a transition area grid, and the high frequency is classified as a dense area grid. The numbers of different regions are marked in the grid matrix, and a mapping relationship table of labels and spatial positions is constructed based on the grid index information. The grid ID, its spatial coordinates and corresponding label are filled into the spatial structure array one by one. Finally, a complete sparse, transition, and dense three-class obstacle labeling layer information is established in the two-dimensional region map to obtain the grid-level obstacle occlusion level label information.

[0067] Please refer to Figure 3The acquisition step of the initial mapping structure of the topological graph passable region is specifically as follows:

[0068] S211: Call the grid-level obstacle occlusion level label information, according to the sparse level labeling unit, based on the spatial adjacency relationship of the sparse labeling grid, perform four-adjacent grid connectivity identification, aggregate processing on the spatial continuous distribution of grid groups, extract the aggregated region and identify the closed region set composed of each group of adjacent grids, and generate the sparse region connectivity cluster number index;

[0069] In the grid-level obstacle occlusion level label information, first extract all the cell data marked as sparse label, establish a spatial mapping table according to the two-dimensional coordinates, and then perform a four-adjacent judgment operation on each sparse grid, that is, detect whether there are other sparse cells in the upper, lower, left and right four directions, if there are, a connection relationship is established, and then all the grid cells connected with each other are aggregated according to the spatial connectivity, and the grid set that can be connected with each other is extracted to form an independent closed region, for example, if G1, G2 and G4 are spatially adjacent and marked as sparse regions, they are classified into a connected cluster C1. After the aggregation of all sparse regions is completed, the aggregated clusters are numbered and identified one by one, and the grid number set, spatial boundary range and other information corresponding to each cluster are recorded, and finally the sparse region connectivity cluster number index is obtained;

[0070] S212: Call the sparse region connectivity cluster number index, according to the grid number information of the cluster region, obtain the boundary grid number and the core grid point number of each cluster region, calculate the ratio of the boundary grid number to the core grid number, and use the formula:

[0071] ;

[0072] The operation obtains the structure compactness difference value, and generates a structure boundary ratio difference sequence;

[0073] Wherein, represents the structure compactness difference value, represents the boundary grid normalization value of the region, represents the core grid point normalization value of the region, represents the normalization value of the boundary direction change frequency in the region, is the growth adjustment term of the core density, and the denominator term is used to construct the structure expansion reference value, and the whole fraction is multiplied by to constitute the structure fluctuation suppression factor;

[0074] Call the grid number information of each cluster region in the sparse region connectivity cluster number index, obtain the corresponding boundary grid number and the core grid point number , and the frequency of boundary direction change After normalization, the formula for calculating the structural compactness difference is as follows:

[0075] ;

[0076] The formula consists of three parts: the numerator part The composite expansion index is used to comprehensively express the relative redundancy of the boundary and the aggregation degree of the core area. When the boundary is large and the core is small, the value increases. At the same time, the logarithmic term controls the increase of the core grid influence when the number increases. The denominator part As a standardization factor of the structural expansion scale, it makes the results of different scale clusters comparable, and multiplied by plays a role in suppressing structural disturbance, reducing the influence of areas with drastic direction changes. The calculation result is the dispersion intensity of the compactness of the cluster structure. The larger the value, the less compact the structure. Taking cluster C2 as an example, assuming that the number of boundary grids is 10, the number of core grid points is 30, and the frequency of direction change is 2, the formula is calculated as follows:

[0077] First, normalize:

[0078] ;

[0079] ;

[0080] ;

[0081] Then substitute the formula to calculate:

[0082] ;

[0083] Table 2: Calculation of structural compactness difference value

[0084]

[0085] As shown in Table 2, the structural compactness difference value of cluster C2 is the largest, indicating that the boundary and core structure are unbalanced and the spatial structure is divergent, and the compactness is the weakest.

[0086] S213: According to the sequence of structural boundary ratio difference, judge whether the structural compactness difference value is less than the structural compactness ratio threshold value, if it is, mark the corresponding cluster area as a priority section, if not, mark it as a constraint section, classify and mark the cluster area, and obtain the initial mapping structure of the topological map passable area.

[0087] According to the results in the sequence of structural boundary ratio difference, the structural difference value of each cluster needs to be calculated The comparison is made with a structural compactness ratio threshold, which uses the average of all differences in the current cluster set plus 0.25 times the standard deviation as a boundary reference. If the mean of structural compactness differences is 0.034 and the standard deviation is 0.016, then the threshold is set to 0.034 + 0.25 × 0.016 = 0.038. Therefore, the judgment boundary is set to 0.038. If the values ​​are less than the threshold, the segment is designated as a priority passage segment; otherwise, it is designated as a constraint segment. Based on the data in Table 2, C1, C3, C4, and C5 are all less than the threshold, thus they are priority passage segments, while C2 is a constraint segment. This result indicates that the core structure within region C2 is excessively loose or has an abnormal boundary distribution, requiring priority avoidance in subsequent path construction. After completing the labeling of all cluster regions, an initial mapping structure of accessible regions in the topology map is established, including the cluster number, boundary compactness classification attribute, and spatial location index. The advantage of the formula is that by introducing boundary change frequency and core expansion terms, the compactness quantification result can cover multi-dimensional structural states, effectively constructing the basic judgment criteria for path connectivity graphs. This result shows that the structural difference value not only reflects whether the boundary is closed but also quantifies whether the core region has coherent directionality, further supporting the hierarchical construction of regional accessibility structures in path planning.

[0088] Please see Figure 4 The specific steps for obtaining the mapping results of the directional stable path segment are as follows:

[0089] S311: Based on the initial mapping structure of the passable region in the topology map, obtain the set of passable nodes included in the priority passage segment, combine double node pairs in the node set according to the spatial coordinate order, extract the start and end coordinates of each path segment and construct the centerline vector of the path segment, combine the robot's current preset global motion direction vector to calculate the direction angle, and generate a sequence of path segment direction offset angles.

[0090] Based on the initial mapping structure of accessible areas in the topology map, the set of all nodes contained in all path segments marked as priority passage segments is obtained, and the two-dimensional spatial coordinates of each node in the map coordinate system are extracted. These nodes are then sequentially arranged from left to right and top to bottom according to their spatial relative positions, constructing pairs of adjacent nodes to form multiple two-node pairs. Each pair represents the start and end points of a path segment. Subsequently, for each pair of nodes, the centerline vector of the path segment is calculated using the difference between the two coordinates, representing the direction from the start point to the end point. For example, if node... , Then the line vector in this path segment is Let the robot's current preset global motion direction be... Next, the angle between the line vector and the direction vector in each path segment is calculated using the vector dot product formula:

[0091] ;

[0092] After obtaining the direction angle, convert it to angular units and establish the direction offset angle sequence of all path segments for subsequent analysis of direction fluctuation intensity, and finally obtain the direction offset angle sequence of the path segment;

[0093] S312: Call the path segment direction offset angle sequence, calculate the maximum and median of the path segment direction offset angle, extract the difference value as the direction fluctuation intensity factor, collect the path length value and the angle change rate of the path segment, and calculate the combined offset factor. The formula is:

[0094] ;

[0095] Calculate the direction offset combined intensity value, compare the direction offset combined intensity value of each path segment with the direction offset stability threshold, filter the path segments that meet the direction consistency, and obtain the direction consistency path segment sequence;

[0096] Wherein, represents the direction offset combined intensity value, represents the maximum value of the path segment direction offset angle, represents the median value of the direction offset angle, represents the normalized length value of the path segment, represents the normalized value of the direction angle change rate, represents the normalized value of the path density between the start and end points of the path segment, and the denominator controls the upper limit of the influence of high-density path segments;

[0097] Call the direction offset angle of all path segments in the path segment direction offset angle sequence, calculate the maximum value and the median value , the difference between the two reflects the overall direction fluctuation degree, which is defined as the direction fluctuation intensity factor, then collect the normalized path length , the normalized value of the direction angle change rate , and the normalized path density value between the start and end points of the path, and the combined offset factor formula is used to calculate the direction offset combined intensity value:

[0098] ;

[0099] The structure and logic of the formula are as follows: Construct the direction change fluctuation range term, used to amplify the influence of path segments with rapid angle changes, introduce the distance influence scale, constitute the denominator to adjust the high-density path segments and suppress the offset contribution of excessively dense paths. The calculation result This represents the joint strength of the directional offset for each path segment; a smaller value indicates greater directional stability. (Based on path segments...) For example, let its parameters be as follows:

[0100] , , , , ;

[0101] Substituting into the formula, we get:

[0102] ;

[0103] Table 3 Calculation Table of Joint Strength Value of Path Segment Directional Offset

[0104]

[0105] As shown in Table 3, path segment D4 has the largest joint strength value of directional offset, indicating that its directional change fluctuation is the most drastic, while path segments D3 and D5 have the best directional stability. The advantage of the formula is that by simultaneously considering path length, directional change amplitude, and density factors, the assessment of directional offset stability is made more spatially continuous and consistent with the motion trend.

[0106] S313: Based on the sequence of directional consistent path segments, mark the selected path segments in the original topology graph structure, assign the path segments the edge weight identifier attribute of the continuity priority edge, and establish the directional stable path segment mapping result.

[0107] Based on the sequence of directional consistent path segments, for all offset intensity values... Path segments below a preset stability threshold are marked. Let's say this threshold is set to 4.000. Referring to the data of path segments D1 to D5, path segments D3 and D5 are selected as meeting the criteria. These path segments are then located and identified in the original topology graph. Their edge identifiers are updated, and their edge weights are marked as continuity-priority edges. This provides a basis for path selection in subsequent path planning, establishing a mapping result for directionally stable path segments. This structure serves as the structural index for directionally consistent path segments in robot navigation path selection, ensuring the consistency of path continuity identifiers and the traceability of the graph structure. The final generated structure includes fields such as path segment number, directional stability identifier, and edge position index, for use by downstream tasks.

[0108] Please see Figure 5 The specific steps for obtaining the identification results of structure-sensitive path segments are as follows:

[0109] S411: Call the direction-stable path segment mapping result, count the number of topological connection edges of each node according to the start and end nodes of the path segment, record the arithmetic mean of the number of edges of the two end nodes of the path segment, and establish the average edge number information of the two end nodes of the path segment;

[0110] Call all path segment information in the direction-stable path segment mapping result, sequentially extract the start node number and end node number of each path segment, locate the node position in the topology graph structure, and count the number of directly connected edges of each node. For the two start and end nodes of each path segment, the number of edges connected by the nodes is respectively and , and the arithmetic mean is calculated, which represents the average topological connection strength of the boundary structure of the path segment, and is used to measure the density of the connection relationship of the two end nodes of the path segment in the overall graph structure. For example, if the number of edges connected by the start node of a path segment is 6 and the number of edges connected by the end node is 8, the average number of edges is . In this way, all path segments are traversed, the path segment number and the corresponding average number of edges are recorded, and complete average edge number information of the two end nodes of the path segment is established, which provides a connection baseline reference for subsequent structure boundary discrimination.

[0111] S412: According to the average edge number information of the two end nodes of the path segment, obtain the third-order adjacent node set of each end node of the path segment, and count the total number of connected edges of each node in the set, combine the network density background and the node distribution center of gravity index of the path segment in the topology graph, and use the formula:

[0112] ;

[0113] The structural boundary deviation strength value is obtained by operation, and a structural boundary deviation strength value sequence is obtained;

[0114] wherein, represents the structural boundary deviation strength value, represents the total connection edge number normalization value of the third-order adjacent nodes of the path segment, represents the average connection edge number normalization value of the two end nodes of the path segment, represents the number of path segments associated with the path segment in the topology graph normalization value, represents the node center of gravity index normalization value of the region where the path segment is located, represents the adjustment factor of the structure center of gravity index, the value range is usually 0.1 to 0.5, which is used to control the weight of the index, is a stable function coupled with the number of path segments, which is used to enhance the boundary difference response under high-density structure;

[0115] According to the acquired average edge number information at both ends of the path segment, the starting node and the ending node of each path segment are determined in the topology graph structure, and the adjacent node set of the third order outward of the node is further found, and all nodes in the set are traversed to count the connection edge number of each node and sum up to obtain the total connection edge number of the path segment in the third order neighborhood, denoted as , which reflects the local topology density. The is combined with the obtained in the previous step, and is brought into the structure boundary deviation enhancement calculation formula:

[0116]

[0117] , wherein represents the normalized value of the connection number of the current path segment and other path segments in the topology graph, controls the number of the path segment as the intersection point in the graph, represents the gravity concentration degree of the path segment area node in the spatial distribution, represents the adjustment factor of the gravity index, and the value range is set to 0.3. The deviation response gain is adjusted according to the concentration degree of the topology graph area structure. The calculation logic of the formula is as follows: the numerator term is the square value of the difference between the neighborhood and the path end, which represents the connection structure deviation degree; the square root in the denominator term is used to control the expansion of the path strength background, is a coupling term, which is used to balance the influence of the connection number on the structure diffusion; and the whole is multiplied by , to further strengthen the response of the path segment with gravity concentration to the boundary deviation. Taking the path segment P1 as an example, the parameters are as follows:

[0118]

[0119] Substituting the calculation gives:

[0120]

[0121] Table 4 shows that the path segment with a higher deviation intensity value, such as P3, represents a larger deviation of the adjacent structure from the average structure at both ends, which is a structure mutation point. The structure boundary deviation intensity value sequence is used for the boundary fluctuation judgment in the subsequent structure discrimination.

[0122]

[0123] As shown in Table 4, the path segment with a higher deviation intensity value, such as P3, represents a larger deviation of the adjacent structure from the average structure at both ends, which is a structure mutation point. The structure boundary deviation intensity value sequence is used for the boundary fluctuation judgment in the subsequent structure discrimination.

[0124] ​​​​​​​S413: Call the structure boundary deviation intensity value sequence, judge whether the structure boundary deviation intensity value is greater than the set boundary fluctuation coefficient threshold value, if the condition is met, mark the corresponding path segment as a structure critical segment, and establish a structure sensitive path segment identification result.

[0125] Call the deviation intensity value corresponding to each path segment in the obtained structure boundary deviation intensity value sequence , and compare it with the system set boundary fluctuation coefficient threshold value to determine whether it belongs to the structure transition area. According to the previous structure dispersion analysis and path segment density fluctuation range, the threshold value is set to 20.0. When the path segment deviation intensity value is greater than this threshold value, it is considered that the boundary is at the position of the structure mutation from dense to sparse or from sparse to dense, and the path segment is marked as a structure critical segment and included in the structure sensitive area path set; otherwise, it is not marked. Taking the path segment as an example, P1, P3, P5 all exceed the threshold value 20.0, and are marked as structure critical segments, and the rest do not meet the conditions. The marking result is associated with the path segment number, and the structure sensitive path segment identification result is established, which is used for path selection or graph structure updating processing in the next step. The final structure identification will be stored in three dimensions of path segment ID, deviation intensity and sensitive segment.

[0126] Please refer to Figure 6 , the acquisition steps of dynamic passable topology path marking information are as follows:

[0127] S511: Obtain the structure sensitive path segment identification result, read the number of calls and the number of successful records of each path segment in the robot continuous path call record, and perform difference calculation on the two data, and process the path segment index to generate a path segment call failure number sequence;

[0128] Obtain all path segment indexes in the structure sensitive path segment identification result, and call the call log of each path segment in the robot path planning system continuous job record in turn, extract the call number and corresponding successful record number of each path segment. The call number is denoted as , and the successful record number is denoted as , which comes from the path planning history table and task completion feedback record table recorded in the control system scheduling module. Taking path segment number P1 as an example, if the total call number is 80 times, of which the number of successful completion records is 68 times, the difference is . The difference represents the failure number of the path segment, which represents the frequency of failure to complete the planning due to navigation failure, obstacle avoidance timeout or dynamic path conflict in use. The failure number of each path segment is collected according to the path segment number to establish a path segment call failure number sequence, which is used as a key basis for subsequent passable state stability judgment.

[0129] According to the generated path segment call failure number sequence, the failure number corresponding to each path segment number is counted respectively and the total number of calls are proportionally converted, and the failure proportion calculation formula is executed:

[0130] ;

[0131] The calculation logic is to first obtain the failure rate of each path segment, and uniformly convert it into percentage format to represent the failure frequency of each path segment. Further, the result is compared with the system set passable state reference proportion threshold value, and the threshold value is 10%. This value is set based on the average path planning error tolerance rate of the robot platform in the indoor working environment and the multiple path segment scheduling data backtesting result. If the failure number of path segment P1 is 12 and the total number of calls is 80, then the failure proportion is:

[0132] ;

[0133] Therefore, the failure proportion of path segment P1 is higher than the set threshold value of 10%, which belongs to the path segment with low potential passable stability. In this way, the failure proportion values of all path segments are calculated, the path segment numbers of all failure proportions exceeding the threshold value are extracted, and the number, failure number, call number, and calculated failure proportion are recorded to form a path segment passable failure proportion value set.

[0134] Table 5: Path segment passable failure proportion example table

[0135]

[0136] As shown in Table 5, the failure proportions of path segments P1, P3, and P6 all exceed the threshold value, and need to enter the state calibration process.

[0137] S512: Based on the path segment call failure number sequence, according to the failure number and the total number of calls of the path segment, the failure frequency proportion calculation is executed, and the set passable state reference proportion is compared item by item, the path segment index higher than the threshold value is extracted and summarized, and the path segment passable failure proportion value set is generated;

[0138] The path segment numbers whose call path segment passing failure ratio value set is higher than the system set threshold value 10% are updated in the topology graph structure, the edges between the path segment nodes are set in the potential inaccessible state, and a specific identifier is given to identify and avoid in the subsequent path scheduling process. The state writing operation is realized by updating the path edge weight additional field in the topology graph data structure, for example, the edge weight field "access_status" of P1, P3 and P6 is adjusted from "normal" to "uncertain". The identifier will be used as a path segment risk weight reference in the dynamic path scheduling module. After completing the state writing of all the marked path segments, the dynamic accessible topology path marking information is formed, which is used for path fusion optimization processing in the next stage.

[0139] S513: The path segment passing failure ratio value set is called, the path segment is marked with a potential inaccessible attribute label according to the path segment index satisfying the threshold condition, the state writing is completed in the topology graph structure, and the dynamic accessible topology path marking information is established.

[0140] The above is only a preferred embodiment of the present application, and does not limit the present application in other forms. Any skilled person in the art can modify or change the disclosed technology content to equivalent embodiments applied to other fields, but any simple modification, equivalent change and modification made according to the technical essence of the present application to the above embodiments without departing from the technical solution content of the present application still belongs to the protection scope of the present application.

Claims

1. A method for robot path planning based on topology awareness, characterized in that, The method comprises the following steps: S1: obtaining a two-dimensional obstacle distribution map in a robot working area, establishing a set of grid units of a fixed size in the area, calculating a four-direction average blocking frequency of each grid, and performing interval judgment on the frequency and a preset blocking frequency threshold to generate grid-level obstacle blocking level label information; S2: calling the grid-level obstacle blocking level label information, calculating the number of boundary grids for each cluster area, marking the cluster area as a passing priority section or a constraint section, and generating an initial mapping structure of a passable area of a topology map; S3: based on the initial mapping structure of the passable area of the topology map, obtaining a line direction vector in each pair of path sections, extracting a maximum deviation value and a median deviation value in an angle sequence and performing difference value operation, screening path sections that meet the stable direction feature, and generating a direction stable path section mapping result; The obtaining step of the direction stable path section mapping result is specifically as follows: S311: based on the initial mapping structure of the passable area of the topology map, obtaining a set of passable nodes included in the passing priority section, combining double-node pairs in the node set in the order of spatial coordinates, extracting start and end coordinate points of each path section and constructing a path section center line vector, performing direction angle calculation in combination with a preset global motion direction vector of a robot, and generating a path section direction deviation angle sequence; S312: calling the path section direction deviation angle sequence, calculating a maximum value and a median value of the path section direction deviation angle, extracting a difference value as a direction fluctuation intensity factor, collecting a path length value and an angle change rate of the path section, performing joint deviation factor calculation, obtaining a direction deviation joint intensity value through operation, comparing the direction deviation joint intensity value of each path section with a direction deviation stability threshold, screening path sections that meet the direction consistency, and obtaining a direction consistency path section sequence; S313: according to the direction consistency path section sequence, performing a marking operation on the screened path sections in the original topology map structure, assigning the path sections with an edge weight identification attribute of a continuity priority edge, and establishing a direction stable path section mapping result; S4: calling the direction stable path section mapping result, according to start and end nodes of a path section, calculating an average value of the number of edges at both ends of the path section, obtaining a structure boundary deviation intensity value through operation, and if the structure boundary deviation intensity value is greater than a set boundary fluctuation coefficient threshold, marking the corresponding path section as a structure critical section to generate a structure sensitive path section recognition result; The obtaining step of the structure sensitive path section recognition result is specifically as follows: S411: calling the direction stable path section mapping result, according to start and end nodes of a path section, calculating the number of topology connection edges of each node, recording an arithmetic average value of the number of edges at both ends of the path section, and establishing average edge number information at both ends of the path section; S412: according to the average edge number information at both ends of the path section, obtaining a three-order adjacent node set at both ends of each path section, calculating a total sum of the number of connection edges of each node in the set, combining a network density background and a node distribution gravity index in which the path section is located in a topology map, obtaining a structure boundary deviation intensity value through operation, and obtaining a structure boundary deviation intensity value sequence; S413: Call the structure boundary deviation intensity value sequence, judge whether the structure boundary deviation intensity value is greater than the set boundary fluctuation coefficient threshold value, if the condition is met, mark the corresponding path segment as a structure critical segment, and establish a structure sensitive path segment identification result. 2.The topology-aware based robot path planning method according to claim 1, wherein, The grid-level obstacle obstruction level label information includes sparse level region annotation, transition level region annotation, dense level region annotation, direction obstruction statistical layer and grid number index, the topological map passable region initial mapping structure specifically is passable priority region set, constraint region set, space connected cluster label, boundary identification bitmap and passable segment space coordinate set, the direction stable path segment mapping result includes continuity priority path segment set, path segment direction offset rate label, path segment target alignment angle information, path segment direction stability score and path segment sequence number table, and the structure sensitive path segment identification result specifically refers to structure critical path segment set, path node connection edge fluctuation label, three-order adjacent node relationship graph, boundary jump segment identification set and topological overload flag. 3.The topology-aware based robot path planning method according to claim 2, wherein, The grid-level obstacle obstruction level label information acquisition step specifically includes: S111: Obtain a two-dimensional obstacle distribution map in a robot work area, establish a fixed edge length grid element set in the area, collect obstacle obstruction frequency count values of each grid in four orthogonal directions, and perform average value calculation on the count values of each direction to establish an average obstruction frequency data set of the grid based on four direction obstructions, and generate grid average obstruction frequency information; S112: Based on the grid average obstruction frequency information, call the average frequency value of each grid and perform interval judgment with the set obstruction frequency threshold value, perform section identification operation of the interval where the frequency value is located, set label attributes after judging whether the corresponding section is a low frequency section, a medium frequency section or a high frequency section, and generate a frequency interval label list; S113: According to the frequency interval label list, classify and label each grid according to the label attribute, call the grid number index and the label corresponding relationship, implement spatial marking of three types of labels of sparse, transition and dense, and obtain grid-level obstacle obstruction level label information. 4.The topology-aware based robot path planning method according to claim 3, wherein, The grid-level obstacle obstruction level label information acquisition step specifically includes: S211: Call the grid-level obstacle obstruction level label information, based on the spatial adjacency relationship of the sparse labeled grid, perform four-adjacent grid connectivity identification according to the sparse level annotation unit, aggregate process the grid groups distributed continuously in space, extract the closed region set composed of each group of adjacent grids, and generate a sparse region connected cluster number index; S212: Call the sparse region connected cluster number index, according to the grid number information of the cluster region, obtain the number of boundary grids and the number of core region grid points of each cluster region, calculate the ratio of the number of boundary grids to the number of core grids, obtain the structure compactness difference value by calculation, and generate a structure boundary ratio difference sequence; S213: According to the structure boundary ratio difference sequence, a judgment process is performed on the set structure compactness ratio threshold value to determine whether the structure compactness difference value is less than the structure compactness ratio threshold value. If yes, the corresponding cluster region is marked as a passing priority section. If not, it is marked as a constraint section. The cluster region is classified and labeled, and the initial mapping structure of the topological map passable region is obtained. 5.The topology-aware based robot path planning method according to claim 4, wherein, The method further comprises the following steps: S5: Obtain the path segment in the structure-sensitive path segment recognition result, calculate the path call failure ratio, and compare the failure ratio with the passable state benchmark ratio. If it is higher than the set proportion threshold value, perform a potential non-passable label marking operation on the path segment to generate dynamic passable topological path labeling information. The dynamic passable topological path labeling information includes potential non-passable path segment index, path segment historical failure comparison table, dynamic path label map, passable state level label, and path segment state change record. 6.The topology-aware based robot path planning method according to claim 5, wherein, The acquisition step of the dynamic passable topological path labeling information is specifically: S511: Obtain the structure-sensitive path segment recognition result, read the number of calls and the number of actual successful records in the robot continuous path call record of each path segment, calculate the difference between the two data, and perform corresponding processing on the path segment index to generate a path segment call failure number sequence; S512: Based on the path segment call failure number sequence, according to the failure number and the total number of calls of the path segment, perform failure frequency ratio calculation, compare each item with the set passable state benchmark ratio, extract the path segment index higher than the threshold value and summarize to generate a path segment passing failure ratio value set; S513: Call the path segment passing failure ratio value set, according to the path segment index that meets the threshold condition, mark the path segment with a potential non-passable attribute label, and complete the state writing in the topological graph structure to establish dynamic passable topological path labeling information.

Images