Automatic driving inspection method and system for fire-fighting robot

Abstract

The present application relates to the technical field of automatic driving inspection, in particular to an automatic driving inspection method and system for a fire-fighting robot, comprising the following steps: marking complex sections by acquiring grid structure, counting connection number, channel width and connection angle, extracting a passing path, calculating a primary path overlap ratio, generating a to-be-reconstructed path segment, combining the shortest path to generate an updated path segment sequence, screening the number of intersection points and path length, collecting fire source smoke coordinates, evaluating heat source and diffusion changes, predicting a shielding path segment, judging whether the path falls into the shielding segment, reconstructing a connection path, calculating redundancy length and overlap times, and generating an automatic driving inspection path through fuzzy comprehensive evaluation screening. In the present application, complex sections are marked by counting connection number, channel width and angle, repeated paths are identified to avoid crossing redundancy, reconstructed paths reduce cross interference, heat source smoke changes are evaluated to predict shielding areas, paths are screened to improve safety and continuity, and path adaptability and execution efficiency are strengthened.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic driving inspection, in particular to an automatic driving inspection method and system for a fire-fighting robot. BACKGROUND

[0002] The technical field of automatic driving inspection includes core technical contents such as autonomous perception, path planning and travel control for target scenes in complex environments; it is mainly applied to places that need to be regularly inspected and monitored, including power systems, petrochemical plant areas, mining areas and large infrastructure, through the integration of multi-sensor fusion perception, autonomous navigation system and environment recognition algorithm, the unmanned equipment can complete the operations such as path autonomous selection, obstacle avoidance and inspection task execution in complex space; the development of this field covers multiple technical links such as robot navigation control, environment map construction, dynamic obstacle avoidance, target recognition and positioning, and relies on high-precision positioning system and environment modeling means to realize accurate understanding and autonomous response to the operating environment.

[0003] Among them, the automatic driving inspection method and system for the fire-fighting robot refers to the environment perception device, path determination device and execution device deployed on the fire-fighting robot, which cooperate with the multi-source data fusion means to identify and determine the information such as obstacle distribution, travel path and heat source area in indoor or special operation place, and generate adaptive path travel instructions combined with the preset inspection route rules and the current environment state; the patent subject specifically covers the area recognition method based on joint analysis of visual image and temperature data, the steering control method realized by path curvature change calculation, and the obstacle avoidance priority sorting mechanism according to the space occupancy evaluation to complete the automatic driving inspection task.

[0004] In the actual inspection path construction process of the prior art, fixed path planning and environment static modeling are relied on, and there is a lack of fine judgment mechanism for path selection in space structure complex area, which is easy to cause path repetition and redundancy in high connection density and frequent angle change section; in addition, the risk area evaluation after the generation of the inspection path generally relies on one-time identification, and lacks periodic heat source and smoke state tracking, which leads to the inability to timely predict the dynamic change of the fire shielding area; in places where the fire situation is complex or the shielding area frequently changes, the static path is easy to fail, and the path instruction cannot be adjusted according to the real-time environment, causing path interruption or task failure; including in some factory buildings or warehouse areas, when the fire source spreads rapidly or the smoke suddenly spreads, the original path is easy to be blocked by shielding, and without continuous evaluation mechanism and path adaptive ability, it will seriously affect the continuity and environmental response efficiency of the fire-fighting robot task. SUMMARY

[0005] In order to solve the technical problems existing in the prior art, the present application provides an automatic driving inspection method and system for a fire-fighting robot.

[0006] To achieve the above object, the present application adopts the following technical solutions:

[0007] An automatic driving inspection method for a fire-fighting robot, comprising the following steps:

[0008] S1: Obtain the grid structure data of the inspection area, count the number of connected nodes, the width of the channel and the connection angle, mark the sections exceeding the average value of the whole area, and generate the structure complex section data;

[0009] S2: Call the structure complex section data, retrieve the node number after retrieving the passable path set, calculate the coincidence ratio of the original path segment and the passable path set and mark the repeated traversal path segment, and construct the path segment set to be reconstructed;

[0010] S3: Call the node number and arrange the path segment set to be reconstructed, call the Dijkstra shortest path algorithm to combine the paths, calculate the number of path intersection points and the path length to filter the paths, and generate the updated path segment sequence;

[0011] S4: Based on the updated path segment sequence, collect the coordinates of the fire source and the smoke, calculate the position increment of the heat source, the diffusion direction and speed change of the smoke, and perform continuous periodical fire source state change evaluation, predict the position of the fire blocking area, and generate the blocking path segment data;

[0012] S5: Determine whether the updated path segment sequence falls into the blocking path segment data, if it falls into, reconstruct the connected path, calculate the node redundancy length and the number of overlapping blocks, filter the path segment through the fuzzy comprehensive evaluation model, and generate the automatic driving inspection path.

[0013] The following is a further optimization of the technical solutions of the present application:

[0014] The structure complex section data includes the number of connected nodes, the position of the channel width abnormal section, and the connection angle fluctuation interval, the path segment set to be reconstructed includes the repeated traversal path segment number sequence, the coincidence ratio of the original path segment, and the passable path difference index, the updated path segment sequence includes the path segment combination order, the number of path intersection points, and the path length, the blocking path segment data specifically refers to the heat source position increment area, the smoke diffusion density segment, and the predicted blocking coverage range, and the automatic driving inspection path includes the node redundancy length, the number of overlapping blocks, and the fuzzy comprehensive evaluation preferred path.

[0015] Further optimization: the step S1 specifically comprises:

[0016] S101: After obtaining the grid structure data of the inspection area, the connection number of all nodes is counted based on the coordinate information and connection relationship of each grid node, and the angle value of each node associated channel is calculated by combining the connection line segments between nodes, to generate a node structure parameter set;

[0017] S102: According to the connection number and angle value in the node structure parameter set, the average value of each type of parameter in the whole area is calculated, and the node number and section number greater than any corresponding average value in the node structure parameter are marked to obtain the super-mean node distribution data;

[0018] S103: Call the node number information in the super-mean node distribution data, extract the corresponding grid structure and surrounding connection information, and merge and arrange according to the structure mode formed by the node in the spatial position to generate the structure complex section data.

[0019] Further optimization: the step of S2 is specifically:

[0020] S201: After calling the structure complex section data, retrieve each passable path set in the area, obtain the node number included in the path and extract the order information of the node, and according to the arrangement relationship of the node in the spatial structure, the numbering sequence is summarized and arranged to generate a passable node numbering set;

[0021] S202: According to the passable node numbering set, calculate the coincidence ratio value between the original path segment and each passable path set, compare all the ratio values with the set coincidence ratio reference value, mark the path segment number that exceeds the coincidence ratio reference value, and obtain the repeated crossing ratio value;

[0022] S203: For the path segment number information in the repeated crossing ratio value, extract the node sequence and spatial trend of the corresponding original path segment, and summarize and merge to form a continuous segment number sequence to establish a to-be-reconstructed path segment set.

[0023] Further optimization: the coincidence ratio reference value is set by extracting the median value or upper quartile value in the distribution interval of the coincidence ratio between all passable path sets and the original path segment in the inspection area as the interval critical point.

[0024] Further optimization: the step of S3 is specifically:

[0025] S301: After calling the to-be-reconstructed path segment set, the node number included in each path segment is extracted in turn, arranged according to the order of the node in the path, and the directed connection relationship of each path segment is constructed according to the numbering order to generate a path node sequence group;

[0026] S302: Based on the node number in each group of path node sequence group, the Dijkstra shortest path algorithm is called to calculate the modified edge weight in the structure network diagram, and the nodes in each path combination process are numbered and recorded in sequence to form the combination chain of continuous path segments, and the shortest path combination chain group is established;

[0027] S303: According to the shortest path combination chain group, the number of repeated numbers in the node number in the path segment and the total length of the path are calculated, and all path combinations are jointly screened according to the number of path intersection points and path length to generate an updated path segment sequence.

[0028] Further optimization: the step of S4 is specifically:

[0029] S401: Based on the updated path segment sequence, the fire source center coordinate point and the smoke area boundary point group in the corresponding area of each path segment are collected, the spatial displacement vector of the fire source point at the adjacent moment is calculated according to the coordinate difference, the spatial displacement change rate in the continuous period is analyzed, and the heat source position increment is obtained;

[0030] S402: The path segment area identified by the heat source position increment is called, the smoke area boundary point group in the corresponding period is extracted, the direction angle of smoke diffusion is calculated based on the angle relationship and time interval distance between the boundary points, the expansion length change rate is calculated and counted, and the smoke diffusion trend parameter group is obtained;

[0031] S403: According to the smoke diffusion trend parameter group and the heat source position increment value, the advancing direction and the shielding coverage extension distance of the shielding boundary in the fire source state change period of the path segment are calibrated, all path segment numbers in the shielding range are extracted, and the shielding path segment data is generated.

[0032] Further optimization: the step of S5 is specifically:

[0033] S501: Based on the updated path segment sequence and the shielding path segment data, the node numbers of each path segment are compared, the path segments in the shielding range are screened and the path numbers are recorded, the mapping table of path segment and shielding relationship is established, and the shielding path segment matching result is generated;

[0034] S502: According to the shielding path segment matching result, the connection relationship of the area with the shielding path segment is reconstructed, the head and tail node numbers and the total length value of each connection path are collected, the number of repeated nodes and the node connection length difference value in the path are calculated, and the node redundancy length and overlapping times parameter group are obtained;

[0035] S503: Call the node redundancy length and overlap times parameter group, score the multi-path segment according to the redundancy length ratio and overlap times coefficient of each path segment through the fuzzy comprehensive evaluation model, filter the path number sequence whose score does not exceed the path segment filtering reference value, and generate the automatic driving inspection path data.

[0036] Further optimization: the path segment filtering reference value is set by statistically calculating the weight score mean and standard deviation of the multi-path segment in the inspection task, and combining the filtering coefficient set by experience.

[0037] On the other hand, the application also provides an automatic driving inspection system for a fire-fighting robot, which is applied to the automatic driving inspection method for the fire-fighting robot, and the system comprises:

[0038] The structure analysis module acquires the grid structure data of the inspection area, counts the number of grid node connections, the width of the channel and the connection angle, marks the sections exceeding the average value of the whole area, generates the structure complex section data and transmits it to the path extraction module.

[0039] The path extraction module calls the structure complex section data, retrieves the passable path set, extracts the node number, calculates the overlap ratio of the original path segment and the passable path set and marks the repeated crossing path segment, constructs the path segment set to be reconstructed and transmits it to the path reconstruction module.

[0040] The path reconstruction module calls the node number of the path segment set to be reconstructed and arranges it, calls the Dijkstra shortest path algorithm for path combination generation, calculates the path intersection point number and the path length screening path, generates the updated path segment sequence and transmits it to the fire condition evaluation module.

[0041] The fire condition evaluation module acquires the fire source and smoke coordinates based on the updated path segment sequence, calculates the heat source position increment, smoke diffusion direction and speed change and performs continuous periodical fire source state change evaluation, predicts the position of the fire blocking area, generates the blocking path segment data and transmits it to the obstacle avoidance planning module.

[0042] The obstacle avoidance planning module judges whether the updated path segment sequence falls into the blocking path segment data, and if it falls into the blocking path segment data, the connection path is reconstructed, the node redundancy length and the number of overlaps are calculated, the path segment is screened through the fuzzy comprehensive evaluation model, and the automatic driving inspection path is generated.

[0043] The technical scheme provided by the embodiment of the application has at least the following beneficial effects:

[0044] In the construction and optimization process of the inspection path, the connection number, the channel width and the connection angle of the regional grid nodes are counted, and the complex structure section is marked, so that the quantitative analysis of the environmental structure complexity is effectively realized, and the path planning can be optimized for the high complexity area. By calculating the overlap ratio of the original path and the passable path, the repeated path segment is identified, the invalid path repetition is avoided, and the path accessibility is improved. Then, the path is reconstructed by the Dijkstra algorithm, and the path length and the number of path intersections are combined to optimize, so that the path redundancy and intersection interference are effectively reduced. Further, after the path is generated, the fire source and smoke diffusion information are collected in real time, the heat source increment and diffusion direction change are calculated, the fire source state is continuously and periodically evaluated, and the possible shielding area is accurately predicted, so that the potential shielding path segment is dynamically marked and the path effectiveness is evaluated. When the path overlaps with the shielding area, the node length and the shielding overlap frequency are automatically evaluated, the path segment is screened by the fuzzy comprehensive evaluation method, and the safety and execution continuity of the path are improved. The overall logic integrates the spatial structure identification, path reconstruction, risk warning and multi-index evaluation, strengthens the pertinence, timeliness and robustness of path selection, and significantly improves the adaptability and execution efficiency of the inspection task in the high-risk environment. BRIEF DESCRIPTION OF DRAWINGS

[0045] Figure 1 The workflow of the embodiment of the present application is shown in the figure.

[0046] Figure 2 The system module diagram in the embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0047] To make the technical problems, technical solutions and advantages of the present application clearer, specific embodiments will be described in detail below with reference to the drawings.

[0048] In the embodiments of the present application, the words such as "example", "include" are used to represent as an example, illustration or description. Any embodiment or design scheme described as "example" in the present application should not be interpreted as more preferred or more advantageous than other embodiments or design schemes. In fact, the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present application, the meaning expressed by "and / or" can be both, or can be one of the two.

[0049] In the embodiments of the present application, "image" and "picture" can be used interchangeably at times. It should be pointed out that when the distinction is not emphasized, the meanings expressed are consistent. "Of", "corresponding" and "relevant" can be used interchangeably at times. It should be pointed out that when the distinction is not emphasized, the meanings expressed are consistent.

[0050] In the embodiments of the present application, sometimes the subscript such as W1 may be written in the form of non-subscript such as W1, and the meanings expressed thereby are consistent when the difference is not emphasized.

[0051] Please refer to Figure 1 An automatic driving inspection method for a fire-fighting robot, comprising the following steps:

[0052] S1: Obtain grid structure data of the inspection area, count the number of grid node connections, channel width and connection angle, and mark the sections exceeding the average value of the entire area, and generate structure complex section data;

[0053] S2: Call the structure complex section data, retrieve the node number after retrieving the passable path set, calculate the coincidence ratio of the original path segment and the passable path set, and mark the repeated traversal path segment, and construct the path segment set to be reconstructed;

[0054] S3: Call the node number and arrange the path segment set to be reconstructed, call the Dijkstra shortest path algorithm to combine the paths, calculate the number of path intersection points and the path length to filter the paths, and generate the updated path segment sequence;

[0055] S4: Based on the updated path segment sequence, collect the fire source and smoke coordinates, calculate the heat source position increment, smoke diffusion direction and speed change, and perform continuous periodical fire source state change evaluation, predict the position of the fire blocking area, and generate the blocking path segment data;

[0056] S5: Determine whether the updated path segment sequence falls into the blocking path segment data, if it falls into, reconstruct the connection path, calculate the node redundancy length and the number of blocking overlaps, filter the path segment through the fuzzy comprehensive evaluation model, and generate the automatic driving inspection path.

[0057] The structure complex section data includes the number of connected nodes, the position of the node, the abnormal section of the channel width, and the connection angle fluctuation interval. The path segment set to be reconstructed includes the repeated traversal path segment number sequence, the coincidence ratio of the original path segment, and the passable path difference index. The updated path segment sequence includes the path segment combination order, the number of path intersection points, and the path length. The blocking path segment data specifically refers to the heat source position increment area, the smoke diffusion density section, and the predicted blocking coverage range. The automatic driving inspection path includes the node redundancy length, the number of blocking overlaps, and the fuzzy comprehensive evaluation preferred path.

[0058] Please refer to Figure 1 The steps of S1 are specifically:

[0059] S101: After obtaining the grid structure data of the inspection area, based on the coordinate information and connection relationship of each grid node, the number of connections of all nodes is counted, the angle value of the channel associated with each node is calculated combined with the connection line segment between nodes, and the node structure parameter set is generated;

[0060] Based on the inspection area grid structure data, the node coordinates and connection relationship are extracted, and the connection number of each node is counted. Taking node A (coordinate ) as an example, the connection nodes B , C , and D are established, the adjacent node ID is recorded, and the connection number statistics table is generated by traversing all nodes. For the three connection lines AB, AC, and AD of node A, the included angle between adjacent line segments is calculated: taking the vector of line segment AB , the vector of line segment AC , and the included angle is calculated by the dot product formula , the included angle is calculated by repeating the calculation of AB-AD , AC-AD , and AD , and the minimum value is taken as the included angle value. Table 1 shows a single area 6 node parameter example, in which the connection number of node E reaches 5 and the included angle value , and a parameter set including node coordinates, connection number, and included angle value is generated.

[0061] Table 1: Node structure parameter example table

[0062] Node number Connection number Angle value (°) X coordinate (m) Y coordinate (m) E 5 28 4.2 7.8 F 2 152 3.1 5.4 G 4 78 2.7 6.3 H 3 65 1.2 3.4 I 2 120 5.6 2.9 J 3 90 7.3 4.7

[0063] As shown in Table 1, node E is identified as a key node due to high connection number and small included angle value, and the parameter set provides basic data for subsequent analysis.

[0064] S102: According to the connection number and included angle value in the node structure parameter set, the average values of multiple parameters in the whole area are calculated respectively, and the node numbers and section numbers in the node structure parameters greater than any corresponding average value are marked to obtain the super-mean node distribution data;

[0065] When calculating the average connection number, take the connection number data set of the 6 nodes in Table 1 {5, 2, 4, 3, 2, 3}, the sum is 19, and the average is 3.17, the super-mean threshold is set to >3.17, node E (connection number 5) and node G (connection number 4) are marked, the average of the included angle value is taken as {28, 152, 78, 65, 120, 90}, and the average is 88.83 degrees, node E is 28 degrees lower than the average but still marked due to the connection number exceeding the threshold, and the section marking data is generated by traversing all nodes, including section X including nodes E, F, and G, 2 / 3 of the node parameters exceed the average value, and the section is marked as a high-risk section.

[0066] S103: Call the node number information in the super-mean node distribution data, extract the corresponding grid structure and surrounding connection information, merge and arrange according to the structure mode formed by the node in the space position, and generate structure complex section data;

[0067] Extract 12 associated nodes within 5 meters radius of node E, construct sub-grid structure, analyze its connection mode: node E forms star structure with nodes J, K (connection ratio 3:1.2), and ring structure with nodes L, M (standard deviation of included angle value up to 42°), calculate density threshold of 0.8 connection points per cubic meter by spatial density algorithm, merge to form complex section with diameter of 4.7 meters, section includes 3 types of topological structure and 4 high load nodes, generated structure complex section data includes spatial coordinate range , topological type distribution table and load level matrix.

[0068] Please refer to Figure 1 , the steps of S2 are as follows:

[0069] S201: After calling the structure complex section data, retrieve each passable path set in the region, get the node number included in the path and extract the node sequence information, according to the arrangement relationship of nodes in the spatial structure, the numbering sequence is summarized and arranged, and the passable node number set is generated;

[0070] Call the topological information of section X spatial range in the structure complex section data, traverse the 12 nodes and 28 connection edges included, retrieve all passable paths with the distance between the first and last nodes less than , including path P001 composed of nodes E-101 coordinates , J-205 , K-308 , the distance between adjacent nodes is calculated: , which verifies the passable condition, and the node sequence is recorded as {E-101, J-205, K-308}, path P002 is composed of nodes E-101, L-412 , M-517 , the distances are and , when summarizing the node sequence information of all paths, the node sequence {J-205, K-308, M-517} of path P003 is verified by spatial arrangement, it is found that the distance between K-308 and M-517 exceeds the threshold, the path is excluded, and the passable node number set including 6 valid paths is generated, table 2 shows the node sequence and coordinate verification results of 3 paths, by traversing the node sequence of all paths, the mapping relationship between path number and node sequence is established, including the coordinate distance of path P004 node sequence {F-201, G-305, H-409} is , , which is verified and included in the numbering set.

[0071] Table 2 verification table of passing path node sequence

[0072] Path number Node sequence Interval verification result (m) P001 E-101, J-205, K-308 0.14， 0.14 P002 E-101, L-412, M-517 0.11， 0.14 P004 F-201, G-305, H-409 0.12， 0.15

[0073] As shown in Table 2, the path node sequence is stored in ascending order according to the path number after being verified by the coordinate interval, forming a structured passing node number set.

[0074] S202: According to the passing node number set, the coincidence ratio value between the original path segment and each passing path set is calculated, all the ratio values are compared with the set coincidence ratio reference value, the path segment number exceeding the coincidence ratio reference value is marked, and the repeated crossing ratio value is obtained;

[0075] In step S202, the coincidence ratio reference value is set by extracting the median value or upper quartile value in the distribution interval between the original path segment and the passing path set in the inspection area, and extracting the median value or upper quartile value in the distribution interval as the interval critical point;

[0076] Extract the node sequence {E-101, J-205, K-308} of path P001 in the passing node number set, and compare it with the segment F1 (node sequence {E-101, J-205, N-609}) in the original path segment library, the number of coincident nodes is 2 (E-101, J-205), the total number of nodes is 3, and the coincidence ratio value is calculated , the coincidence ratio reference value is set to 0.6 (according to the data: analyze the coincidence degree distribution of 100 path segments, take the threshold value corresponding to the 80% quantile), traverse other path segments, including segment F2 (node sequence {L-412, M-527, Q-721}) and node sequence {E-101, L-412, M-517} of path P002, the number of coincident nodes is 1 (L-412), the ratio value , which is lower than the reference value, when marking the path segment exceeding the reference value, comparing segment F3 (node sequence {J-205, K-308, M-517}) with path P001, the number of coincident nodes is 2 (J-205, K-308), the ratio value , trigger the marking condition, and generate the repeated crossing ratio value data set by summarizing all the marking results, wherein the path segments F1 and F3 are marked as high repetition segments, and their numbers and ratio values are stored in the result table, including segment F4 (node sequence {G-305, H-409, I-503}) and the maximum coincidence ratio value of any passing path , which exceeds the threshold value.

[0077] S203: For the path segment number information in the repeated crossing ratio value, the node sequence and spatial trend of the corresponding original path segment are extracted, and a continuous segment number sequence is formed by summarizing and merging, and a to-be-reconstructed path segment set is established;

[0078] Extract the node sequence {E-101, J-205, N-609} of the marked path segment F1, and calculate its spatial orientation: the azimuth angle of the node E-101 to J-205 , the azimuth angle of the node J-205 to K-308 , the average orientation , merge the path segments with a deviation less than of the adjacent orientation, including the azimuth angle sequence of the segment F5 (node sequence {K-308, M-517, N-609}) , , the average , the deviation of F1 , trigger the merging condition, generate the continuous segment number sequence {S01, S02}, and when the spatial range is drawn, take the extreme coordinates of all merged segment nodes: the minimum X = 4.1, the maximum X = 4.5, the minimum Y = 7.8, and the maximum Y = 8.1. Calculate the node density: there are 8 nodes in the area, and the area , the density , combined with the connection number distribution (mean 3.2, standard deviation 1.1), filter the nodes {E-101, J-205, M-517} with connection number ≥4, establish the set of path segments to be reconstructed, including 2 core segments, 3 high connection nodes and 5 associated edges, covering 82% of the high load connections in the section X.

[0079] Please refer to Figure 1 , the steps of S3 are as follows:

[0080] S301: After calling the set of path segments to be reconstructed, extract the node numbers included in each path segment in turn, arrange them according to the order of the nodes in the path, and construct the directed connection relationship of each path segment according to the number order, generating a group of path node sequences;

[0081] Call the segments S01 (node sequence {E-101, J-205, K-308}) and S02 (node sequence {K-308, M-517, N-609}) in the set of path segments to be reconstructed, construct the directed connection according to the order of node appearance, and extract the edge weight of E-101→J-205 , the edge weight of J-205→K-308 , record as directed sequence {E→J→K}, when verifying node order, check whether there are other fragments downstream of K-308 (such as S02 K-308→M-517), if there are, mark as connection point, Table 3 shows the node order and edge weight of fragment S01, generate path node sequence group by traversing all fragments, including the edge weight sequence of fragment S03 (node sequence {G-305, H-409, I-503}) {0.12, 0.15}, after verification, it is included in the sequence group.

[0082] Table 3 Path fragment node order and edge weight

[0083] Fragment number Node sequence Edge weight (km) S01 E-101, J-205, K-308 0.14， 0.14 S02 K-308, M-517, N-609 0.18， 0.22

[0084] As shown in Table 3, the node sequence is strictly sorted in the direction of travel, providing structured input for subsequent path calculation.

[0085] S302: Based on each node number in the path node sequence group, call Dijkstra shortest path algorithm to calculate the modified edge weight in the structure network graph, and sequentially number and record the nodes in each path combination process to form the combination chain of continuous path segments, and establish the shortest path combination chain group;

[0086] Based on the path node sequence group, representative node to the original edge weight value, which is obtained by measuring the distance between nodes with a laser range finder, including the distance between nodes E-101 and J-205 , representative node degree centrality, the number of directly connected edges is counted, such as node E-101 connecting 5 edges, , representative path turning number, defined as the number of times the azimuth angle of adjacent segments in the path changes more than , including the azimuth angle difference between J→K and K→M in the path E→J→K→M , not counted, , representative path smoothing coefficient, set to according to the median of the smoothness of the path data set (100 paths), the value range is , representative turning penalty coefficient, through experimental test of the influence of multiple values on path efficiency, select the value that makes the variance of travel time minimum , representative angle deviation of the th turning, calculate the absolute value of the azimuth angle difference of adjacent segments, including the angle difference of single turning , then , the edge weight of nodes E-101 ( ) to J-205 ( ) is taken as an example, the original edge weight (measured value), the degree centrality term: , the path E→J→K exists once turn ( ), , the path smoothing coefficient : Based on the path smoothing degree score (1-5 points), the path ratio corresponding to the score ≥4 points is 30%, , the turning penalty coefficient : Through the control variable experiment, test the path efficiency when , select to make the average travel time shortest, substitute into the formula to calculate the shortest path, , the denominator calculation: , , .

[0087] Table 4 Comparison of edge weight correction

[0088] Path segment Original length (km) Corrected length (km) Whether selected E-101→J-205 0.14 0.283 Yes J-205→K-308 0.14 0.265 Yes K-308→M-517 0.18 0.401 No

[0089] As shown in Table 4, the total length of the corrected path E→J→K is still lower than that of K→M , but the latter is excluded due to insufficient connection number, and the result verifies the effective trade-off of the formula for the importance of node structure.

[0090] S303: According to the shortest path combination chain group, calculate the number of repeated node numbers in the path segment and the total length of the path, and then filter the path combination according to the number of path intersection points and path length. Generate an updated path segment sequence;

[0091] Extract the shortest path combination chain group path E→J→K→M (total length 1.2km, repeated node K-308 appears twice), calculate the repeated node ratio , set the filtering threshold: repeated ratio ≤0.4 and path length ≤1.5km, traverse other paths, such as path J→K→M→N (length 1.8km, repeated node 1, ratio 0.25), because it exceeds the length threshold, path G→H→I (length 1.1km, no repeated node) also meets the two conditions, when joint filtering, count the number of path intersection points: the intersection points of path E→J→K→M and path K→M→N are K-308 and M-517 (number 2), which exceeds the benchmark value 1, keep path G→H→I and E→J→K (intersection point number 1), generate an updated path segment sequence {S01-1, S03-1}, covering 92% of the efficient travel demand in the network graph.

[0092] Referring to Figure 1 , the steps of S4 are specifically:

[0093] S401: Based on the updated path segment sequence, the fire source center coordinate point and the smoke area boundary point group in the corresponding area of each path segment are collected, the spatial displacement vector of the fire source point at the adjacent time is calculated according to the coordinate difference value, the spatial displacement change rate in the continuous period is analyzed, and the heat source position increment is obtained;

[0094] Based on the path S01-1 (node sequence {E-101, J-205, K-308}) and S03-1 (node sequence {G-305, H-409, I-503}) in the updated path segment sequence, the fire source center coordinate point data is collected, and the path S01-1 area is taken as an example. At time The fire source point The coordinates are At time (interval ), the fire source point is offset to , the spatial displacement vector is calculated, the displacement change rate , the fire source point coordinates are , , , in three continuous periods (T1, T2, T3), and the lengths of the adjacent increment vectors are , , the average change rate is , and Table 5 shows the fire source point displacement data in the path S01-1 area. The heat source position increment table is generated by traversing all path segments, including the fire source point in the path S03-1 area from to , the increment is , and the change rate is .

[0095] Table 5: Fire source point spatial displacement data table

[0096] Path number Time (s) X coordinate (m) Y coordinate (m) Offset (m) S01-1 0 4.25 7.82 - S01-1 60 4.28 7.85 0.042 S01-1 120 4.32 7.88 0.042

[0097] As shown in Table 5, the fire source point shows a stable northeast migration trend, and the heat source position increment data provides input for subsequent diffusion analysis.

[0098] S402: Call the path segment area identified by the heat source position increment, extract the smoke area boundary point group in the corresponding period, calculate the direction angle of smoke diffusion based on the angle relationship and time interval distance between the boundary points, calculate the expansion length change rate and perform statistics, and obtain the smoke diffusion trend parameter group;

[0099] Call the path S01-1 area from to the set of smoke boundary points of the time period, Definition: the polar angle radian value of the th boundary point relative to the heat source reference point, calculated by coordinate conversion, formula is , where is the heat source reference point coordinate, is the boundary point coordinate, the heat source reference point , the boundary point coordinate , then , Definition: the time interval (unit: seconds) between the collection time of adjacent boundary points, directly obtained by the difference value of sensor time stamp example: , , then , Definition: time normalization reference quantity, fixed to , Definition: the absolute value of the change quantity of smoke expansion length in adjacent time periods, measure the radius difference of smoke area in adjacent time periods, example: time period 1 radius , time period 2 radius , then , Definition: the Euclidean distance (unit: meters) between the th and the th boundary point, calculation method: , xy represents the coordinates of p and q points, example: the distance between boundary points and : , Definition: the initial smoke area diameter (unit: meters), the maximum horizontal and vertical span of the initial detected smoke area boundary points are and , take the maximum value , substitute into the formula , the first term calculation: , , , the second term calculation: , result: , example result compared with the preset threshold threshold based on fire data statistics, indicates that the diffusion direction is highly unstable, the current value , indicating that the smoke diffusion direction is relatively stable in this time period, which can be marked as "low risk path segment".

[0100] S403: According to the smoke diffusion trend parameter group and the heat source position increment value, the advancing direction of the shielding boundary and the shielding coverage extension distance in the fire source state change period of the path segment are calibrated, all path segment numbers in the shielding range are extracted, and shielding path segment data is generated;

[0101] Combined with the heat source position increment (path S01-1 average offset rate ) and the smoke diffusion trend parameter ( ), the advancing direction of the shielding boundary is northeast , and the extension distance is calculated: , representing the average offset rate, representing the time, the shielding range is a circular area with the fire source point as the center and the radius , the path segment node coordinates are extracted, including the node K-308 ( ) distance from the fire source center is less than the shielding radius, and is marked as in the shielding range, after traversing all nodes, shielding path segment data is generated, including {K-308} of path S01-1 and {H-409} of path S03-1, path number S01-1, shielding node number K-308, distance from fire source (m) 0.143, path number S03-1, shielding node number H-409, distance from fire source (m) 0.210.

[0102] Please refer to Figure 1 , the steps of S5 are as follows:

[0103] S501: Based on the updated path segment sequence and the shielding path segment data, the node numbers of each path segment are compared, the path segments in the shielding range are screened and the path numbers are recorded, the mapping table of path segment and shielding relationship is established, and the shielding path segment matching result is generated;

[0104] Based on the path S01-1 (node sequence {E-101, J-205, K-308}) and S03-1 (node sequence {G-305, H-409, I-503}) in the updated path segment sequence, the shielding nodes K-308 and H-409 marked in the shielding path segment data are called, the path segment node sequence is traversed, and it is checked whether the node K-308 of S01-1 exists in the shielding node table (table 6), if it exists, the path number S01-1 is recorded, and the node H-409 (0.210 m away from the fire source, less than the shielding radius 2.542 m) of S03-1 is checked in the same way, the path number S03-1 is marked, and the mapping relationship is established, including the path S02-2 (node sequence {M-517, N-609}) without shielding node, which is not included in the mapping table, table 6 shows the matching result, covering 80% of all updated path segments.

[0105] Table 6: Shielded path segment matching table

[0106] Path number Shield node number Shielding state S01-1 K-308 Yes S03-1 H-409 Yes S02-2 - No

[0107] As shown in Table 6, paths S01-1 and S03-1 are marked for path reconstruction because they include shielding nodes.

[0108] S502: According to the shielding path segment matching result, the connection relationship of the area with shielding path segment is reconstructed, the first and last node numbers and the total length value of each connection path are collected, and the number of repeated nodes and the length difference of node connection in the path are calculated to obtain the node redundancy length and overlapping times parameter group;

[0109] The connection relationship of the shielding path segment S01-1 is reconstructed, the first and last nodes E-101 and K-308 are extracted, and the new path candidate scheme is calculated: 1. Path P1: E-101→F-201→K-308, total length 0.14+0.15=0.29km, repeated node number 0, 2. Path P2: E-101→J-205→K-308, total length 0.14+0.14=0.28km, repeated node number 2 (J-205, K-308), calculate redundancy length ratio: repeated node number / path P2 total node number=2 / 3≈0.67, path P1 is 0 / 3=0, length difference=|0.29-0.28|=0.01km, traverse other paths, such as path S03-1 reconstruction candidate path G-305→H-409→I-503 (length 0.12+0.15=0.27km, repeated node number 1), redundancy length ratio 1 / 3≈0.33, filter paths with redundancy length ratio>0.3 and length difference>0.005km, path number P1, redundancy length ratio 0.00, length difference (km) 0.01, path number P2, redundancy length ratio 0.67, length difference (km) 0.00, path number S03-1, redundancy length ratio 0.33, length difference (km) 0.02, path P2 is marked because of high redundancy length.

[0110] S503: Call the node redundancy length and overlapping times parameter group, score multiple path segments according to the redundancy length ratio and overlapping times coefficient of each path segment through the fuzzy comprehensive evaluation model, select path number sequences with scores not exceeding the path segment screening reference value, and generate automatic driving inspection path data;

[0111] In step S503, the path segment screening reference value is set by statistically calculating the weight score mean and standard deviation of multiple path segments in the inspection task, and combining the screening coefficient set by experience;

[0112] The redundancy length ratio of the path P2 in the call parameter group is 0.67, and the overlap times coefficient is 1 (nodes J-205 and K-308 appear twice in the original path), the fuzzy evaluation weight is set: the redundancy length ratio weight is 0.6, and the overlap times coefficient weight is 0.4, and the scoring formula is: , wherein is the redundancy length ratio, is the overlap times coefficient (the value is 0-1, and 1 represents the highest overlap), and the path P2 is calculated as: , the path segment screening reference value is set to 0.5 (according to data: 80% of the effective path score is less than or equal to 0.5), the score of the path P2 is 0.802, which is higher than the threshold value and is eliminated, and the score of the path P1 is is retained, and after traversing all paths, the automatic driving inspection path data is generated, including the path P1, S03-1 and the like 6 low-score paths, covering 92% of the safe travel demand of the area.

[0113] Please refer to Figure 2 , an automatic driving inspection system for a fire-fighting robot, the automatic driving inspection system for the fire-fighting robot is used for executing the automatic driving inspection method for the fire-fighting robot, and the system comprises:

[0114] a structure analysis module, which acquires grid structure data of an inspection area, counts the number of grid node connections, the width of a channel and the connection angle, marks sections exceeding the average value of the whole area, generates structure complex section data and delivers the structure complex section data to a path extraction module;

[0115] a path extraction module, which calls the structure complex section data, extracts node numbers after searching a passable path set, calculates the coincidence ratio of the original path segment and the passable path set and marks the repeated crossing path segment, constructs a to-be-reconstructed path segment set and delivers the to-be-reconstructed path segment set to a path reconstruction module;

[0116] a path reconstruction module, which calls the to-be-reconstructed path segment set to acquire node numbers and arrange the node numbers, calls a Dijkstra shortest path algorithm to combine paths to generate, calculates the number of path intersections and the length of the path to screen the path, generates an updated path segment sequence and delivers the updated path segment sequence to a fire condition evaluation module;

[0117] a fire condition evaluation module, which acquires fire source and smoke coordinates based on the updated path segment sequence, calculates the heat source position increment, the smoke diffusion direction and the speed change and performs continuous periodical fire source state change evaluation, predicts the position of a fire disaster shielding area, generates shielding path segment data and delivers the shielding path segment data to an obstacle avoidance planning module;

[0118] an obstacle avoidance planning module, which judges whether the updated path segment sequence falls into the shielding path segment data, and if yes, reconstructs a connection path, calculates the node redundancy length and the shielding overlap times, screens the path segment through a fuzzy comprehensive evaluation model, and generates an automatic driving inspection path.

[0119] If the functions are realized in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium; based on such understanding, the technical solutions of the present application essentially or the parts of the prior art or the parts of the technical solutions can be embodied in the form of software products, the computer software products are stored in a storage medium, and include a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in various embodiments of the present application; and the foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (read-only memory, ROM), a random access memory (random access memory, RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0120] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. An automatic driving inspection method for a firefighting robot, characterized by, The method comprises: S1: acquiring the grid structure data of the inspection area, counting the number of connected nodes, the width of the channel and the connection angle, marking the sections exceeding the average value of the whole area, and generating the complex structure section data; S2: calling the complex structure section data, retrieving the node number after extracting the set of passable paths, calculating the coincidence ratio of the original path segment and the set of passable paths and marking the repeated crossing path segment, and constructing the set of path segments to be reconstructed; S3: calling the set of path segments to be reconstructed to obtain the node number and arrange, calling the Dijkstra shortest path algorithm to combine the paths and generate, and calculating the number of path intersection points and the path length to filter the paths, and generating the updated path segment sequence; S4: based on the updated path segment sequence, collecting the coordinates of the fire source and the smoke, calculating the position increment of the heat source, the diffusion direction and speed change of the smoke, and performing continuous periodical fire source state change evaluation, predicting the position of the fire shelter area, and generating the shelter path segment data; S5: judging whether the updated path segment sequence falls into the shelter path segment data, if it falls into, reconstructing the connection path, calculating the node redundancy length and the number of shelter overlaps, filtering the path segment through the fuzzy comprehensive evaluation model, and generating the automatic driving inspection path.

2. The method for automatic driving inspection of a firefighting robot according to claim 1, wherein The complex structure section data comprises the number of connected nodes, the position of the connected nodes, the abnormal section of the channel width, and the connection angle fluctuation interval. The set of path segments to be reconstructed comprises the repeated crossing path segment number sequence, the coincidence ratio of the original path segment, and the difference index of the passable path. The updated path segment sequence comprises the path segment combination order, the number of path intersection points, and the path length. The shelter path segment data specifically refers to the heat source position increment area, the smoke diffusion density section, and the predicted shelter coverage range. The automatic driving inspection path comprises the node redundancy length, the number of shelter overlaps, and the fuzzy comprehensive evaluation preferred path.

3. The method of claim 1, wherein the method further comprises: The steps of S1 are specifically: S101: after obtaining the grid structure data of the inspection area, based on the coordinate information and connection relationship of each grid node, the number of connections of all nodes is counted, the angle value of the channel associated with each node is calculated combined with the connection line segment between nodes, and the node structure parameter set is generated; S102: according to the connection number and angle value in the node structure parameter set, the average value of multiple parameters in the whole area is calculated respectively, the node number and section number in the node structure parameter greater than any corresponding average value are marked, and the super-mean node distribution data is obtained; S103: calling the node number information in the super-mean node distribution data, extracting the corresponding grid structure and surrounding connection information, merging and organizing according to the structure mode formed by the node in the spatial position, and generating the complex structure section data.

4. The method of claim 1, wherein the method further comprises: The steps of S2 are specifically: S201: after calling the complex structure section data, retrieving each set of passable paths in the area, obtaining the node number included in the path and extracting the order information of the node, and according to the arrangement relationship of the node in the spatial structure, the numbering sequence is summarized and organized, and the set of passable node numbers is generated; S202: According to the set of passable node numbers, the coincidence ratio value between the original path segment and each set of passable paths is calculated, all the ratio values are compared with the set coincidence ratio reference value, the path segment numbers exceeding the coincidence ratio reference value are marked, and the repeated crossing ratio value is obtained; S203: For the path segment number information in the repeated crossing ratio value, the node sequence and the spatial trend corresponding to the original path segment are extracted, and the continuous segment number sequence is formed by summarizing and merging, and the set of path segments to be reconstructed is established.

5. The method of claim 4, wherein the method further comprises: The coincidence ratio reference value is set by extracting the median value or the upper quartile value in the distribution interval between the set of all passable paths in the inspection area and the original path segment as the interval critical point.

6. The method of claim 1, wherein the method further comprises: The steps of S3 are specifically: S301: After calling the set of path segments to be reconstructed, the node numbers included in each path segment are extracted in turn, arranged according to the order of the nodes in the path, and the directed connection relationship of each path segment is constructed according to the number order to generate the path node sequence group; S302: Based on each group of node numbers in the path node sequence group, the Dijkstra shortest path algorithm is called to calculate the modified edge weight in the structural network graph, and the nodes in each path combination process are numbered and recorded in turn to form the combination chain of continuous path segments, and the shortest path combination chain group is established; S303: According to the shortest path combination chain group, the number of repeated numbers in the node numbers in the path segment and the total length of the path are calculated, and all path combinations are jointly screened according to the number of path intersection points and the length of the path to generate the updated path segment sequence.

7. The method of claim 1, wherein the method further comprises: The steps of S4 are specifically: S401: Based on the updated path segment sequence, the fire source center coordinate point and the smoke area boundary point group in the corresponding area of each path segment are collected, the spatial displacement vector of the fire source point at the adjacent moment is calculated according to the coordinate difference, the spatial displacement change rate in the continuous period is analyzed, and the heat source position increment is obtained; S402: The path segment area identified by the heat source position increment is called, the smoke area boundary point group in the corresponding period is extracted, the direction angle of smoke diffusion is calculated based on the angle relationship and time interval distance between the boundary points, the expansion length change rate is calculated and counted, and the smoke diffusion trend parameter group is obtained; S403: According to the smoke diffusion trend parameter group and the heat source position increment value, the advancing direction and the shielding coverage extension distance of the shielding boundary in the heat source state change period of the path segment are calibrated, all the path segment numbers in the shielding range are extracted, and the shielding path segment data is generated.

8. The method of claim 1, wherein the method further comprises: The steps of S5 are specifically: S501: Based on the updated path segment sequence and the shielding path segment data, the node numbers of each path segment are compared, the path segments in the shielding range are screened and the path numbers are recorded, the mapping table of the path segment and the shielding relationship is established, and the shielding path segment matching result is generated; S502: According to the shielding path segment matching result, the path connection relationship of the area with the shielding path segment is reconstructed, the head and tail node numbers and the total path segment length value of each connection path are collected, and the node redundancy length and the node connection length difference value are calculated, and the node redundancy length and the overlapping times parameter group are obtained; S503: The node redundancy length and the overlapping times parameter group are called, and according to the redundancy length ratio and the overlapping times coefficient of each path segment, the multi-path segment is scored through the fuzzy comprehensive evaluation model, the path number sequence with a score not exceeding the path segment screening reference value is screened, and the automatic driving inspection path data is generated.

9. The method of claim 8, wherein the method further comprises: The path segment screening reference value is set by statistically analyzing the weight score mean and standard deviation of the multi-path segment in the inspection task, and combining the screening coefficient set by experience.

10. An automatic driving inspection system for a fire-fighting robot, characterized by, The system is used to realize the automatic driving inspection method of the fire-fighting robot according to any one of claims 1-9, and the system comprises: A structure analysis module is configured to obtain grid structure data of an inspection area, count the number of grid node connections, channel width and connection angle, mark sections exceeding the average value of the whole area, generate structure complex section data and pass it to the path extraction module; A path extraction module is configured to call the structure complex section data, retrieve the node number of the passable path set, calculate the coincidence ratio of the original path segment and the passable path set and mark the repeated crossing path segment, construct the path segment set to be reconstructed and pass it to the path reconstruction module; A path reconstruction module is configured to call the node number of the path segment set to be reconstructed and arrange it, call the Dijkstra shortest path algorithm to combine the paths, calculate the path intersection point number and the path length screening path, generate the updated path segment sequence and pass it to the fire condition evaluation module; A fire condition evaluation module is configured to collect the fire source and smoke coordinates based on the updated path segment sequence, calculate the heat source position increment, smoke diffusion direction and speed change and perform continuous periodical fire source state change evaluation, predict the fire shielding area position, generate the shielding path segment data and pass it to the obstacle avoidance planning module; An obstacle avoidance planning module is configured to determine whether the updated path segment sequence falls into the shielding path segment data, and if it falls into the shielding path segment data, to reconstruct the connection path, calculate the node redundancy length and the shielding overlapping times, screen the path segment through the fuzzy comprehensive evaluation model, and generate the automatic driving inspection path.

Images