Path planning method, system, computer device and computer readable storage medium

Abstract

The application discloses a path planning method and system, computer equipment and a computer readable storage medium, and relates to the technical field of path planning. The method comprises the following steps: calculating a passing slope cost, a ground roughness cost and a communication signal cost according to a digital elevation raster map of a target area; determining a comprehensive cost map according to the passing slope cost, the ground roughness cost and the communication signal cost; and performing hierarchical security degradation path search according to the comprehensive cost map to obtain a planning path of the target area. In this way, the hierarchical security degradation path search mechanism is introduced by fusing the three-dimensional costs of the slope, the roughness and the communication to construct the comprehensive cost map, and a safe path that avoids extreme slopes, high-roughness ground and communication blind areas is planned for the target area.

Description

Technical Field

[0001] This invention relates to the field of path planning technology, and in particular to a path planning method, system, computer device, and computer-readable storage medium. Background Technology

[0002] While existing humanoid mining robots possess basic environmental perception and search capabilities during path planning, they face multi-dimensional safety challenges when autonomously navigating in complex tunnel environments. This imposes numerous constraints on the comprehensiveness and reliability of path planning. Paths planned by existing humanoid mining robot path planning technologies may traverse communication blind spots, areas with extreme slopes, or rough and unstable ground, directly impacting the robot's operational safety and reliability in mine tunnels. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to overcome the shortcomings of the prior art and provide a path planning method, system, computer device and computer-readable storage medium.

[0004] This invention provides the following technical solution: In a first aspect, the present invention provides a path planning method, the method comprising: Calculate the traffic slope cost, surface roughness cost, and communication signal cost based on the digital elevation raster map of the target area; A comprehensive cost map is determined based on the cost of traffic slope, the cost of surface roughness, and the cost of communication signals. Based on the comprehensive cost map, a hierarchical security degradation path search is performed to obtain the planned path for the target area.

[0005] In an optional implementation, a comprehensive cost map is determined based on traffic slope cost, surface roughness cost, and communication signal cost, including: Set the preset passage threshold as the first threshold, and for each grid corresponding to the digital elevation raster map, determine whether the communication slope cost or the communication signal cost of the current grid is greater than or equal to the preset passage threshold. If the communication slope cost or the communication signal cost of the current grid is greater than or equal to the preset passage threshold, then the total cost of the current grid is the preset cost. If the communication slope cost and the communication signal cost of the current grid are both less than the preset passage threshold, then based on the preset weight, the passage slope cost, the surface roughness cost, and the communication signal cost of the current grid are weighted and summed to obtain the comprehensive cost of the current grid. A comprehensive cost map is obtained based on the comprehensive cost of each grid.

[0006] In an optional implementation, a hierarchical security degradation path search is performed based on the comprehensive cost map to obtain a planned path for the target area, including: Path search is performed based on the comprehensive cost map; if a feasible path is found, it is used as the planned path; if no feasible path is found, the preset passage threshold is set as the second threshold, and the steps of determining whether the communication slope cost or the communication signal cost of the current grid is greater than or equal to the preset passage threshold are re-executed for each grid corresponding to the digital elevation raster map, so as to obtain a new comprehensive cost map. The path search continues based on the new comprehensive cost map. If a feasible path is found, it is used as the planned path, and a downgrade search prompt and critical area information are generated. If no feasible path is found, the comprehensive cost map is adjusted based on the preset penalty coefficient to obtain the adjusted comprehensive cost map. The path search continues based on the adjusted comprehensive cost map to obtain the planned path.

[0007] In an optional implementation, the comprehensive cost map is adjusted based on a preset penalty coefficient to obtain an adjusted comprehensive cost map, including: Identify at least one grid cell in the comprehensive cost map whose comprehensive cost is a preset cost, and use it as a candidate grid cell; For each candidate grid, based on preset weights, the original cost of the candidate grid is obtained by weighted summation of the traffic slope cost, the surface roughness cost, and the communication signal cost. Calculate the first difference between the original cost and the first threshold, and calculate the second difference between the preset value and the first threshold; calculate the ratio of the first difference to the second difference, and add the product of the ratio and the preset penalty coefficient to the preset value to obtain the soft penalty value; The overall cost of candidate grid cells in the overall cost map is adjusted to a soft penalty value to obtain the adjusted overall cost map.

[0008] In an optional implementation, the access slope cost is calculated based on a digital elevation raster map of the target area, including: Calculate the X-axis gradient and Y-axis gradient of each grid cell in the digital elevation raster map; The slope angle of each grid is calculated based on the X-direction gradient and the Y-direction gradient of each grid. Map the slope angle of each grid cell to the passage slope cost of each grid cell.

[0009] In an optional implementation, the surface roughness cost is calculated based on a digital elevation raster map of the target area, including: Calculate the mean local elevation of each grid cell in a digital elevation raster map; The variance of each grid cell is calculated based on the mean local elevation of each grid cell. The variance of each grid cell is mapped to the surface roughness cost of each grid cell.

[0010] In an optional implementation, the communication signal cost is calculated based on a digital elevation raster map of the target area, including: Calculate the planar distance from the preset communication base station to each grid in the digital elevation raster map based on the digital elevation raster map; Based on the logarithmic distance path loss model, the planar distance from the preset communication base station to each grid is calculated to obtain the base station received power corresponding to each grid. For grid cells whose elevation values ​​are greater than a preset elevation threshold, the received power of the base station corresponding to the grid cell is attenuated. The base station received power corresponding to each grid is mapped to the communication signal cost of each grid.

[0011] Secondly, the present invention provides a path planning system, the system comprising: The calculation module is used to calculate the access slope cost, surface roughness cost, and communication signal cost based on the digital elevation raster map of the target area; The determination module is used to determine the comprehensive cost map based on the traffic slope cost, surface roughness cost, and communication signal cost; The search module is used to perform hierarchical security degradation path search based on the comprehensive cost map to obtain the planned path for the target area.

[0012] Thirdly, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, it implements a path planning method as described in any of the foregoing embodiments.

[0013] Fourthly, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements a path planning method as described in any of the foregoing embodiments.

[0014] The path planning method, system, computer equipment, and computer-readable storage medium disclosed in this invention calculate the travel slope cost, surface roughness cost, and communication signal cost based on a digital elevation raster map of the target area; determine a comprehensive cost map based on the travel slope cost, surface roughness cost, and communication signal cost; and perform a hierarchical safety degradation path search based on the comprehensive cost map to obtain the planned path for the target area. In this way, by integrating the costs of slope, roughness, and communication to construct a comprehensive cost map and introducing a hierarchical safety degradation path search mechanism, safe paths are planned for the target area, avoiding extreme slopes, high-roughness ground, and communication blind spots. This significantly improves the safety and robustness of path planning, both rigidly avoiding single-point fatal risks such as overturning, instability, and disconnection, and providing step-by-step controllable degradation when no feasible path is available, ensuring task continuity. Simultaneously, interpretable risk gradient information is output to support real-time motion decision-making and manual intervention. Attached Figure Description

[0015] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope of protection of the present invention. In the various drawings, similar components are numbered similarly.

[0016] Figure 1 A flowchart of the path planning method proposed in this embodiment is shown; Figure 2 Another flowchart of the path planning method proposed in this embodiment is shown; Figure 3 This illustration shows another flowchart of the path planning method proposed in this embodiment; Figure 4 Another flowchart of the path planning method proposed in this embodiment is shown; Figure 5 A schematic diagram of the path planning system proposed in this embodiment is shown.

[0017] Explanation of reference numerals in the attached diagram: 500 - Path planning system; 501 - Calculation module; 502 - Determination module; 503 - Search module. Detailed Implementation

[0018] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0019] The components of the embodiments of the invention described and illustrated herein can typically be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0020] In the following, the terms “comprising,” “having,” and their cognates, which may be used in various embodiments of the invention, are intended only to indicate a particular feature, number, step, operation, element, component, or combination thereof, and should not be construed as excluding, firstly, the presence of one or more other features, numbers, steps, operations, elements, components, or combinations thereof, or adding the possibility of one or more features, numbers, steps, operations, elements, components, or combinations thereof.

[0021] Furthermore, the terms "first," "second," and "third" are used only to distinguish descriptions and should not be interpreted as indicating or implying relative importance.

[0022] Unless otherwise specified, all terms used herein (including technical and scientific terms) shall have the same meaning as commonly understood by one of ordinary skill in the art to which the various embodiments of the invention pertain. Terms (such as those defined in commonly used dictionaries) shall be interpreted as having the same meaning as in their contextual meaning in the relevant technical field and shall not be interpreted as having an idealized or overly formal meaning, unless clearly defined in the various embodiments of the invention.

[0023] Example 1 This disclosure provides a path planning method for realizing path planning of humanoid robots in mine roadways based on multi-dimensional cost fusion.

[0024] Please see Figure 1 The path planning method includes steps S101 to S103, and each step is described in detail below.

[0025] Step S101: Calculate the traffic slope cost, surface roughness cost, and communication signal cost based on the digital elevation raster map of the target area.

[0026] In this embodiment, if the target area is a mine roadway, the cost of traversing slope, surface roughness, and communication signal are calculated based on a digital elevation raster map. Simultaneously considering three dimensions—slope (affecting overturning safety), roughness (affecting gait stability), and communication signal coverage (affecting remote communication and control capabilities)—a multi-dimensional environmental cost model that more closely reflects actual working conditions is constructed. This enables the path planning system to select a safe and reliable weighted traversal route for the humanoid robot.

[0027] Among them, gradient differential operation based on digital elevation raster map is used to calculate the passage slope cost of each grid, local statistical modeling based on digital elevation raster map is used to calculate the surface roughness cost of each grid, and path loss model based on 5G communication in mine roadway is used to calculate the communication signal cost of each grid.

[0028] It should be noted that, in addition to the three dimensions of slope, roughness, and signal, environmental factors such as temperature, humidity, and dust concentration can be added to construct a more multi-dimensional cost fusion model.

[0029] In one specific embodiment, step S101 includes: calculating the X-direction gradient and the Y-direction gradient of each grid in the digital elevation grid map; calculating the slope angle of each grid based on the X-direction gradient and the Y-direction gradient of each grid; and mapping the slope angle of each grid to the access slope cost of each grid.

[0030] In this embodiment, to address the cost of traffic slope, the digital elevation raster map is first subjected to Gaussian smoothing with a smoothing parameter σ = 1.5 pixels. The purpose is to eliminate false steep slopes caused by random noise and ensure the accuracy of subsequent gradient calculations.

[0031] Furthermore, using the smoothed digital elevation raster map, the X-direction gradient and Y-direction gradient of each raster are calculated using the central difference method. The calculation formula is as follows: In the formula, The gradient is in the X direction. The gradient is in the Y direction, and Δ = 0.1m is the pixel spacing. This is a smoothed digital elevation raster map.

[0032] Furthermore, the slope angle of each grid is calculated based on the X-direction gradient and the Y-direction gradient of each grid. .

[0033] Furthermore, a piecewise linear mapping method is used to calculate the traffic slope cost. The calculation formula is: In the formula, = 15° is the maximum safe climbing angle for a humanoid robot. This parameter is set based on the dynamic constraints of the humanoid robot's bipedal gait. When the slope exceeds this angle, the robot is at risk of tipping over. At this time, the slope cost is forcibly set to 1, indicating that the area is impassable.

[0034] In one specific embodiment, step S101 includes: calculating the local elevation mean of each grid in the digital elevation raster map; calculating the variance of each grid based on the local elevation mean of each grid; and mapping the variance of each grid to the surface roughness cost of each grid.

[0035] In this embodiment, convolution is used to calculate each grid cell. Local average elevation: .

[0036] Furthermore, the digital elevation data of each grid in the digital elevation raster map are calculated frequently to obtain the elevation values ​​of each grid. .

[0037] Furthermore, the variance of each grid cell The calculation formula is: .

[0038] Furthermore, the surface roughness cost is obtained through mapping. The calculation formula is: ;in, The maximum tolerable roughness threshold can be set to 0.005, corresponding to a local elevation difference of approximately 7 cm. This threshold is based on the stability of the humanoid robot's foot area in contact with the ground.

[0039] In one specific embodiment, step S101 includes: calculating the planar distance from a preset communication base station to each grid in the digital elevation grid map based on the digital elevation grid map; calculating the planar distance from the preset communication base station to each grid based on the logarithmic distance path loss model to obtain the base station received power corresponding to each grid; attenuating the base station received power corresponding to the grid to be processed for grids whose elevation values ​​are greater than a preset elevation threshold; and mapping the base station received power corresponding to each grid to the communication signal cost of each grid.

[0040] In this embodiment, two 5G communication base stations are deployed in the mine roadway, located at 30m and 130m longitudinally respectively, to ensure full signal coverage.

[0041] Furthermore, the planar distances of each grid cell from the two 5G communication base stations are calculated separately.

[0042] Furthermore, the single base station received power of each grid The logarithmic distance path loss model is used for calculation: Wherein, the carrier frequency f = 3.5 GHz, and the reference distance... =1m, Path Loss Index =2.45 (experience value of alleyway environment). This represents the planar distance corresponding to the grid. Transmit power Pt = -32dBm, transmit antenna gain Gt = 5dBi, receive antenna gain Gr = 0dBi.

[0043] Furthermore, for rock wall areas, i.e., grid cells whose digital elevation values ​​exceed a preset elevation threshold (e.g., 0.3m), an additional 30dB of wall penetration attenuation is added to the received power of the two single-base stations corresponding to the grid cell, based on the path loss. The base station received power for each grid cell includes the dual-base station received power. Merging via power superposition: , and This represents the final received power of the two single base stations in the current grid.

[0044] Furthermore, the cost of communication signals Piecewise linear mapping is used: , This is the upper limit value. This is a lower limit value, and no limitation is made here in this embodiment.

[0045] In addition to the logarithmic distance path loss model, the ITU-R tunnel propagation model, ray tracing model, or empirical models based on measured data can be used to calculate the signal field strength. The communication standard can also be replaced by other communication technologies such as WiFi 6 and UWB, instead of 5G.

[0046] Understandably, this is the first time that the actual propagation characteristics of the 3.5GHz band in mine tunnels (path loss index 2.45, wall penetration attenuation 30dB) have been incorporated into the path planning cost system, overcoming the oversimplification of signal coverage modeling in traditional methods. The dual-base station power superposition model can accurately reflect the differences in communication quality at different locations within the tunnel.

[0047] It should be added that, for cost mapping, in addition to piecewise linear mapping, the cost of each dimension can be adjusted by using the Sigmoid function, logarithmic function or piecewise nonlinear mapping to adjust the cost sensitivity of different intervals.

[0048] Step S102: Determine the comprehensive cost map based on the traffic slope cost, surface roughness cost, and communication signal cost.

[0049] In this embodiment, the combined cost map is determined by combining the cost of traversing slope, the cost of surface roughness, and the cost of communication signals, serving as the data basis for hierarchical safety degradation path search.

[0050] Please see Figure 2 In one specific embodiment, step S102 includes steps S1021 to S1024, and each step is described in detail below.

[0051] Step S1021: Set the preset passage threshold as the first threshold. For each grid corresponding to the digital elevation raster map, determine whether the communication slope cost or the communication signal cost of the current grid is greater than or equal to the preset passage threshold.

[0052] In this embodiment, the preset passage threshold is a non-passable threshold. First, the preset passage threshold is set as a first threshold, corresponding to the lowest security tolerance, used to provide limited assurance of the highest security level. Further, for each grid corresponding to the digital elevation raster map, it is determined whether the communication slope cost or the communication signal cost of the current grid is greater than or equal to the preset passage threshold (first threshold) to perform a first-level search judgment. The first threshold can be 0.999, and this embodiment does not limit it to this value.

[0053] Understandably, the pre-set passage threshold and veto rule provide a hard guarantee of safety. Even if a location performs well in terms of slope and roughness, if insufficient signal coverage leads to a risk of communication interruption, that location is marked as impassable, fundamentally preventing humanoid robots from being planned to pass through areas with safety hazards in a single dimension.

[0054] Step S1022: If the communication slope cost or the communication signal cost of the current grid is greater than or equal to the preset passage threshold, then the comprehensive cost of the current grid is the preset cost.

[0055] In this embodiment, if the communication slope cost or the communication signal cost of the current grid is greater than or equal to a preset passage threshold, it is known that the current grid is impassable. The comprehensive cost of the current grid is then set to the preset cost for rejection, ensuring that the humanoid robot will not be planned to pass through areas with extreme slopes or communication blind spots, thus fundamentally avoiding single-dimensional safety hazards. At the same time, an impassable logical barrier is naturally formed in subsequent path search, eliminating the need for additional obstacle labeling or topology pruning, and greatly simplifying the map preprocessing process.

[0056] The preset cost can be, for example, the value "1" or other values, and this embodiment does not limit it.

[0057] It should be noted that, in addition to the rejection based on the slope and signal dimensions, it can be extended to trigger rejection when any dimension reaches the threshold, or it can be set to reject multiple dimensions together, such as rejecting only when both slope and roughness exceed 80% of their respective thresholds.

[0058] Step S1023: If the communication slope cost and the communication signal cost of the current grid are both less than the preset passage threshold, then based on the preset weight, the passage slope cost, the surface roughness cost, and the communication signal cost of the current grid are weighted and summed to obtain the comprehensive cost of the current grid.

[0059] In this embodiment, if both the communication slope cost and the communication signal cost of the current grid are less than a preset passage threshold, then the current grid is allowed to pass. The overall cost of the current grid is then... The calculation formula is: ,in, , , All weights are preset and determined using the Analytic Hierarchy Process (AHP).

[0060] Exemplary, =0.40、 =0.25、 =0.35, satisfying the consistency ratio (CR) <0.1. The weight allocation is based on: inviting 4 mining experts to compare the three indicators of slope, roughness and communication signal using the scaling method, and constructing a judgment matrix. Slope has the greatest impact on the safety of humanoid robots (overturning risk), followed by signal cost (communication interruption risk), and roughness has a relatively small impact on gait stability.

[0061] Understandably, the AHP weighted fusion method provides a scientific basis for weight allocation across the three dimensions (CR < 0.1 ensures consistency), avoiding the subjectivity and arbitrariness of empirical weighting. The piecewise linear cost mapping function ensures that the costs of each dimension have a unified physical meaning within the range [0,1]: 0 represents optimal conditions, and 1 represents complete impassability.

[0062] It should be noted that, in addition to the Analytic Hierarchy Process (AHP), entropy weighting, coefficient of variation (COP) methods, or a combination of AHP and entropy weighting can be used to determine the preset weights of each dimension to further reduce subjectivity. The entropy weighting method relies on the information entropy of the cost data for each dimension to calculate the weights, completely avoiding subjective interference from human experience. If the numerical difference of a certain cost dimension is greater at different locations in the roadway, it indicates that the dimension has a more prominent effect on path selection, and the corresponding weight should be higher. Compared to the entropy weighting method, the COP method determines the index weights using the coefficient of variation (the ratio of standard deviation to mean) of the cost of each dimension. It does not require calculating information entropy and a weight matrix, making the calculation process simpler and suitable for embedded deployment scenarios with high real-time requirements. Its core idea is consistent with the entropy weighting method: the greater the data fluctuation of a dimension, the higher its weight. The AHP-entropy weighting combination method integrates the subjective weights of the Analytic Hierarchy Process (AHP). Objective weights of the entropy weight method To balance expert experience with data objectivity, two common methods are linear and weighted normalization, and multiplicative normalization. , .

[0063] Step S1024: Obtain the comprehensive cost map based on the comprehensive cost of each grid.

[0064] In this embodiment, the output comprehensive cost map is a dense representation of full-stack security semantics, which also contains triple physical constraint information of slope, roughness, and communication, and retains the spatial resolution of the original DEM. It supports high-precision local obstacle avoidance and provides a continuous, smooth, and jump-free optimized target field for subsequent hierarchical search.

[0065] Step S103: Perform a hierarchical security degradation path search based on the comprehensive cost map to obtain the planned path for the target area.

[0066] In this embodiment, in a real mine roadway environment, due to the veto mechanism in step S1021 and the narrowness of the roadway space, multiple adjacent grids may be simultaneously marked as impassable (comprehensive cost = preset cost), thus forming a continuous veto area. When this area lies precisely between the starting point and the target, the path search algorithm may return no feasible path, causing the system to get stuck. To solve this problem, a hierarchical safety degradation search strategy is adopted, designing the path search process as an N-level progressive structure, with each level corresponding to a different safety tolerance. The highest safety level is prioritized, and when no path is found, the level is downgraded step by step. At the same time, degradation information is recorded for decision-making reference, and finally, a planned path is obtained.

[0067] Please see Figure 3 In one specific embodiment, step S103 includes steps S1031 to S1033, and each step is described in detail below.

[0068] Step S1031: Perform path search based on the comprehensive cost map; if a feasible path is found, use the feasible path as the planned path; if no feasible path is found, set the preset passage threshold as the second threshold, and re-execute the step of determining whether the communication slope cost or the communication signal cost of the current grid is greater than or equal to the preset passage threshold for each grid corresponding to the digital elevation raster map, so as to obtain a new comprehensive cost map.

[0069] In this embodiment, a first-level path search is performed on the comprehensive cost map based on a preset path search algorithm.

[0070] If a path is found, a feasible path is successfully found and is used as the planned path. If no feasible path is found, a preset passage threshold is set as the second threshold, and step S1021, which involves determining whether the communication slope cost or communication signal cost of the current grid cell is greater than or equal to the preset passage threshold for each grid cell corresponding to the digital elevation raster map, is executed again to obtain a new comprehensive cost map. The second threshold can be 0.9999, and this embodiment does not limit this value.

[0071] Step S1032: Continue path search based on the new comprehensive cost map; if a feasible path is found, use the feasible path as the planned path and generate downgrade search prompt information and critical area information; if no feasible path is found, adjust the comprehensive cost map based on the preset penalty coefficient to obtain the adjusted comprehensive cost map.

[0072] In this embodiment, a second-level path search is performed on the new comprehensive cost map based on a preset path search algorithm, allowing passage through critical rejection zones (i.e., grids with comprehensive costs between a first threshold and a second threshold). This level of search still prioritizes bypassing rejection zones, only traversing critical zones when there is no other way to proceed. If a path is found, a feasible path is successfully found, and the currently found feasible path is used as the planned path. A degraded search prompt message indicating a degraded warning is generated, along with critical zone information containing the number and location of critical zones traversed by the planned path.

[0073] If no feasible path is found, the comprehensive cost of the rejected grid in the comprehensive cost map is adjusted based on the preset penalty coefficient to obtain the adjusted comprehensive cost map.

[0074] Step S1033: Continue path search based on the adjusted comprehensive cost map to obtain the planned path.

[0075] In this embodiment, a third-level path search is performed on the adjusted comprehensive cost map based on a preset path search algorithm to obtain the final planned path.

[0076] It should be noted that the preset path search algorithm can include Dijkstra's algorithm, A* algorithm, and RRT algorithm, etc., to compare and verify the effectiveness of the comprehensive cost map. Specifically, the A* algorithm uses the Euclidean distance heuristic function, 8-neighborhood expansion, and the movement cost is the Euclidean distance between adjacent pixels plus the target pixel cost multiplied by a coefficient of 10. The Dijkstra algorithm removes the heuristic function (f=g). The RRT algorithm uses 5% target bias sampling, a step size of 5 pixels, and a maximum of 5000 iterations.

[0077] Please see Figure 4In one specific embodiment, step S1032 includes steps S401 to S404, and each step is described in detail below.

[0078] Step S401: Determine at least one grid cell in the comprehensive cost map whose comprehensive cost is a preset cost, and use it as a candidate grid cell.

[0079] In this embodiment, at least one grid cell in the comprehensive cost map whose comprehensive cost is a preset cost is obtained as a candidate grid cell, forming a set of objects for soft penalty. This avoids unnecessary disturbance to grid cells that have already met the safety conditions, ensuring the targeted nature of cost adjustment and the principle of minimal intervention.

[0080] Step S402: For each candidate grid cell, based on preset weights, the original cost of the candidate grid cell is obtained by weighted summation of the traffic slope cost, the surface roughness cost, and the communication signal cost.

[0081] In this embodiment, for each candidate grid, the comprehensive cost is calculated as the original cost of the candidate grid according to the comprehensive cost calculation formula in step S1023.

[0082] Step S403: Calculate the first difference between the original cost and the first threshold, and calculate the second difference between the preset value and the first threshold; calculate the ratio of the first difference to the second difference, and add the product of the ratio and the preset penalty coefficient to the preset value to obtain the soft penalty value.

[0083] In this embodiment, the soft penalty value The calculation formula is: In the formula, λ is the preset penalty coefficient (default λ=10). The first threshold, This represents the original cost before the grid was rejected. Even with the rejection lifted, there is a strong tendency to detour, only traversing the rejected area when there is absolutely no other way. The search objective becomes the path with the minimum cumulative rejection cost.

[0084] Understandably, a veto power ensures no risk is overlooked, tiered demotion prevents deadlocks, and risk reporting ensures no hidden dangers are concealed. These three elements work together to guarantee the feasibility and safety of path planning for humanoid robots in extreme mine tunnel environments.

[0085] Step S404: Adjust the comprehensive cost of candidate grids in the comprehensive cost map to a soft penalty value to obtain the adjusted comprehensive cost map.

[0086] In this embodiment, the comprehensive cost (preset cost) of the candidate grids in the comprehensive cost map is adjusted to a soft penalty value to obtain the adjusted comprehensive cost map. Its cost distribution exhibits continuous, monotonous, and physically interpretable characteristics. The cost value of each adjusted grid directly corresponds to its "excess severity × penalty intensity". The path generated by the path planner naturally carries risk gradient information, providing a real-time basis for the robot to autonomously adjust motion parameters.

[0087] Understandably, the hierarchical safety degradation search mechanism addresses the practical engineering problem of path breakage caused by a veto. In narrow mine tunnels, consecutive veto zones can block all feasible paths, causing traditional path search algorithms to return no solution. This embodiment provides a systematic fallback path while prioritizing safety through a three-level progressive degradation strategy. This mechanism forms a synergistic relationship of constraint and degradation with the veto: the veto identifies risks, hierarchical degradation addresses unsolvable dilemmas caused by risks, and risk reports transparently transmit degradation information to operators. Both are indispensable; without degradation, the system will completely fail in extreme scenarios; without veto, the system will lose its safety baseline. This collaborative design fundamentally guarantees the availability and safety of the path planning system under various working conditions.

[0088] It should be noted that the number of degradation layers N can be expanded to any configurable number of layers; in addition to no path, the degradation trigger condition can also be set to automatically trigger when the risk indicators of all paths exceed the threshold; in addition to linear penalty, the degradation cost function can adopt exponential penalty, logarithmic penalty, etc.; the preset penalty coefficient λ can also be dynamically adjusted according to the mine safety level.

[0089] The path planning method proposed in this embodiment calculates the travel slope cost, surface roughness cost, and communication signal cost based on the digital elevation raster map of the target area; determines a comprehensive cost map based on these costs; and performs a hierarchical safety degradation path search based on the comprehensive cost map to obtain the planned path for the target area. In this way, by integrating the costs of slope, roughness, and communication to construct a comprehensive cost map and introducing a hierarchical safety degradation path search mechanism, safe paths are planned for the target area, avoiding extreme slopes, high-roughness surfaces, and communication blind spots. This significantly improves the safety and robustness of path planning, both rigidly avoiding single-point fatal risks such as overturning, instability, and disconnection, and providing step-by-step controllable degradation when no feasible path is available, ensuring task continuity. Simultaneously, it outputs interpretable risk gradient information to support real-time motion decision-making and manual intervention.

[0090] Example 2 Furthermore, this disclosure provides a path planning system 500, please refer to [link to relevant documentation]. Figure 5 The system includes: The calculation module 501 is used to calculate the traffic slope cost, surface roughness cost and communication signal cost based on the digital elevation raster map of the target area. Module 502 is used to determine a comprehensive cost map based on the traffic slope cost, surface roughness cost, and communication signal cost. Search module 503 is used to perform hierarchical security degradation path search based on the comprehensive cost map to obtain the planned path for the target area.

[0091] In an optional implementation, the determining module 502 is further configured to set a preset passage threshold as a first threshold, and for each grid corresponding to the digital elevation raster map, determine whether the communication slope cost or the communication signal cost of the current grid is greater than or equal to the preset passage threshold; if the communication slope cost or the communication signal cost of the current grid is greater than or equal to the preset passage threshold, then the comprehensive cost of the current grid is the preset cost; if both the communication slope cost and the communication signal cost of the current grid are less than the preset passage threshold, then based on preset weights, the passage slope cost, the surface roughness cost, and the communication signal cost of the current grid are weighted and summed to obtain the comprehensive cost of the current grid; and a comprehensive cost map is obtained based on the comprehensive costs of each grid.

[0092] In an optional implementation, the search module 503 is further configured to perform path search based on the comprehensive cost map; if a feasible path is found, it is used as the planned path; if no feasible path is found, a preset passage threshold is set as a second threshold, and the step of determining whether the communication slope cost or communication signal cost of the current grid is greater than or equal to the preset passage threshold is re-executed for each grid corresponding to the digital elevation raster map, to obtain a new comprehensive cost map; path search continues based on the new comprehensive cost map; if a feasible path is found, it is used as the planned path, and downgraded search prompt information and critical area information are generated; if no feasible path is found, the comprehensive cost map is adjusted based on a preset penalty coefficient to obtain an adjusted comprehensive cost map; path search continues based on the adjusted comprehensive cost map to obtain the planned path.

[0093] In an optional implementation, the search module 503 is further configured to determine at least one grid cell in the comprehensive cost map whose comprehensive cost is a preset cost, as a candidate grid cell; for each candidate grid cell, based on a preset weight, a weighted sum is performed according to the candidate grid cell's traffic slope cost, the candidate grid cell's surface roughness cost, and the candidate grid cell's communication signal cost to obtain the candidate grid cell's original cost; a first difference between the original cost and a first threshold is calculated, and a second difference between a preset value and the first threshold is calculated; the ratio of the first difference to the second difference is calculated, and the product of the ratio and a preset penalty coefficient is added to the preset value to obtain a soft penalty value; the comprehensive cost of the candidate grid cells in the comprehensive cost map is adjusted to the soft penalty value to obtain an adjusted comprehensive cost map.

[0094] In an optional implementation, the calculation module 501 is further configured to calculate the X-direction gradient and the Y-direction gradient of each grid in the digital elevation grid map; calculate the slope angle of each grid based on the X-direction gradient and the Y-direction gradient of each grid; and map the slope angle of each grid to the access slope cost of each grid.

[0095] In an optional implementation, the calculation module 501 is further configured to calculate the local elevation mean of each grid in the digital elevation raster map; calculate the variance of each grid based on the local elevation mean of each grid; and map the variance of each grid to the surface roughness cost of each grid.

[0096] In an optional implementation, the calculation module 501 is further configured to calculate the planar distance from the preset communication base station to each grid in the digital elevation grid map based on the digital elevation grid map; calculate the planar distance from the preset communication base station to each grid based on the logarithmic distance path loss model to obtain the base station received power corresponding to each grid; attenuate the base station received power corresponding to the grid to be processed for the grid whose elevation value is greater than the preset elevation threshold; and map the base station received power corresponding to each grid to the communication signal cost of each grid.

[0097] The system provided in this embodiment can execute the steps of the path planning method provided in Embodiment 1. To avoid repetition, these steps will not be repeated.

[0098] The path planning system proposed in this embodiment calculates the travel slope cost, surface roughness cost, and communication signal cost based on the digital elevation raster map of the target area; determines a comprehensive cost map based on the travel slope cost, surface roughness cost, and communication signal cost; and performs a hierarchical safety degradation path search based on the comprehensive cost map to obtain the planned path for the target area. In this way, by integrating the costs of slope, roughness, and communication to construct a comprehensive cost map and introducing a hierarchical safety degradation path search mechanism, it plans safe paths for the target area that avoid extreme slopes, high-roughness surfaces, and communication blind spots, significantly improving the safety and robustness of path planning. It not only rigidly avoids single-point fatal risks such as overturning, instability, and disconnection, but also provides a step-by-step controllable degradation when no feasible path is available, ensuring task continuity. Simultaneously, it outputs interpretable risk gradient information to support real-time motion decision-making and manual intervention.

[0099] Example 3 Furthermore, this disclosure provides a computer device including a memory and a processor, wherein the memory stores a computer program, and the computer program, when executed by the processor, implements the path planning method described in Embodiment 1.

[0100] The device provided in this embodiment can execute the steps of the path planning method provided in Embodiment 1. To avoid repetition, it will not be described again.

[0101] Example 4 This disclosure provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the path planning method described in Embodiment 1.

[0102] In this embodiment, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0103] The computer-readable storage medium provided in this embodiment can implement the path planning method provided in Embodiment 1. To avoid repetition, it will not be described again here.

[0104] In all examples shown and described herein, any specific values ​​should be interpreted as merely exemplary and not as limitations; therefore, other examples of exemplary embodiments may have different values.

[0105] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0106] The above-described embodiments are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention.

Claims

1. A path planning method, characterized in that, The method includes: Calculate the traffic slope cost, surface roughness cost, and communication signal cost based on the digital elevation raster map of the target area; A comprehensive cost map is determined based on the traffic slope cost, the surface roughness cost, and the communication signal cost. Based on the comprehensive cost map, a hierarchical security degradation path search is performed to obtain the planned path for the target area.

2. The path planning method according to claim 1, characterized in that, The process of determining the comprehensive cost map based on the traffic slope cost, the surface roughness cost, and the communication signal cost includes: Set a preset passage threshold as the first threshold, and for each grid corresponding to the digital elevation grid map, determine whether the communication slope cost or the communication signal cost of the current grid is greater than or equal to the preset passage threshold. If the communication slope cost or the communication signal cost of the current grid is greater than or equal to the preset passage threshold, then the total cost of the current grid is the preset cost. If the communication slope cost and the communication signal cost of the current grid are both less than the preset passage threshold, then based on the preset weight, the passage slope cost, the surface roughness cost, and the communication signal cost of the current grid are weighted and summed to obtain the comprehensive cost of the current grid. The comprehensive cost map is obtained based on the comprehensive cost of each of the grids.

3. The path planning method according to claim 2, characterized in that, The step of performing a hierarchical security degradation path search based on the comprehensive cost map to obtain the planned path for the target area includes: Path search is performed based on the comprehensive cost map; if a feasible path is found, it is used as the planned path; if no feasible path is found, the preset passage threshold is set to the second threshold, and the step of determining whether the communication slope cost or communication signal cost of the current grid is greater than or equal to the preset passage threshold is re-executed for each grid corresponding to the digital elevation grid map, so as to obtain a new comprehensive cost map. The path search continues based on the new comprehensive cost map. If a feasible path is found, it is used as the planned path, and a downgraded search prompt and critical area information are generated. If no feasible path is found, the comprehensive cost map is adjusted based on a preset penalty coefficient to obtain an adjusted comprehensive cost map. The path search continues based on the adjusted comprehensive cost map to obtain the planned path.

4. The path planning method according to claim 3, characterized in that, The adjustment of the comprehensive cost map based on a preset penalty coefficient to obtain the adjusted comprehensive cost map includes: Identify at least one grid cell in the comprehensive cost map whose comprehensive cost is the preset cost as a candidate grid cell; For each candidate grid, based on the preset weight, the original cost of the candidate grid is obtained by weighted summation of the traffic slope cost, the surface roughness cost, and the communication signal cost of the candidate grid. Calculate the first difference between the original cost and the first threshold, and calculate the second difference between the preset value and the first threshold; calculate the ratio of the first difference to the second difference, and add the product of the ratio and the preset penalty coefficient to the preset value to obtain the soft penalty value; The comprehensive cost of the candidate grid in the comprehensive cost map is adjusted to the soft penalty value to obtain the adjusted comprehensive cost map.

5. The path planning method according to claim 1, characterized in that, The calculation of the access slope cost based on the digital elevation raster map of the target area includes: Calculate the X-direction gradient and the Y-direction gradient of each grid cell in the digital elevation raster map; The slope angle of each grid is calculated based on the X-direction gradient and the Y-direction gradient of each grid. The slope angle of each of the grids is mapped to the passage slope cost of each of the grids.

6. The path planning method according to claim 1, characterized in that, The surface roughness cost is calculated based on the digital elevation raster map of the target area, including: Calculate the mean local elevation of each grid cell in the digital elevation raster map; The variance of each of the grid cells is calculated based on the mean local elevation of each grid cell. The variance of each of the graticules is mapped to the surface roughness cost of each of the graticules.

7. The path planning method according to claim 1, characterized in that, The communication signal cost is calculated based on the digital elevation raster map of the target area, including: Calculate the planar distance from the preset communication base station to each grid in the digital elevation grid map based on the digital elevation grid map; The planar distance from the preset communication base station to each of the grids is calculated based on the logarithmic distance path loss model to obtain the base station received power corresponding to each grid. For grid cells whose elevation values ​​are greater than a preset elevation threshold, the base station received power corresponding to the grid cell is attenuated. The base station received power corresponding to each of the grids is mapped to the communication signal cost of each of the grids.

8. A path planning system, characterized in that, The system includes: The calculation module is used to calculate the access slope cost, surface roughness cost, and communication signal cost based on the digital elevation raster map of the target area; The determination module is used to determine a comprehensive cost map based on the traffic slope cost, the surface roughness cost, and the communication signal cost. The search module is used to perform a hierarchical security degradation path search based on the comprehensive cost map to obtain the planned path for the target area.

9. A computer device, characterized in that, It includes a memory and a processor, the memory storing a computer program that, when executed by the processor, implements the path planning method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, implements the path planning method as described in any one of claims 1 to 7.

Images