A method for group path planning of tracked vehicles fusing beach passability evaluation

Abstract

A kind of fusion beach passability evaluation tracked vehicle group path planning method, the method includes tracked vehicle group passing through beach area, first draw global grid map, the type of beach is identified by first vehicle, draw passability semantic map, and the path of current vehicle is planned, if the bearing capacity of soil can support the passage of tracked vehicle, and there is no collision between vehicle and obstacle on the path, then the path planning is executed;The passability risk of the path of preceding vehicle is evaluated based on the tracked vehicle behind, continue to use the path of preceding vehicle, or update passability semantic map and carry out path re-planning, until all tracked vehicles pass through completely.The method fully considers the change of soil characteristics in the path after the passage of preceding vehicle, realizes the repeatability judgment of path in the process of tracked vehicle group on land, effectively improves the efficiency and safety of tracked vehicle group passing through on beach road surface.

Description

Technical Field

[0001] This invention relates to the field of vehicle control technology, and in particular to a method for path planning of tracked vehicle groups for assessing tidal flat passability. Background Technology

[0002] As a unique natural landform, tidal flats place high demands on vehicle performance due to their distinctive environmental characteristics. Tidal flat surfaces are often soft and have low load-bearing capacity, especially after rain when they become even muddier and unable to support vehicle weight. The low coefficient of friction on soft mud makes vehicles prone to slipping, particularly when climbing slopes or turning, where insufficient tire grip can lead to loss of traction. Furthermore, tidal flat surfaces are unstable and prone to localized collapse or subsidence, especially after repeated compaction by heavy vehicles, altering the surface structure and further reducing passability. Researching path planning methods for tracked vehicle groups on tidal flats can help improve vehicle adaptability. Summary of the Invention

[0003] This disclosure provides a path planning method for tracked vehicle groups that integrates tidal flat passability assessment to address the problem of poor passability caused by repeated compaction of the tidal flat surface during the landing process of tracked vehicle groups. This results in low soil bearing capacity, low friction coefficient, local collapse or subsidence. The method improves the tidal flat passability of tracked vehicle groups in terms of top-level planning.

[0004] The path planning method for tracked vehicle groups disclosed herein includes the following steps:

[0005] S1, Global Grid Map of Tidal Flats: Draws a global grid map of the tidal flat area that a group of tracked vehicles is about to pass through, containing obstacles that need to be avoided in the path planning;

[0006] S2, the vehicle in front identifies the type of mudflat and assesses its carrying capacity; draws a passability semantic map based on the global grid map of the mudflat; plans the path of the current vehicle based on the passability semantic map; if the soil carrying capacity can support the passage of the tracked vehicle and there is no collision between the vehicle and the obstacle on the path, then the path planning is started.

[0007] S3: After the preceding vehicle passes through according to the planned route, the following vehicle assesses the passability risk coefficient of the preceding vehicle's route. If the risk coefficient is lower than the set threshold, the following vehicle does not need to replan the route and continues to use the preceding vehicle's route; otherwise, the following vehicle updates the passability semantic map and replans the route.

[0008] Until all tracked vehicles have passed.

[0009] Furthermore, in step S2, the steps of identifying the type of tidal flat and assessing its carrying capacity specifically include:

[0010] An algorithm with classification capabilities is used to identify the type and water content of tidal flats;

[0011] Based on the identification results, values ​​were assigned to the cohesion and frictional characteristics parameters in the vertical indentation model and horizontal shear model of tidal flat soil.

[0012] Based on the weight of the tracked vehicle and the ground contact area of ​​the track, the soil resistance and track traction force of the vehicle are further calculated.

[0013] When the driving force of the tracked vehicle is greater than the soil traction force and also greater than the soil resistance, it is determined that the current tidal flat bearing capacity can support the tracked vehicle to pass through.

[0014] Furthermore, in step S2, the method for drawing a transitivity semantic map based on the global raster map of the tidal flats includes:

[0015] A hazard coefficient dimension is added to the global raster 2D map of the tidal flats. A hazard coefficient of 1 indicates an impassable area or obstacle, while a hazard coefficient of 0 indicates that there is no hazard in the passable area.

[0016] Furthermore, in step S2, the step of planning the current vehicle's path based on the passability semantic map uses A-Star or RRT path planning algorithms for global path planning.

[0017] Furthermore, step S3 specifically includes:

[0018] S31: Historical Path Identification: Before planning the path of the following vehicle, the path coordinates of the preceding vehicle are identified in the grid map to form a continuous path that the preceding vehicle has already executed.

[0019] S32: Passability Risk Assessment:

[0020] As the surface shear strength of the tidal flat soil gradually decreases while soil resistance gradually increases during repeated compaction along the planned route, the ratio of the difference between the soil traction force and soil resistance of the following vehicle and the preceding vehicle is used as the passability risk coefficient of the route after the preceding vehicle passes. This quantifies the passability risk of the route. The passability risk coefficient is assessed starting from the second vehicle after the first vehicle passing along the current route. The assessment model is as follows:

[0021]

[0022] In the formula, ξ is the passability hazard coefficient; η is the safety margin coefficient, which is 1.5 if the soil is judged to be cohesive, and 1.2 if it is judged to be frictional soil; Ft For the soil traction of tracked vehicles, F r The soil longitudinal resistance; i indicates the nth vehicle passing on the current path;

[0023] When the passability risk factor is greater than the set threshold, it is determined that the current path cannot support the passage of the following vehicle. At this time, the following vehicle continues to execute steps S33 and S34 to replan the path and find a passable path again.

[0024] S33: Update the accessibility semantic map:

[0025] Historical path markers are overlaid on the initial passability semantic map, and passability risk coefficients are marked on the paths. The risk coefficient represents the risk of continuing to travel along the path, and the passability risk coefficient ranges from 0 to 1. As a group of tracked vehicles continues to pass through a path, the passability risk coefficient gradually increases. When the passability risk coefficient is greater than a set threshold, it is considered that the path can no longer support the passage of the next vehicle. In the next semantic map update, the risk coefficient of the path is updated to 1, that is, it is identified as an obstacle or an impassable area.

[0026] S34: The following vehicle re-plans and executes its route based on the updated passability semantic map using the method in step S2.

[0027] Compared with the prior art, the beneficial effects of this disclosure are: ① It proposes a raster map semantic representation method and update process that combines mechanical information; ② It considers the changes in soil properties in the path after the preceding vehicle passes, and realizes the path repeatability judgment during the landing process of tracked vehicle groups; ③ It improves the efficiency and safety of tracked vehicle groups passing through tidal flats. Attached Figure Description

[0028] The above and other objects, features and advantages of this disclosure will become more apparent from the more detailed description of exemplary embodiments of this disclosure taken in conjunction with the accompanying drawings, in which the same reference numerals generally represent the same components.

[0029] Figure 1 This is a general flowchart of an exemplary embodiment according to the present disclosure;

[0030] Figure 2 A flowchart for evaluating the beach passability of tracked vehicles as an exemplary embodiment;

[0031] Figure 3 A flowchart for updating the passability semantic map as an exemplary embodiment;

[0032] Figure 4 A vehicle tidal flat traversal hazard factor diagram for an exemplary embodiment;

[0033] Figure 5 A semantic map illustration of fused tidal flat passability as an exemplary embodiment;

[0034] Figure 6 A schematic diagram of tidal flat path planning for an exemplary embodiment, (a) first path planning, (b) second path replanning, (c) third path replanning, and (d) fourth path replanning;

[0035] Figure 7 For an exemplary embodiment of the semantic map oriented towards mudflat accessibility, (a) drawing the semantic map, (b) first semantic map update, (c) second semantic map update, and (d) third semantic map update. Detailed Implementation

[0036] Preferred embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While preferred embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that the present disclosure will be thorough and complete, and will fully convey the scope of the present disclosure to those skilled in the art.

[0037] This disclosure provides a method for path planning of tracked vehicle groups that integrates tidal flat passability assessment. In one exemplary embodiment, the process is as follows: Figure 1 As shown, it includes the following steps S1-S8:

[0038] 1. S1: Global raster map of the mudflats;

[0039] In this embodiment, a global grid map is drawn showing the tracked vehicle group passing through the tidal flat area, containing obstacles that need to be avoided in the path planning;

[0040] The starting and ending points of the tracked vehicle group are set according to actual needs. The starting position of the tracked vehicle group is set according to the actual position. In this embodiment, grid (1,1) is set as the starting point and grid (30,30) is set as the ending point.

[0041] 2. S2: Assessment of the carrying capacity of tidal flats:

[0042] In this embodiment, a deep residual network ResNet 18 (any algorithm with classification function can be used) is applied to classify and identify the type and water content level of the tidal flats;

[0043] Based on the identification results, the subsidence capacity and shear resistance of the tidal flats are calibrated by using a lookup table method, and then the ground resistance and ground traction force during the movement of the tracked vehicle are calculated.

[0044] The current mudflat crossing capability of a vehicle is determined based on ground resistance, ground traction, and tracked vehicle driving force.

[0045] In this embodiment, the carrying capacity of the tidal flat needs to be assessed before the tracked vehicle group path planning. First, the ResNet18 network or other algorithms with classification functions are used to identify the type and water content of the tidal flat. Based on the identification results, the cohesion and friction characteristics parameters in the vertical indentation model and horizontal shear model of the tidal flat soil are assigned values. Combined with the weight of the tracked vehicle and the track ground contact area, the soil resistance and track soil traction force of the vehicle are further calculated. When the driving force of the tracked vehicle is greater than the soil traction force and also greater than the soil resistance, it is determined that the current tidal flat carrying capacity can support the passage of the tracked vehicle, thereby completing the assessment of the tracked vehicle tidal flat passage performance.

[0046] like Figure 2 As shown, the specific steps are S21-S27:

[0047] S21: Tidal Flat Type Identification:

[0048] In this embodiment, the ResNet 18 deep residual network (any algorithm with classification function can be used) is first applied to identify the type of tidal flat. The tidal flat types are divided into frictional soil and cohesive soil, with sandy beaches being frictional soil and silt being cohesive soil.

[0049] S22: Moisture content calibration of tidal flats:

[0050] In this embodiment, the moisture content of the tidal flats is first identified by the deep residual network ResNet18 (any algorithm with classification function can be used). The moisture content of the tidal flats is divided into three levels: low, medium and high.

[0051] The recommended ranges for low, medium, and high moisture content of frictional soils such as sand are 3%, 10%, and 25%, respectively; the recommended ranges for low, medium, and high moisture content of cohesive soils such as silt are 15%, 30%, and 60%, respectively.

[0052] S23: Vertical indentation capacity calibration of tidal flats:

[0053] In this embodiment, soil indentation capacity is related to both soil type and moisture content, and the indentation model can be expressed as:

[0054]

[0055] In the formula: k c k is the cohesive modulus. φ denoted as the frictional deformation modulus; b is the track width; L is the track length; n is the settlement index; and z is the settlement amount.

[0056] For frictional soils such as coarse and fine sand, the recommended cohesive deformation modulus for low, medium, and high moisture contents is 3.7e-2 MPa / m³. n+1 3.5e-2MPa / m n+1 3.4e-2MPa / m n+1 The recommended value for the frictional deformation modulus is 1.4e-3 MPa / m. n+2 1.1e-3MPa / m n+2 0.8e-3MPa / m n+2 The recommended values ​​for the subsidence index are 0.4, 0.6, and 0.8.

[0057] For cohesive soils such as silt, the recommended cohesive deformation modulus for low, medium, and high moisture contents is 0.41 MPa / m. n+1 3.17MPa / m n+1 0.1MPa / m n+1 The recommended value for the frictional deformation modulus is 2.1e-2 MPa / m. n+2 43e-2MPa / m n+2 0.5e-2MPa / m n+2 The recommended values ​​for the subsidence index are 0.4, 0.8, and 1.4.

[0058] S24: Calculation model for resistance of tracked vehicles traveling on mudflats:

[0059] In this embodiment, the tidal flats are characterized by soft soil and significant subsidence. The bulldozing resistance in front of the tracks will account for the majority of the vehicle's walking resistance, which can be expressed as:

[0060]

[0061] In the formula: F r ρ is the longitudinal resistance of the soil; c is the soil cohesion; ρ is the soil density; φ is the soil internal friction angle.

[0062] For frictional soils such as coarse and fine sand, the recommended values ​​for soil cohesion at low, medium, and high moisture content are 3.3 kPa, 3.1 kPa, and 2.8 kPa, respectively, and the recommended values ​​for soil internal friction angle are 34°, 30°, and 28°, respectively.

[0063] For cohesive soils such as silt, the recommended values ​​for soil cohesion at low, medium, and high moisture content are 8 kPa, 15 kPa, and 5 kPa, respectively, and the recommended values ​​for soil internal friction angle are 8°, 10°, and 3°, respectively.

[0064] S25: Horizontal shear capacity calibration of tidal flats:

[0065] In this embodiment, soil shear resistance refers to the soil's ability to resist shear stress without failure during the shearing process between the track and the soil. Tidal flats are loose soils, and disturbed soil can be used to describe the shear stress resistance of tracked vehicles.

[0066] τ=(c+ptanφ)(1-e -j(x) / K )

[0067] In the formula: τ is the track shear stress; p is the normal pressure of the tracked vehicle; x is the distance between the contact point and the front end of the track; and K is the shear displacement corresponding to the occurrence of the maximum shear stress.

[0068] S26: Traction calculation model for tracked vehicles traveling on mudflats:

[0069] In this embodiment, the soil traction force of the tracked vehicle can be obtained by integrating over the track length based on the soil shear stress model of the tracked vehicle:

[0070]

[0071] In the formula: F t It is the driving force for the tracks.

[0072] S27: Performance evaluation of tidal flats:

[0073] In this embodiment, when the driving force of the tracked vehicle is greater than the soil traction force and also greater than the soil resistance, it is determined that the current tidal flat bearing capacity can support the passage of the tracked vehicle.

[0074] F m >F t >F r

[0075] Among them, F m The maximum driving force provided for tracked vehicles.

[0076] 3. S3: Throughness semantic map drawing / updating:

[0077] In this embodiment, by drawing a semantic map of the tidal flats and relying on a global grid map of the tidal flats, obstacles are marked, and a hazard factor dimension is added to the two-dimensional global grid map of the tidal flats, such as... Figure 4 As shown in the diagram. A danger coefficient of 1 indicates an impassable area, while a danger coefficient of 0 indicates that the area is passable without danger. The initial semantic map is as follows: Figure 5 As shown, the area with a danger level of 1 is an obstacle.

[0078] After the preceding vehicle passes along the planned route, the following vehicle needs to assess the soil shear conditions of that route, predict the reduction in the tidal flat passage capacity of that route, and mark the risk factor of that route. The risk factor ranges from 0 to 1, with a higher risk factor indicating poorer passability.

[0079] Therefore, in this embodiment, each tracked vehicle in the vehicle group needs to record its map coordinates during the path planning process. Before the following vehicle passes, the path coordinates of the preceding vehicle need to be marked in the grid map to form a continuous path already executed by the preceding vehicle. As the surface shear strength of the tidal flat soil gradually decreases while the soil resistance gradually increases during the repeated compaction of the planned path by the tracked vehicle group, it is necessary to use the ratio of the difference between the soil traction force and soil resistance of the following and preceding vehicles as the risk coefficient of the path to quantify the passability risk of the path (the risk coefficient represents that there is a certain risk in continuing to travel along the path, and the passability risk coefficient ranges from 0 to 1. As the tracked vehicle group continues to pass on a path, the passability risk coefficient gradually increases). When the passability risk coefficient is greater than the set threshold of 0.6, it is considered that the path can no longer support the passage of the next vehicle. The following vehicle needs to update the passability semantic map, replan the path, and find a new passable path on the tidal flat. When the semantic map is updated, the risk coefficient of the path is updated to 1, that is, it is identified as an obstacle.

[0080] The above process for updating the passivity semantic map is as follows: Figure 3 As shown, the specific steps are S31-S33:

[0081] S31: Historical Path Identifier:

[0082] In this embodiment, each tracked vehicle needs to record its map coordinates during path planning; before planning the path for the next vehicle, the path coordinates of the preceding vehicle need to be marked on the grid map to form a continuous path already executed by the preceding vehicle, such as... Figure 6 As shown, where, Figure 6 'a' represents the route of the first vehicle after the first vehicle has completed its route; Figure 6 b represents the path of the second vehicle as identified by the third vehicle after the second vehicle has completed its path execution. Figure 6 c represents the path of the third vehicle identified by the fourth vehicle after the third vehicle has completed its path.

[0083] S32: Passability Risk Assessment:

[0084] In this embodiment, during the repeated compaction of the tracked vehicle group's planned path, the surface shear strength of the tidal flat soil gradually decreases, while the soil resistance gradually increases. The ratio of the difference between the soil traction force and soil resistance of the following and preceding vehicles is used as the tidal flat risk factor to quantify the passability risk of the path. The passability risk factor is assessed starting from the second vehicle preparing to pass along the current path. The assessment model is as follows:

[0085]

[0086] In the formula, η is the safety margin coefficient. If the soil is judged to be cohesive, then η = 1.5; if the soil is judged to be frictional, then η = 1.2. When the passability risk coefficient is greater than 0.6, it is determined that the current path cannot support the passage of the following vehicle. At this time, the following vehicle needs to replan its route and find a new passable path.

[0087] The path hazard assessment results in the semantic map are derived from... Figure 6 As shown, Figure 6 'a' represents the path risk coefficient of the second vehicle after the first vehicle completes its route, which is 0.26 according to the prediction model. Figure 6 b represents the path hazard coefficient of the third vehicle after the second vehicle completes its route, which is 0.45 according to the prediction model. Figure 6 c represents the path hazard coefficient of the fourth vehicle after the third vehicle completes its route. The prediction model estimates it to be 0.62. Since the passability hazard coefficient is greater than 0.6, it means that the current path can no longer support the passage of the fourth vehicle. The fourth vehicle needs to replan its route and find a new passable path on the mudflats.

[0088] S33: Update the passability semantic map:

[0089] In this embodiment, a global grid map of the tidal flats with tackling obstacles is used as the basis, where obstacles are marked as 1, i.e., impassable areas, and other areas are marked as 0, i.e., passable areas.

[0090] Historical path markers are overlaid, and a passability risk coefficient is marked on the path. The risk coefficient represents the risk of continuing to travel along the path, and the passability risk coefficient ranges from 0 to 1. As a group of tracked vehicles continues to pass through a path, the passability risk coefficient gradually increases. When the passability risk coefficient is greater than 0.6, it is considered that the path can no longer support the passage of the next vehicle. In the next semantic map update, the risk coefficient of the path is updated to 1, which means it is identified as an obstacle.

[0091] 4. S4: Tracked vehicle path planning / replanning:

[0092] In this embodiment, the first vehicle's route planning is performed using the A-Star route planning algorithm (which can be implemented using other route planning methods such as RRT) based on the passability semantic map.

[0093] When replanning the route for tracked vehicles, the risk factor of the previous planned route must be considered. If the risk factor is low, it means that the route is not dangerous and the soil bearing capacity can support passage, so the following vehicle does not need to replan the route.

[0094] If the risk factor is high, it means that the soil capacity of the route cannot support the passage of subsequent vehicles, and the route needs to be replanned.

[0095] Depend on Figure 7 As shown, Figure 7 'a' represents the planned route for the first vehicle; Figure 7 b means that after the first vehicle passes through, the semantic map determines that the risk coefficient of this path is 0.26, which is a low risk coefficient. This path can continue to be used, so the second vehicle does not need to replan its path. Figure 7 c represents the path risk coefficient of 0.45 determined by the semantic map after the second vehicle passes. This is considered a medium risk coefficient, indicating that the path has a passability risk but can still be used. Therefore, the third vehicle does not need to replan its route. Figure 7 After the third vehicle passes, the semantic map determines that the risk coefficient of this path is 0.62, which is a high risk coefficient. This path has a significant risk to its passability. The original path is determined to be impassable, so the fourth vehicle needs to replan its route.

[0096] 5. S5: Determine if the planned path is passable:

[0097] In this embodiment, the soil carrying capacity of the planned route is checked again to see if it is passable. If it is determined that it is passable, the next step is executed; if it is determined that it is not passable, the route is replanned.

[0098] 6. S6: Determine if the planned path collides.

[0099] In this embodiment, the planned path is checked again to see if it collides with obstacles. If no collision is found, the next step is executed; if a collision is found, the path is replanned.

[0100] 7. S7: Did the entire group pass?

[0101] In this embodiment, it is determined whether the entire group has passed through. If it is determined that the entire group has not passed through, the following vehicles need to reassess the soil carrying capacity and replan the route. If it is determined that the entire group has passed through, the route planning for the tracked vehicles in the tidal flat environment is completed.

[0102] The above technical solutions are merely exemplary embodiments of the present invention. For those skilled in the art, based on the application methods and principles disclosed in the present invention, it is easy to make various types of improvements or modifications, and not limited to the methods described in the specific embodiments of the present invention. Therefore, the methods described above are merely preferred and not restrictive.

Claims

1. A method for path planning of tracked vehicle groups that integrates tidal flat passability assessment, comprising the following steps: S1, Global Grid Map of Tidal Flats: Draws a global grid map of the tidal flat area that a group of tracked vehicles is about to pass through, containing obstacles that need to be avoided in the path planning; S2, the vehicles at the front of the line identify the type of mudflats and assess their carrying capacity; a transit semantic map is drawn based on the global raster map of the mudflats; Based on the passability semantic map, the current vehicle's path is planned; if the soil's carrying capacity can support the passage of the tracked vehicle, and there is no collision between the vehicle and the obstacle on the path, then the path planning is started. S3: After the preceding vehicle passes through according to the planned route, the following vehicle assesses the passability risk factor of the preceding vehicle's route. If the risk factor is lower than the set threshold, the following vehicle does not need to replan the route and continues to use the preceding vehicle's route. Otherwise, the following vehicle updates the passability semantic map and performs route replanning; Until all tracked vehicles have passed; Step S3 specifically includes: S31: Historical Path Identification: Before planning the path of the following vehicle, the path coordinates of the preceding vehicle are identified in the grid map to form a continuous path that the preceding vehicle has already executed. S32: Passability Risk Assessment: The ratio of the difference in soil traction and soil resistance between the following vehicle and the preceding vehicle is used as the passability risk coefficient of the path after the preceding vehicle passes, quantifying the passability risk of the path. The passability risk coefficient is assessed starting from the second vehicle after the first vehicle passing along the current path. The assessment model is as follows: In the formula, ξ is the passability risk factor; η As a safety margin factor, if the soil is determined to be cohesive, then η =1.5, if it is determined to be frictional soil, then η =1.2; F t For the soil traction of tracked vehicles, F r The longitudinal resistance of the soil; i indicates the nth vehicle passing on the current path; When the passability risk factor is greater than the set threshold, it is determined that the current path cannot support the passage of the following vehicle. At this time, the following vehicle continues to execute steps S33 and S34 to replan the path and find a passable path again. S33: Update the accessibility semantic map: Historical path markers are overlaid on the initial passability semantic map, and passability risk coefficients are marked on the paths. The risk coefficient represents the risk of continuing to travel along the path, and the passability risk coefficient ranges from 0 to 1. As a group of tracked vehicles continues to pass through a path, the passability risk coefficient gradually increases. When the passability risk coefficient is greater than a set threshold, it is considered that the path can no longer support the passage of the next vehicle. In the next semantic map update, the risk coefficient of the path is updated to 1, that is, it is identified as an obstacle or an impassable area. S34: The following vehicle re-plans and executes its route based on the updated passability semantic map using the method in step S2.

2. The method according to claim 1, characterized in that, In step S2, the steps of identifying the type of tidal flat and assessing its carrying capacity include the following specific methods: An algorithm with classification capabilities is used to identify the type and water content of tidal flats; Based on the identification results, values ​​were assigned to the cohesion and frictional characteristics parameters in the vertical indentation model and horizontal shear model of tidal flat soil. Based on the weight of the tracked vehicle and the ground contact area of ​​the track, the soil resistance and track traction force of the vehicle are further calculated. When the driving force of the tracked vehicle is greater than the soil traction force and also greater than the soil resistance, it is determined that the current tidal flat bearing capacity can support the tracked vehicle to pass through.

3. The method according to claim 1 or 2, characterized in that, In step S2, the method for drawing a transit semantic map based on the global raster map of the tidal flats includes: A hazard coefficient dimension is added to the global raster 2D map of the tidal flats. A hazard coefficient of 1 indicates an impassable area or obstacle, while a hazard coefficient of 0 indicates that there is no hazard in the passable area.

4. The method according to claim 1, characterized in that, In step S2, the step of planning the current vehicle's path based on the passability semantic map uses A-Star or RRT path planning algorithms for global path planning.

Images