A vehicle adaptive intelligent guidance method and system based on tunnel positioning

Abstract

The application discloses a kind of vehicle adaptive intelligent guidance method and system based on tunnel positioning, it is related to vehicle adaptive guidance technical field, the present application includes ST1, by tunnel control platform sends task instruction to several mobile vehicles and parses task instruction, according to the initial position information of several mobile vehicles and target point position information in parsing task instruction, the optimal execution path of several mobile vehicles is automatically generated;ST2, each mobile vehicle real-time positioning and perception, update high-precision local map and semantic map, and utilize the sensor return data of each mobile vehicle to carry out local path planning and obstacle avoidance, the present application updates high-precision local map and semantic map in real time in the movement process of each mobile vehicle, and in local path planning and obstacle avoidance process, the state of super under-excavation area, geology change point, face and support section surrounding rock in tunnel is introduced as constraint condition, and truly flows into vehicle control closed loop.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle adaptive guidance, in particular to a vehicle adaptive intelligent guidance method and system based on tunnel positioning. BACKGROUND

[0002] The space of a tunnel construction site is often limited. In the limited space, engineering vehicles shuttle like a loom. The overhead is unhardened rock mass, and the foot is a rugged and uneven road surface. The traditional guidance relies on manual scheduling and fixed sign management mode, which is increasingly unable to meet the needs in this dynamic and high-risk environment. Introducing a working vehicle adaptive guidance system is the core support for ensuring life and production safety, improving efficiency, and promoting engineering lean.

[0003] The prior art, such as the invention application patent published with the publication number CN118800050B, discloses a generic vehicle road coordination method and system based on roadside perception and visual guidance. The current vehicle speed, position, and distance information are used to achieve fine and adaptive controllable characteristics of key channel scenes, and to improve road flow stability and safety. The invention application patent published with the publication number CN107605219B discloses a self-adaptive indoor parking navigation and automatic parking system and method based on BLE. Through the research on the adaptive multi-space scale vehicle position positioning strategy, the parking process is more fine and quantitative, which lays a foundation for subsequent more automated and intelligent indoor reverse vehicle search and automatic parking business.

[0004] In combination with the above-mentioned scheme, it can be found that the existing technology still has the following defects. The guidance in the existing technology is mostly in the form of from A to B, which is difficult to realize real-time, accurate, and semantic perception and map construction of the construction environment, and to drive the guidance decision. The attention to the state of the overbreak and underbreak area, the geological change point, the working face, and the surrounding rock of the support section in the tunnel is not high, and it does not really flow into the vehicle control closed loop, thereby it is difficult to ensure the safety and stability of the tunnel internal engineering vehicle driving, and to a certain extent, it causes damage to the engineering vehicle. SUMMARY

[0005] The purpose of the present application is to provide a vehicle adaptive intelligent guidance method and system based on tunnel positioning, which solves the problems in the background art.

[0006] To solve the above technical problems, the present application adopts the following technical scheme: the present application provides a vehicle adaptive intelligent guidance method based on tunnel positioning, comprising: ST1, sending task instructions to a plurality of mobile vehicles by a tunnel control platform and analyzing the task instructions, and automatically generating the optimal execution path of the plurality of mobile vehicles according to the initial position information of the plurality of mobile vehicles and the target point position information in the analyzed task instructions.

[0007] ST2, each mobile vehicle real-time positioning and perception, update high-precision local map and semantic map, and use the sensor return data of each mobile vehicle for local path planning and obstacle avoidance.

[0008] The second aspect of the present application provides a system for performing the tunnel positioning-based vehicle adaptive intelligent guidance method described in the present application, comprising: a task instruction analysis module, which sends task instructions to a plurality of mobile vehicles and analyzes the task instructions, and automatically generates the optimal execution path of the plurality of mobile vehicles according to the initial position information of the plurality of mobile vehicles and the target point position information in the analyzed task instructions.

[0009] A map updating and path planning module, each mobile vehicle real-time positioning and perception, update high-precision local map and semantic map, and use the sensor return data of each mobile vehicle for local path planning and obstacle avoidance.

[0010] The beneficial effects of the present application are as follows: (1) The present application analyzes the task instructions, automatically generates the optimal execution path of each mobile vehicle according to the task instructions, and updates the high-precision local map and semantic map in real time during the movement of each mobile vehicle, realizes real-time, accurate, and semantic perception and map construction of the construction environment, and lays a foundation for subsequent local path planning and obstacle avoidance.

[0011] (2) The present application introduces the state of the overbreak and underbreak area, the geological change point, the working face, and the surrounding rock of the support section in the tunnel as a constraint condition in the process of local path planning and obstacle avoidance, truly forms a vehicle control closed loop, so as to ensure the safety and stability of the tunnel internal engineering vehicle while avoiding damage to the engineering vehicle to a certain extent. BRIEF DESCRIPTION OF DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiment or prior art description. Obviously, the drawings in the following description only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0013] Figure 1 The present application is a method for implementing step flowchart.

[0014] Figure 2 The present application is a system structure connection diagram. DETAILED DESCRIPTION

[0015] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described, obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work belong to the scope of protection of the present application.

[0016] With reference to Figure 1 The first aspect of the present application provides a tunnel positioning-based vehicle adaptive intelligent guidance method, comprising: ST1, sending task instructions to a plurality of mobile vehicles by a tunnel control platform and analyzing the task instructions, and automatically generating optimal execution paths of the plurality of mobile vehicles according to initial position information of the plurality of mobile vehicles and target point position information in the analyzed task instructions.

[0017] In specific embodiments of the present application, the specific analysis content of the analyzed task instructions comprises: identifying a target point and a work type according to the task instructions, and positioning to the target point position information in combination with a tunnel three-dimensional model.

[0018] ST2, real-time positioning and perception of each mobile vehicle, updating a high-precision local map and a semantic map, and performing local path planning and obstacle avoidance by using sensor return data of each mobile vehicle.

[0019] In specific embodiments of the present application, the specific content of ST2 comprises: ST201, each mobile vehicle scans the inside environment of the opened tunnel based on a mounted scanner in real time, performs rapid high-precision matching with a design BIM model, automatically identifies the change type and change parameter of the overbreak and underbreak area and the geological change point, and dynamically updates the high-precision local map.

[0020] ST202, using computer vision to identify, classify and track key elements in the tunnel in real time, the key elements comprising mobile vehicles, materials, personnel and temporary obstacles, and dynamically labeling the identification results on the semantic map.

[0021] The present application analyzes the task instructions, automatically generates the optimal execution paths of each mobile vehicle according to the task instructions, and dynamically updates the high-precision local map and the semantic map during the movement of each mobile vehicle, thereby realizing real-time, accurate and semantic perception and map construction of the construction environment, and laying a foundation for subsequent local path planning and obstacle avoidance.

[0022] ST203, in combination with the sensor return data of each mobile vehicle, real-time evaluation of the stability of the working face and the state of the surrounding rock of the support section, and obtaining a risk evaluation result in the tunnel.

[0023] ST204: generating constraint conditions for guiding path planning according to the high-precision local map, the semantic map and the risk assessment result in the tunnel, and performing local path planning and obstacle avoidance according to the constraint conditions.

[0024] In specific embodiments of the application, the constraint conditions for guiding path planning are generated according to the high-precision local map, the semantic map and the risk assessment result of each mobile vehicle. Specifically, the first risk constraint parameter of each mobile vehicle is evaluated as constraint condition 1 for guiding path planning according to the three-dimensional model of the over-excavation area, the change type and the change parameter of the geological change point in the high-precision local map.

[0025] The second risk constraint parameter of each mobile vehicle is evaluated as constraint condition 2 for guiding path planning according to the key elements in the tunnel in the semantic map.

[0026] The risk assessment result in the tunnel is taken as constraint condition 3 for guiding path planning.

[0027] In specific embodiments of the application, the first risk constraint parameter of each mobile vehicle is evaluated as follows. Based on the three-dimensional model of the over-excavation area in the high-precision local map of each mobile vehicle, the volume, the impact parameter and the depth of the over-excavation area, and the volume, the impact parameter and the depth of the under-excavation area are extracted. The impact parameter is specifically the volume of the prohibited area or the ground flatness, and the impact parameter is specifically constrained by the position of the over-excavation area and the under-excavation area in the tunnel.

[0028] Specifically, if the over-excavation area is not in the ground area, the impact parameter thereof is the volume of the prohibited area, if the over-excavation area is in the ground area, the impact parameter thereof is the ground flatness, if the under-excavation area is not in the ground area, the impact parameter thereof is the volume of the prohibited area, and if the under-excavation area is in the ground area, the impact parameter thereof is the ground flatness.

[0029] It should be noted that the ground flatness is quantified by the fluctuation of the ground height. For example, the ground height data set can be quantified by the dispersion, or the variance and the standard deviation.

[0030] Based on the change type and the change parameter of the geological change point in the high-precision local map of each mobile vehicle, and quantifying the risk factor of the geological change point, the risk factor specifically reflects the risk degree of the geological change, and the greater the change parameter, the greater the risk factor.

[0031] It should be noted that the change type includes softness and fissure, and the change parameter is area.

[0032] It should also be noted that the change parameter is uniformly processed to obtain the risk factor of the geological change point.

[0033] The volume, influence parameter and depth of the overbreak area, the volume, influence parameter and depth of the underbreak area, and the risk factor of the geological change point of each mobile vehicle are uniformly processed and weightedly summed to obtain the first risk constraint parameter of each mobile vehicle.

[0034] It should be noted that the weighted sum is specifically the multiplication and accumulation of the volume, influence parameter and depth of the overbreak area, the volume, influence parameter and depth of the underbreak area, and the risk factor of the geological change point of each mobile vehicle after uniform processing and the corresponding weight factor, and the weight factor specifically includes an influence direction and an influence degree, is stored in a data warehouse, and the influence direction includes a positive direction and a negative direction, and represents the relationship between the volume, influence parameter and depth of the overbreak area, the volume, influence parameter and depth of the underbreak area, and the risk factor of the geological change point and the first risk constraint parameter. If it is positive, it means positive correlation, and if it is negative, it means negative correlation.

[0035] In specific embodiments of the application, the second risk constraint parameter of each mobile vehicle is evaluated, and the specific evaluation method is as follows: based on the key elements in the tunnel in the semantic map of each mobile vehicle, the basic volume of the key elements is obtained, and the predicted moving trajectory of several key elements is predicted to fit the predicted moving volume trajectory of several key elements.

[0036] It should be noted that the predicted moving trajectory of several key elements is specifically predicted according to the historical moving trajectory of several key elements recognized by each mobile vehicle, and can be realized by relying on physical models, machine learning, deep learning and other methods, and is prior art, which will not be described here.

[0037] The optimal execution path of each mobile vehicle is compared with the predicted moving volume trajectory of several key elements, the coincident volume trajectory of each mobile vehicle and several key elements is extracted, and the second risk constraint parameter of each mobile vehicle is obtained after uniform processing and mean processing.

[0038] In specific embodiments of the application, the real-time evaluation of the stability of the working face and the state of the surrounding rock of the support section obtains the risk evaluation result in the tunnel, and the specific content includes: obtaining the surface three-dimensional point cloud of the working face from the sensor return data of each mobile vehicle, dividing the working face into several regions, and then calculating the surface displacement rate of several regions of the working face through point cloud registration, and identifying the crack length, width and expansion rate of the working face, and obtaining the data set of the working face state evaluation, and obtaining the working face state constraint parameter Q1 through data processing, wherein the Q1 contains the numerical value of 0 and 1, and when Q1=1, it means that the working face is in a risk state, and when Q1=0, it means that the working face is in a normal state.

[0039] It also needs to be explained that based on the three-dimensional point cloud of the working face surface of each mobile vehicle, the three-dimensional point cloud of the working face is sorted in chronological order according to the time stamp of the sensor return data of each mobile vehicle, and then the surface displacement rate of several regions of the working face is calculated through point cloud registration. The calculation method in the prior art is relatively mature, and will not be repeated here. The crack length and width of the working face specifically refer to the crack length and width identified according to the last time stamp return data. The expansion rate specifically refers to the average of the crack area growth rate between adjacent time stamps. The identification method in the prior art is relatively mature, and will not be repeated here.

[0040] It needs to be explained that the working face state constraint parameter obtained through data processing is specifically processed as follows: if the surface displacement rate of a certain region exceeds the allowable displacement rate or the crack length, width, or expansion rate of the working face is greater than the allowable crack length, width, or expansion rate, the working face state constraint parameter is recorded as 1; otherwise, the working face state constraint parameter is recorded as 0.

[0041] The stress distribution map and the water pressure distribution map of the supporting section surrounding rock are obtained through the embedded optical fiber strain sensor array and the pore water pressure sensor array of the supporting section surrounding rock.

[0042] The stress mutation points and the water pressure mutation points are identified according to the stress distribution map and the water pressure distribution map of the supporting section surrounding rock, the stress mutation thermal map and the water pressure mutation thermal map are drawn, the thermal grade of each stress mutation region and the thermal grade of each water pressure mutation region are extracted, and the stress mutation region area ratio and the water pressure mutation region area are extracted. The data set for evaluating the state of the supporting section surrounding rock is obtained by summarizing, and the supporting section surrounding rock state constraint parameter Q2 is obtained through data processing. The Q2 contains the values of 0 and 1. When Q2=1, it means that the supporting section surrounding rock is in a risk state, and when Q2=0, it means that the supporting section surrounding rock is in a normal state.

[0043] It needs to be explained that the data processing of the large supporting section surrounding rock state constraint parameter is specifically processed as follows: if the thermal grade of a certain stress mutation region belongs to the thermal risk grade stored in the data warehouse, the thermal grade of a certain water pressure mutation region belongs to the thermal risk grade, the stress mutation region area ratio is greater than the stress mutation bearing area ratio, or the water pressure mutation region area is greater than the water pressure mutation bearing area ratio, the supporting section surrounding rock state constraint parameter is recorded as 1; otherwise, it is recorded as 0.

[0044] It also needs to be explained that the stress mutation point specifically refers to a point whose stress is not within the safe stress interval, and the water pressure mutation point specifically refers to a point whose water pressure is not within the safe water pressure interval. The farther the stress of the stress mutation point is from the middle value of the safe stress interval, the higher the thermal grade is. The farther the water pressure of the water pressure mutation point is from the middle value of the safe water pressure interval, the higher the thermal grade is.

[0045] If , the risk assessment result in the tunnel is recorded as 1, otherwise, the risk assessment result in the tunnel is recorded as 0, is a logical symbol or.

[0046] In specific embodiments of the application, the local path planning and obstacle avoidance according to the constraint conditions specifically include: calculating the passing margin of the global map according to constraint condition 1, constraint condition 2 and constraint condition 3, the passing margin of the global map is specifically the proportion of the allowed passing map volume to the global map volume, and the passing margin of the global map is introduced into the guidance mode model , is a passing margin threshold value stored in the data warehouse, F1 represents that the guidance mode is global planning, and F2 represents that the guidance mode is local planning.

[0047] It also needs to be supplemented that when the risk assessment result in the tunnel is 1, the passing margin of the global map is forced to be 0.

[0048] Exemplarily, if the first risk constraint parameter of a mobile vehicle in the detection range is greater than or equal to the first risk constraint parameter threshold value, the detection range is marked as the color corresponding to constraint condition 1, and the color of constraint condition 2 is marked in the same way, and the total range volume of the constraint condition 1 color of the global map and the total range volume of the constraint condition 2 color are obtained to calculate the passing margin of the global map.

[0049] The global planning specifically suspends the actions of all vehicles, and performs special processing on the inside of the tunnel, including but not limited to maintenance of the working face, maintenance of the support section, backfilling of overbreak areas, re-excavation of underbreak areas, etc.

[0050] The local planning specifically dynamically plans the dynamic passing path of each mobile vehicle according to the current position, high-precision local map and semantic map of each mobile vehicle, and the dynamic passing path specifically is a safe passing path that avoids overbreak areas, underbreak areas and geological change points, etc.

[0051] The application introduces the states of overbreak and underbreak areas, geological change points, working faces and support section surrounding rocks in the inside of the tunnel as constraint conditions in the process of local path planning and obstacle avoidance, truly forms a closed loop of vehicle control, so as to ensure the safety and stability of the driving of the engineering vehicles in the inside of the tunnel, and to a certain extent, avoid damage to the engineering vehicles.

[0052] Referring to Figure 2As shown, the second aspect of the present application provides a system for executing the tunnel positioning-based vehicle adaptive intelligent guidance method described in the present application, comprising: a task instruction analysis module, which sends task instructions to a plurality of mobile vehicles by a tunnel control platform and analyzes the task instructions, and automatically generates optimal execution paths of the plurality of mobile vehicles according to initial position information of the plurality of mobile vehicles and target point position information in the analyzed task instructions.

[0053] a map updating and path planning module, which updates high-precision local maps and semantic maps in real time according to positioning and perception of each mobile vehicle, and performs local path planning and obstacle avoidance by using sensor return data of each mobile vehicle.

[0054] It should be noted that the system further comprises a data warehouse for storing relevant data for realizing the tunnel positioning-based vehicle adaptive intelligent guidance.

[0055] The above is only an example and description of the concept of the present application, and those skilled in the art can make various modifications or supplements or use similar ways to replace the described specific embodiments, as long as they do not deviate from the concept of the present application or exceed the scope defined by the present application, which shall fall within the protection scope of the present application.

Claims

1. A tunnel positioning based vehicle adaptive intelligent guidance method, characterized in that, Comprise: ST1, send task instructions to a plurality of mobile vehicles by a tunnel control platform and analyze the task instructions, automatically generate the optimal execution path of the plurality of mobile vehicles according to the initial position information of the plurality of mobile vehicles and the target point position information in the analyzed task instructions; ST2, each mobile vehicle real-time positioning and perception, update high-precision local map and semantic map, and use sensor return data of each mobile vehicle for local path planning and obstacle avoidance; The specific content of ST2 comprises: ST201: each mobile vehicle scans the internal environment of the tunnel based on the scanner carried by the mobile vehicle in real time, quickly and accurately matches with the designed BIM model, automatically identifies the change type and change parameter of the overbreak and underbreak area and the geological change point, and dynamically updates the high-precision local map; ST202: using computer vision, real-time identification, classification and tracking of key elements in the tunnel, the key elements include mobile vehicles, materials, personnel and temporary obstacles, and the identification results are dynamically labeled on the semantic map; ST203: combining the sensor return data of each mobile vehicle, real-time evaluation of the stability of the working face and the state of the surrounding rock of the support section, and obtaining the risk evaluation result in the tunnel; ST204: according to the high-precision local map, the semantic map and the risk evaluation result in the tunnel, generate constraint conditions for guiding path planning, and conduct local path planning and obstacle avoidance according to the constraint conditions; The specific generation method of the constraint conditions for guiding path planning according to the high-precision local map, the semantic map and the risk evaluation result of each mobile vehicle is: According to the three-dimensional model of the overbreak and underbreak area, the change type and change parameter of the geological change point in the high-precision local map, evaluate the first risk constraint parameter of each mobile vehicle as constraint condition 1 for guiding path planning; According to the key elements in the tunnel in the semantic map, evaluate the second risk constraint parameter of each mobile vehicle as constraint condition 2 for guiding path planning; The risk evaluation result in the tunnel is used as constraint condition 3 for guiding path planning; The specific content of the local planning and obstacle avoidance according to the constraint conditions comprises: According to constraint condition 1, constraint condition 2 and constraint condition 3, the passing margin of the global map is calculated, and the passing margin of the global map is specifically the proportion of the allowed passing map volume to the global map volume, and the passing margin of the global map is introduced into the guidance mode model is a threshold value of the passing margin stored in the data warehouse, F1 indicates that the guidance mode is global planning, and F2 indicates that the guidance mode is local planning.​ The global planning is to suspend the action of all vehicles and perform special processing on the internal environment of the tunnel; The local planning is to dynamically plan the dynamic traffic path of each mobile vehicle according to the current position of each mobile vehicle, the high-precision local map and the semantic map.

2. The tunnel positioning based vehicle adaptive intelligent guidance method according to claim 1, wherein, The specific analysis content of the analyzed task instructions comprises: identifying the target point and the operation type according to the task instructions, and positioning to the target point position information combined with the three-dimensional model of the tunnel.

3. The tunnel positioning based vehicle adaptive intelligent guidance method according to claim 2, wherein, The specific evaluation method of the first risk constraint parameter of each mobile vehicle is: Based on the three-dimensional model of the overbreak and underbreak area in the high-precision local map of each mobile vehicle, the volume, influence parameter and depth of the overbreak area, and the volume, influence parameter and depth of the underbreak area are extracted, the influence parameter is specifically the volume of the prohibited traffic area or the ground flatness, and the influence parameter is specifically constrained by the position of the overbreak and underbreak area in the tunnel; Based on the change type and change parameter of the geological change point in the high-precision local map of each mobile vehicle, and quantifying the risk factor of the geological change point, the risk factor specifically embodies the risk degree of the geological change, and the greater the change parameter, the greater the risk factor; The volume, influence parameter and depth of the overbreak area, the volume, influence parameter and depth of the underbreak area, and the risk factor of the geological change point of each mobile vehicle are uniformly processed and weighted summed to obtain the first risk constraint parameter of each mobile vehicle.

4. The tunnel positioning based vehicle adaptive intelligent guidance method according to claim 3, wherein, The second risk constraint parameter of each mobile vehicle is evaluated, and the specific evaluation method is: Based on the key elements in the tunnel in the semantic map of each mobile vehicle, the basic volume of the key elements is obtained, and the predicted moving trajectory of several key elements is predicted, and the predicted moving volume trajectory of several key elements is fitted; The optimal execution path of each mobile vehicle is compared with the predicted moving volume trajectory of several key elements, the overlapping volume trajectory of each mobile vehicle and several key elements is extracted, and the overlapping volume trajectory is uniformly processed and averaged to obtain the second risk constraint parameter of each mobile vehicle.

5. The tunnel positioning based vehicle adaptive intelligent guidance method according to claim 4, wherein, The real-time evaluation of the stability of the working face and the state of the surrounding rock of the support section is obtained, and the risk evaluation result in the tunnel, and the specific content includes: The surface three-dimensional point cloud of the working face is obtained from the sensor return data of each mobile vehicle, the working face is divided into several regions, and then the surface displacement rate of the working face is calculated by point cloud registration, and the crack length, width and expansion rate of the working face are identified, and the data set of the working face state evaluation is obtained, and the working face state constraint parameter Q1 is obtained by data processing, and the Q1 contains the values of 0 and 1, when Q1=1, it means that the working face is in a risk state, and when Q1=0, it means that the working face is in a normal state; The stress distribution map and water pressure distribution map of the surrounding rock of the support section are obtained through the implanted fiber optic strain sensor array and pore water pressure sensor array of the surrounding rock of the support section; The stress mutation points and water pressure mutation points are identified according to the stress distribution map and water pressure distribution map of the surrounding rock of the support section, the stress mutation thermal map and water pressure mutation thermal map are drawn, the thermal grade of each stress mutation area and the thermal grade of each water pressure mutation area are extracted, and the stress mutation area ratio and water pressure mutation area are extracted, and the data set of the surrounding rock state evaluation of the support section is obtained, and the surrounding rock state constraint parameter Q2 of the support section is obtained by data processing, and the Q2 contains the values of 0 and 1, when Q2=1, it means that the surrounding rock of the support section is in a risk state, and when Q2=0, it means that the surrounding rock of the support section is in a normal state; If then the risk assessment result in the tunnel is recorded as 1, otherwise, the risk assessment result in the tunnel is recorded as 0, is a logical symbol or.

6. A system for performing the tunnel-based positioning based vehicle adaptive intelligent guidance method of any one of claims 1-5, characterized in that, It includes: The task instruction analysis module sends task instructions to several mobile vehicles from the tunnel control platform and analyzes the task instructions, and automatically generates the optimal execution path of several mobile vehicles according to the initial position information of several mobile vehicles and the position information of the target point in the analyzed task instructions; The map updating and path planning module updates the high-precision local map and semantic map in real time, and plans the local path and avoids obstacles by using the sensor return data of each mobile vehicle.

Images