Underground pipe gallery intelligent inspection robot path planning method

Abstract

The invention relates to the field of information processing, and discloses an underground pipe gallery intelligent inspection robot path planning method, which comprises the steps of judging an environment change risk based on underground pipe gallery structural characteristics, and dynamically adjusting a detection period; triggering local path reconstruction in combination with sensor data change and target conflict conditions; and orderly collaboration of path safety, energy consumption and smoothness is realized through a layered optimization architecture. According to the method, triggering logic is used for replacing a traditional periodic planning mechanism, risks and conflict points in operation can be recognized more accurately, and active adaptation and quick response of path decision are achieved. Through a multi-dimensional decoupling architecture, mutual interference of targets is avoided, and the system stability and execution efficiency of path planning are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of information processing methods, and in particular to a path planning method for an intelligent inspection robot in an underground utility tunnel. Background Art

[0002] As an important part of the urban lifeline project, the underground utility tunnel carries multiple types of pipelines such as power, communication, gas, water supply and drainage. The operating environment is complex, the space is narrow, and it extends over a long distance. It is vulnerable to environmental disturbances. Therefore, daily inspection work is particularly important. In recent years, with the development of intelligent robot technology, intelligent inspection robots in underground utility tunnels have gradually replaced the traditional manual inspection mode. Path planning, as one of the core capabilities of robots to execute tasks, its optimization effect is directly related to the inspection efficiency, safety and the utilization rate of system resources.

[0003] In the prior art, path planning usually relies on a periodic replanning mechanism, that is, during the operation of the robot, the path is globally updated at a preset time interval. Although this method has a certain degree of stability, it often faces problems such as untimely response and resource waste in a dynamic environment. For example, when sudden obstacles or environmental mutations occur inside the utility tunnel, the robot may collide or misjudge because it fails to adjust the path in time.

[0004] To improve the flexibility and adaptive ability of path planning, some studies have introduced multi-objective optimization algorithms, attempting to balance between path safety, energy consumption and path smoothness. However, since these objectives often conflict in actual operation, such as the path with the optimal energy consumption may sacrifice path safety, the traditional weighted method often has difficulty in coordinating the priorities among multiple objectives, resulting in unstable planning results. In addition, in the face of abnormal sensor interference, such as data mutations caused by methane leakage or position offsets caused by electromagnetic interference, the existing path planning mechanism is difficult to make an effective response. Such problems are particularly prominent in underground utility tunnels because their environment has typical characteristics such as high humidity and high electromagnetic interference. Although some solutions in the industry have tried to improve the system robustness by adding redundant hardware sensors or external navigation devices, this often brings problems of increased cost and deployment complexity in actual engineering.

[0005] Therefore, there is an urgent need in the industry for a new path planning mechanism that can trigger path updates dynamically based on the environmental state and achieve structural decoupling among multiple objectives to improve the real-time response ability of the system and the overall path execution quality. Summary of the Invention

[0006] The technical problem solved by the present invention is to provide a path planning method for an intelligent inspection robot in an underground utility tunnel to solve the problems of late response of the path update mechanism, conflict of target optimization and poor coherence of path reconstruction as mentioned in the above background art.

[0007] To solve the above technical problems, a path planning method for an intelligent inspection robot in an underground utility tunnel of the present invention includes the following steps: Based on the preset structural feature information of the underground utility tunnel, the environmental area where the inspection robot is currently located is judged in real time. When it is judged that the inspection robot enters a structural feature area where environmental mutations are likely to occur, the environmental information detection period is automatically adjusted to a preset shortened period to trigger path update;

[0008] The data changes of at least one key sensor carried by the inspection robot are monitored in real time. When the data change rate of any key sensor exceeds a preset abnormal threshold, local path replanning is forcibly started;

[0009] During path planning, the safety factor S and energy consumption factor E of the current path are obtained in real time, and the absolute difference between the two is calculated. When the absolute difference is greater than the dynamic threshold Td determined based on the current environmental information, that is, when ∣S - E∣>Td is satisfied, the target weight adaptive module is activated, and the weights of the safety target and energy consumption target in path planning are dynamically adjusted according to preset rules;

[0010] A hierarchical optimization architecture is used for path decision-making. The hierarchical optimization architecture includes a safety layer, an energy consumption layer, and a path smoothness layer in sequence. Among them, the safety layer has the highest priority and is used to perform collision prediction and dangerous area avoidance using a lightweight algorithm. The energy consumption layer performs energy consumption optimization using a lightweight algorithm on the premise of meeting safety constraints. The path smoothness layer optimizes the smoothness of the path using a lightweight algorithm on the premise of meeting safety and energy consumption constraints, and finally generates the planned path of the inspection robot.

[0011] Preferably, the step of determining the dynamic threshold Td includes: establishing a three-dimensional model of the underground utility tunnel containing electromagnetic interference intensity information. The three-dimensional model divides the underground utility tunnel into multiple regions and marks the electromagnetic interference intensity I of each region; according to the real-time positioning coordinates of the inspection robot, the electromagnetic interference intensity information I corresponding to the region where the inspection robot is currently located is queried and loaded from the three-dimensional model; based on the electromagnetic interference intensity information I, the dynamic threshold Td is calculated according to a preset functional relationship, that is, Td = f(I), where f is a preset function about the electromagnetic interference intensity.

[0012] Preferably, the step of forcibly starting local path replanning includes: when the inspection robot detects an obstacle, controlling the inspection robot to move a preset number of path points in the reverse direction of the current planned path; during the reverse movement of the inspection robot, new local path planning calculations are synchronously executed to generate a local path bypassing the obstacle.

[0013] Preferably, the safety layer decision-making introduces the simulation data of the stress field of the utility tunnel structure, including: accessing the pre-established BIM model of the underground utility tunnel to obtain the structural stress distribution data therein; when performing the safety layer path planning, using the structural stress distribution data as a constraint condition to guide the path planning algorithm to generate a safe path that avoids stress concentration areas.

[0014] Preferably, when it is detected that an obstacle triggers local path replanning, the following steps are further included: controlling the inspection robot to move backward along the original planned path for 3 to 5 path points; during the backward movement, real-time inertial data of the movement of the inspection robot is collected, and the mechanical characteristics of the backward movement trajectory are extracted by analyzing the inertial data to generate a constraint condition for path smoothness; embedding the constraint condition for path smoothness into the local path planning algorithm to synchronously generate a new path that matches the mechanical characteristics of the backward movement trajectory.

[0015] Preferably, the mechanical characteristics of the backward movement trajectory include acceleration and angular velocity, and the inertial data is collected by the inertial measurement unit IMU sensor built in the inspection robot.

[0016] Preferably, the preset structural feature areas include the vault area and the bend area.

[0017] Preferably, the safety layer adopts the grid collision prediction method, and the energy consumption layer adopts the inertial motion trajectory fitting.

[0018] Preferably, the key sensor is a methane concentration sensor, and the preset abnormal threshold is that the methane concentration gradient change rate is greater than 0.5% / s.

[0019] Compared with the prior art, the beneficial effects of the present invention are:

[0020] 1. The mechanism of path update no longer depends on a single time drive, but judges and responds according to the dynamic relationship among the actual environmental state, sensor performance, and optimization objectives. This path planning method with trigger logic as the core can actively identify the mutation risks and safety conflicts existing in the operation process, break through the passive operation bottleneck of traditional path update, make the path adjustment behavior more targeted and timely, and thus significantly improve the environmental adaptability of the inspection system.

[0021] 2. The multi-dimensional hierarchical optimization structure proposed by the present invention breaks through the inherent limitation that the various objectives in the traditional multi-objective optimization method restrict each other and cannot be coordinated. By independently processing the safety, energy consumption control, and smoothness in sequence and separately at different levels, the optimization process at each level can be carried out orderly under clear preconditions, fundamentally realizing the logical decoupling among multiple objectives, and avoiding the path conflict problem caused by unclear optimization priorities. At the same time, this hierarchical design is also easier to deploy and update modularly, which is conducive to subsequent system expansion.

[0022] 3. By fusing the three-dimensional space features with the electromagnetic interference intensity mapping, an adaptive threshold adjustment mechanism is established, effectively avoiding false triggering and missed triggering in path planning, enhancing the adaptability of the planning system to complex working conditions. The introduced reverse motion strategy creates a physical data acquisition window for path feature analysis through the backward operation of the robot's current path, and transforms the trajectory inertia characteristics reflected in the reverse motion into constraint conditions and embeds them into the path generation process, realizing the synchronous closed-loop of motion as data, data as constraints, and constraints as paths. This coupling mechanism not only improves the naturalness of path connection but also enhances the dynamic continuity and stability of the overall path.

[0023] 4. Using the structural stress field data as an important decision-making basis for the path safety layer reflects the in-depth expansion of the inspection safety from surface obstacle avoidance to structural hazard avoidance. By guiding the path to avoid stress concentration areas and constructing a logically safe buffer zone for the structure, the system's forward-looking avoidance ability for potential hazards is effectively improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0025] Figure 1 It is a flowchart of the path planning of the intelligent inspection robot for underground pipe galleries of the present invention;

[0026] Figure 2 It is a flowchart of the decision-making of the safety layer of the present invention;

[0027] Figure 3 It is a flowchart of the path replanning triggered by the data of the methane concentration sensor of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0028] The following will clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of them. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0029] A method for path planning of an intelligent inspection robot for underground pipe galleries provided in this embodiment includes the following steps:

[0030] Based on the preset structural feature information of the underground utility tunnel, the environmental area where the inspection robot is currently located is judged in real time. When it is judged that the inspection robot enters the structural feature area where environmental mutations are likely to occur, the environmental information detection period is automatically adjusted to the preset shortened period to trigger path update;

[0031] The data changes of at least one key sensor carried by the inspection robot are monitored in real time. When the data change rate of any key sensor exceeds the preset abnormal threshold, local path replanning is forced to start;

[0032] During the path planning process, the safety factor S and energy consumption factor E of the current path are obtained in real time, and the absolute difference between the two is calculated. When the absolute difference is greater than the dynamic threshold Td determined based on the current environmental information, that is, when ∣S - E∣>Td is satisfied, the target weight adaptive module is activated, and the weights of the safety target and energy consumption target in path planning are dynamically adjusted according to the preset rules;

[0033] A hierarchical optimization architecture is adopted for path decision-making. The hierarchical optimization architecture includes a safety layer, an energy consumption layer, and a path smoothness layer in sequence. Among them, the safety layer has the highest priority and is used to perform collision prediction and dangerous area avoidance using a lightweight algorithm. The energy consumption layer performs energy consumption optimization using a lightweight algorithm on the premise of meeting safety constraints. The path smoothness layer optimizes the smoothness of the path using a lightweight algorithm on the premise of meeting safety and energy consumption constraints, and finally generates the planned path of the inspection robot.

[0034] Preferably, the determination step of the dynamic threshold Td includes: establishing a three-dimensional model of the underground utility tunnel containing electromagnetic interference intensity information. The three-dimensional model divides the underground utility tunnel into multiple regions and marks the electromagnetic interference intensity I of each region; according to the real-time positioning coordinates of the inspection robot, query and load the electromagnetic interference intensity information I corresponding to the region where the inspection robot is currently located from the three-dimensional model; based on the electromagnetic interference intensity information I, calculate the dynamic threshold Td according to the preset functional relationship, that is, Td = f(I), where f is a preset function about the electromagnetic interference intensity.

[0035] Preferably, the steps for forced local path replanning include: when the inspection robot detects an obstacle, control the inspection robot to move a preset number of path points in the reverse direction of the current planned path; during the reverse movement of the inspection robot, synchronously perform new local path planning calculations to generate a local path bypassing the obstacle.

[0036] Preferably, the safety layer decision-making introduces the simulation data of the stress field of the utility tunnel structure, including: accessing the pre-established BIM model of the underground utility tunnel to obtain the structural stress distribution data therein; when planning the safety layer path, using the structural stress distribution data as a constraint condition to guide the path planning algorithm to generate a safe path that avoids stress concentration areas.

[0037] Preferably, when it is detected that an obstacle triggers local path replanning, the following steps are further included: controlling the inspection robot to move backward along the original planned path for 3 to 5 path points; during the backward movement, real-time collecting the inertial data of the movement of the inspection robot, extracting the mechanical characteristics of the backward movement trajectory by analyzing the inertial data, and generating the constraint conditions of the path smoothness; embedding the constraint conditions of the path smoothness into the local path planning algorithm to synchronously generate a new path that matches the mechanical characteristics of the backward movement trajectory.

[0038] Preferably, the mechanical characteristics of the backward movement trajectory include acceleration and angular velocity, and the inertial data is collected by the inertial measurement unit IMU sensor built in the inspection robot.

[0039] Preferably, the preset structural feature areas include the vault area and the bend area.

[0040] Preferably, the safety layer adopts the grid collision prediction method, and the energy consumption layer adopts the inertial motion trajectory fitting.

[0041] Preferably, the key sensor is a methane concentration sensor, and the preset abnormal threshold is that the methane concentration gradient change rate is greater than 0.5% / s.

[0042] Example 1: In this example, by dynamically judging the environmental changes, sensor data changes and multi-objective optimization strategies, this method can respond and adjust the path of the inspection robot in real time. Specifically, according to the preset structural feature information of the underground utility tunnel, when the robot enters a specific structural feature area (such as the vault area, bend area, etc.), the system will automatically adjust the detection period. By real-time monitoring the environmental changes, it is judged whether there are areas prone to environmental mutations, such as areas with too high humidity or strong electromagnetic interference. When these risk areas are detected, the system shortens the detection period of path update, so as to be able to start the path update mechanism in time and ensure that there will be no delay in path adjustment during the inspection process; at the same time, if the data change rate of any key sensor (such as a methane concentration sensor) exceeds the preset abnormal threshold (such as the methane concentration change rate exceeds 0.5% / s), local path replanning is triggered. Specifically, the robot will forcibly start local path replanning according to the change of the sensor to bypass sudden obstacles or dangerous areas.

[0043] The system calculates the safety factor S and energy consumption factor E of the current path in real time. According to the environmental information of the current path, when the difference between S and E (|S - E|) is greater than the preset dynamic threshold Td, the target weight adaptive module is activated. This module dynamically adjusts the weights of the safety target and energy consumption target in path planning according to preset rules to ensure that the system can balance the requirements of safety and energy consumption under different environmental conditions; and in the process of path planning, a hierarchical optimization architecture is adopted, including a safety layer, an energy consumption layer, and a path smoothness layer. Different lightweight algorithms are used for optimization in each layer;

[0044] Safety layer: The collision prediction algorithm is adopted to prevent the robot from entering dangerous areas. This layer uses a grid-based collision prediction method to guide the robot to avoid obstacles by calculating the collision risk of the path in real time; Energy consumption layer: On the premise of safety constraints, the energy consumption of the path is optimized through the inertial motion trajectory fitting method to reduce unnecessary energy waste; Path smoothness layer: On the basis of ensuring safety and energy consumption, the path smoothness algorithm is used to optimize the continuity and comfort of the path and reduce the movement vibration of the robot. Through the above three-layer optimization, this embodiment effectively solves the conflict problem among safety, energy consumption, and path smoothness in traditional path planning methods.

[0045] The electromagnetic interference intensity information in the underground utility tunnel is incorporated into the 3D model, and the electromagnetic interference intensity I of each area is accurately marked. In the process of path planning, the system queries the electromagnetic interference intensity of the current area from the 3D model according to the real-time positioning information of the inspection robot, and calculates the dynamic threshold Td through the preset functional relationship Td = f(I). The introduction of this dynamic threshold can effectively avoid false triggering and missed triggering, and enhance the adaptability of the path planning system to complex working conditions; when the inspection robot encounters an obstacle, the local path replanning mechanism will be triggered. The robot will move backward along the current planned path by several path points (usually 3 to 5 points). At the same time, during the backward movement, the system will collect and analyze the inertial data of the robot in real time, and extract the mechanical characteristics of the backward movement trajectory. These mechanical characteristics (such as acceleration and angular velocity) will be used as the constraint conditions for path smoothness to further optimize the smoothness and naturalness of the path.

[0046] Embodiment 2: Figures 1 to 3 It shows the key processes in the path planning method of the underground utility tunnel intelligent inspection robot, including path optimization, local path replanning, and safety guarantee strategies. The following is the detailed implementation process of this method:

[0047] Figure 1Shows the overall process of path planning for an intelligent inspection robot in an underground utility tunnel. First, the robot starts from the starting node and goes through the safety layer inspection process to ensure the safety of the path. If the safety inspection is passed, the path enters the energy consumption layer optimization stage, where the energy consumption during the inspection is reduced by optimizing the energy consumption. Subsequently, it enters the smoothness optimization stage to further optimize the smoothness of the path to ensure the comfort and efficiency of the robot's travel. After completing these optimization steps, the final inspection path will be generated in the path output stage. If the safety layer inspection fails, the system will return to the path reconstruction step to re-plan the path and ensure the safe execution of the inspection task.

[0048] Figure 2 Shows the decision-making process of the safety layer in path planning. At this stage, the system obtains the current position of the robot through the real-time positioning module and combines the BIM model and the mine collapse prediction function to evaluate the structural stress in the utility tunnel. Combining the structural stress distribution data and the electromagnetic interference intensity information, the path generator calculates and generates a safe inspection path on the premise of ensuring avoidance of potential dangerous areas. This safety guarantee strategy is to ensure that the inspection robot can avoid risk areas and maintain the safety of the path in a complex underground environment.

[0049] Figure 3 Shows the specific process of path re-planning triggered by methane concentration sensor data. When the methane sensor detects that the concentration change rate exceeds the threshold of 0.5% / s, the relevant data will be sent to the control center. After receiving the abnormal data, the control center sends a forced re-planning instruction to the path planner to start the local path calculation. After calculation, the system will confirm and generate a new path and transmit it to the actuator for execution. If the methane concentration change does not reach the threshold, the system will continue to execute the original path to ensure the continuity and stability of the inspection task.

[0050] Embodiment 3: In this embodiment, the specific implementation steps of the path planning method for the intelligent inspection robot in the underground pipe gallery will be described in detail. Step 1: The inspection robot first determines the current environmental area according to the preset structural feature information of the underground pipe gallery. For example, the robot monitors environmental changes such as humidity and electromagnetic interference intensity in real time through sensors. Especially in the vault area or the bend area, when the environmental change is greater than the preset threshold, the system will automatically adjust the detection period. In this process, the change rate of sensor data will directly affect the adjustment of the path planning period to ensure timely response to environmental changes. At the same time, the robot is equipped with at least one key sensor to monitor the change of sensor data in real time. If the change rate of a certain sensor exceeds the preset abnormal threshold, the system will forcibly start local path replanning. Specifically, the robot will move backward 3 to 5 path points along the original path to avoid potential obstacles or hazard sources, and simultaneously perform new path planning calculations to generate a new path bypassing the obstacles. Through this strategy, the robot can flexibly respond to sudden environmental changes or sudden obstacles and maintain the continuity of the inspection task.

[0051] Step 2: During the path planning process, the robot will calculate the safety factor (S) and energy consumption factor (E) of the current path in real time. As shown in the formula:

[0052] |S - E| > T d ,

[0053] where S is the safety factor, E is the energy consumption factor, and T_d is the dynamic threshold based on the current environmental information. The value of the dynamic threshold T_d will be automatically adjusted according to different environmental conditions (such as electromagnetic interference intensity, battery power, etc.). When the calculated absolute difference is greater than T_d, the target weight adaptive module is activated to dynamically adjust the weights of the safety target and the energy consumption target. For example, in an area with a large electromagnetic interference intensity, the system may increase the weight of path safety to ensure that the robot avoids dangerous areas; in the case of insufficient battery power, the system may give priority to optimizing the energy consumption path.

[0054] Step 3: The electromagnetic interference intensity information in the underground pipe gallery is accurately marked in the 3D model. The 3D model divides the pipe gallery into multiple areas, and each area is marked with the electromagnetic interference intensity I. During the path planning process, the inspection robot obtains its current position through real-time positioning, then queries the electromagnetic interference intensity I of the current area from the 3D model, and calculates the dynamic threshold T according to the following preset functional relationship d :

[0055] T d = f(I),

[0056] where f(I) is a preset function about the electromagnetic interference intensity I, and T dDynamically adjust according to the change of electromagnetic interference intensity I. This mechanism can effectively avoid mis-triggering or missed triggering of path planning and enhance the adaptability of the path planning system to complex working conditions; when the robot detects an obstacle and triggers local path replanning, it will move backward 3 to 5 path points along the current planned path. During the backward movement, the robot will collect and analyze inertial data of the movement (such as acceleration and angular velocity) in real time, so as to extract the mechanical characteristics of the backward movement trajectory. These mechanical characteristics will be used as the constraint conditions for path smoothness. By embedding these constraint conditions into the local path planning algorithm, a smoother and more natural path can be generated, reducing vibrations or discontinuities caused by path mutations. In the decision-making process of the path safety layer, in addition to basic collision prediction, the system also introduces simulation data of the stress field of the utility tunnel structure. This data accesses the pre-established BIM model of the underground utility tunnel to obtain the structural stress distribution data. Based on these data, the path planning system can identify stress concentration areas and avoid these areas during path planning to generate a safe path. This mechanism greatly improves the safety of the robot in complex environments and ensures that it will not enter potential dangerous areas during the inspection process.

[0057] Embodiment 4: In this embodiment, for the dynamic adjustment of the target weight and the optimization of path smoothness in the path planning process, the following specific implementation steps are proposed. Dynamic adjustment of the weight of the path planning target: During the path planning process, the safety factor S and the energy consumption factor E of the current path are calculated in real time, and their weights are adjusted according to environmental changes. When the difference |S - E| between S and E is greater than the set dynamic threshold Td, the target weight adaptive module is activated. Specifically, when the system detects an increase in the electromagnetic interference intensity I in the environment, the safety target weight will be appropriately increased to ensure that the robot avoids high electromagnetic interference areas. When the battery power is insufficient, the system will prioritize optimizing energy consumption and reduce the weight of the energy consumption target. This dynamic adjustment module balances the safety and energy consumption targets in different scenarios by analyzing sensor data, environmental changes, battery status and other factors in real time, ensuring the flexibility and adaptability of path planning.

[0058] In this embodiment, by constructing a three-dimensional model of the underground utility tunnel, the utility tunnel is divided into multiple regions and the electromagnetic interference intensity I is marked. During path planning, the robot obtains the electromagnetic interference intensity I of the current region according to its real-time positioning information and calculates the dynamic threshold Td according to the preset functional relationship Td = f(I), so that the path planning system can more accurately respond to changes in different environments and avoid mis-triggering or missed triggering.

[0059] During the path replanning process, the robot moves 3 to 5 path points in the reverse direction along the current path. To improve path smoothness and avoid discontinuities caused by reverse movement, this embodiment further proposes a mechanical feature analysis method based on inertial data. The inertial measurement unit (IMU) sensor built into the robot collects inertial data such as acceleration and angular velocity in real time. By analyzing the mechanical characteristics of the reverse movement trajectory, the system can generate a new path that conforms to physical properties, embed the path smoothness as a constraint condition into the local path planning algorithm to ensure that the path is more natural and continuous, and by accessing the BIM model to obtain the stress distribution data in the pipe gallery, the path planning system can identify potential stress concentration areas and avoid these high-risk areas when generating a safe path. This mechanism can effectively prevent the inspection robot from entering areas that may have structural hidden dangers, improve the safety and stability of the system, and during the optimization process of the path smoothness layer, the robot not only considers the safety and energy consumption of the path, but also further optimizes the path smoothness through inertial data analysis. By analyzing the inertial characteristics of the reverse movement trajectory, calculating the acceleration and angular velocity of the path, and generating path constraints that conform to the movement characteristics. These constraint conditions are embedded into the local path planning algorithm, thereby ensuring that the path is smoother and more natural during execution, reducing vibrations and discomfort, and all belong to the extended implementation methods known to those of ordinary skill in the art.

[0060] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and without departing from the spirit or basic characteristics of the present invention, the present invention can be implemented in other specific forms. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, it is intended to embrace all changes that fall within the meaning and scope of the equivalent elements of the claims within the present invention.

Claims

1. An intelligent inspection robot path planning method for an underground pipe gallery, characterized in that, Including the following steps: Based on the preset structural feature information of the underground utility tunnel, the environment area where the inspection robot is currently located is judged in real time. When it is judged that the inspection robot enters the structural feature area where environmental mutations are likely to occur, the environmental information detection period is automatically adjusted to the preset shortened period to trigger path update; The data changes of at least one key sensor carried by the inspection robot are monitored in real time. When the data change rate of any key sensor exceeds the preset abnormal threshold, local path replanning is forcibly started; During path planning, the safety factor S and energy consumption factor E of the current path are obtained in real time, and the absolute difference between the two is calculated. When the absolute difference is greater than the dynamic threshold Td determined based on the current environmental information, that is, when ∣S - E∣>Td is satisfied, the target weight adaptive module is activated, and the weights of the safety target and energy consumption target in path planning are dynamically adjusted according to the preset rules; A hierarchical optimization architecture is adopted for path decision-making. The hierarchical optimization architecture includes a safety layer, an energy consumption layer, and a path smoothness layer in sequence. Among them, the safety layer has the highest priority and is used to perform collision prediction and dangerous area avoidance using a lightweight algorithm. The energy consumption layer performs energy consumption optimization using a lightweight algorithm on the premise of meeting safety constraints. The path smoothness layer optimizes the smoothness of the path using a lightweight algorithm on the premise of meeting safety and energy consumption constraints, and finally generates the planned path of the inspection robot.

2. The path planning method of the intelligent inspection robot for the underground pipe gallery according to claim 1, wherein The steps for determining the dynamic threshold Td include: establishing a three-dimensional model of the underground utility tunnel containing electromagnetic interference intensity information. The three-dimensional model divides the underground utility tunnel into multiple regions and marks the electromagnetic interference intensity I of each region; according to the real-time positioning coordinates of the inspection robot, query and load the electromagnetic interference intensity information I corresponding to the region where the inspection robot is currently located from the three-dimensional model; based on the electromagnetic interference intensity information I, calculate the dynamic threshold Td according to the preset functional relationship, that is, Td = f(I), where f is a preset function about the electromagnetic interference intensity.

3. The method for path planning of the intelligent inspection robot for the underground pipe gallery according to claim 1, wherein, The steps for forcibly starting local path replanning include: when the inspection robot detects an obstacle, controlling the inspection robot to move a preset number of path points in the reverse direction of the current planned path; during the reverse movement of the inspection robot, synchronously execute new local path planning calculations to generate a local path bypassing the obstacle.

4. The path planning method of the intelligent inspection robot for the underground pipe gallery according to claim 1, characterized in that, The safety layer decision introduces the simulation data of the pipe gallery structure stress field, including: accessing the pre-established BIM model of the underground utility tunnel and obtaining the structural stress distribution data therein; when performing safety layer path planning, using the structural stress distribution data as a constraint condition to guide the path planning algorithm to generate a safe path avoiding stress concentration areas.

5. The method for path planning of an intelligent inspection robot for an underground pipe gallery according to claim 1, wherein When local path replanning is triggered by detecting an obstacle, the following steps are further included: controlling the inspection robot to move backward along the original planned path for 3 to 5 path points; during the backward movement, inertial data of the movement of the inspection robot is collected in real time, and the mechanical characteristics of the backward movement trajectory are extracted by analyzing the inertial data to generate constraint conditions for path smoothness; embedding the constraint conditions for path smoothness into the local path planning algorithm to synchronously generate a new path that matches the mechanical characteristics of the backward movement trajectory.

6. The path planning method for the intelligent inspection robot in the underground pipe gallery according to claim 5, characterized in that, The mechanical characteristics of the backward movement trajectory include acceleration and angular velocity, and the inertial data is collected by an inertial measurement unit (IMU) sensor built in the inspection robot.

7. The path planning method of the intelligent inspection robot for the underground pipe gallery according to claim 1, characterized in that, The preset structural feature areas include the vault area and the bend area.

8. The path planning method of the intelligent inspection robot for the underground pipe gallery according to claim 1, characterized in that, The safety layer uses the grid collision prediction method, and the energy consumption layer uses inertial motion trajectory fitting.

9. The path planning method of the intelligent inspection robot for the underground pipe gallery according to claim 1, wherein, The key sensor is a methane concentration sensor, and the preset abnormal threshold is that the methane concentration gradient change rate is greater than 0.5% / s.

Images