Unmanned aerial vehicle navigation dynamic planning method and system based on inspection vehicle

Abstract

The invention provides an unmanned aerial vehicle navigation dynamic planning method and system based on an inspection vehicle, and relates to the technical field of tunnel inspection, and the method comprises the following steps: S1, obtaining the task load of the current inspection vehicle to inspect the interior of a tunnel; s2, the current navigation state of the unmanned aerial vehicle is a global positioning mode, and the unmanned aerial vehicle obtains position information through multi-sensor fusion in the global positioning mode; s3, detecting whether the residual electric quantity of the current unmanned aerial vehicle is reduced to a preset threshold value or not; s4, judging whether the current unmanned aerial vehicle can complete the task load in the global positioning mode or not in real time according to the first flied distance and the first electric quantity consumption; and S5, according to the interior of the tunnel, obtaining a target line which is located in the tunnel and extends along the tunnel, enabling the inspection trolley to inspect the interior of the tunnel in the extending direction of the target line, changing the current navigation state of the unmanned aerial vehicle into a target tracking mode, and enabling the unmanned aerial vehicle to obtain position information through the inspection trolley in the target tracking mode.

Description

Technical Field

[0001] The present invention relates to the technical field of tunnel inspection, and particularly relates to a method and system for dynamic planning of unmanned aerial vehicle (UAV) navigation based on an inspection vehicle. Background Art

[0002] Due to the structural characteristics and terrain limitations of tunnels, the complexity and difficulty of their inspection are particularly prominent among the inspections of various transportation infrastructure. With the maturity of UAV technology, UAVs have been widely used in rescue, monitoring and other scenarios in recent years. UAVs have high flexibility and mobility, can easily cope with the complex environment in tunnels, and have obvious advantages in the tunnel inspection operation scenario due to their low cost, small and flexible equipment, and high operation convenience. However, there is no GNSS signal in tunnels, and the scene repetition rate is high and the geometric features of the movement direction are few, resulting in difficulty for traditional LiDAR SLAM algorithms to complete positioning. Therefore, at present, tunnel inspection generally needs to be assisted by wheeled vehicles, odometers, RFID, track markers, etc. to complete positioning.

[0003] However, the above methods can only achieve positioning in tunnels and cannot meet the actual intelligent requirements of tunnel inspection. The UAV also needs to complete autonomous path planning and flight in the tunnel, so as to realize one-key inspection and improve the tunnel inspection efficiency. Among them, UAVs that can fly stably in tunnels are usually limited by the aircraft size and battery technology, and the flight duration is generally 15 - 30 minutes, which is difficult to meet the needs of long-tunnel autonomous inspection. Summary of the Invention

[0004] The purpose of the present invention is to provide a method and system for dynamic planning of UAV navigation based on an inspection vehicle. A navigation method is set for the UAV on the inspection vehicle to enable it to perform autonomous path planning and flight. Moreover, with emphasis on the energy consumption problem of the UAV, different navigation methods are set for the UAV, so that the UAV can automatically select and switch according to the energy consumption situation to meet the needs of long-tunnel autonomous inspection.

[0005] To solve the above technical problems, the present invention adopts the following solutions: A method for dynamic planning of UAV navigation based on an inspection vehicle, where the inspection vehicle inspects the interior of a tunnel, and an UAV chassis and an UAV for charging the UAV are arranged on the inspection vehicle. The dynamic planning method includes the following steps: S1. Obtain the task volume of the current inspection vehicle for inspecting the interior of the tunnel; S2. During the process of the current inspection vehicle inspecting the tunnel, match the navigation state of the current UAV. If the navigation state of the current UAV is the global positioning mode, and the UAV obtains position information through multi-sensor fusion in the global positioning mode, then go to step S3; S3. Detect whether the remaining power of the current UAV has dropped to a preset threshold. If so, go to step S4; S4. Obtain the first flight distance and the first power consumption generated by the UAV during the current inspection of the tunnel by the inspection vehicle. According to the first flight distance and the first power consumption, determine in real time whether the current UAV can complete the task volume in the global positioning mode. If not, go to step S5; S5. Based on the target line obtained inside the tunnel that extends along the tunnel, the inspection vehicle inspects the inside of the tunnel in the extension direction of the target line, then change the navigation state of the current UAV to the target tracking mode. In the target tracking mode, the UAV obtains the position information through the inspection vehicle.

[0006] Further, in S2, the process of the UAV obtaining the position information through multi-sensor fusion in the global positioning mode is as follows: In the global positioning mode, when the UAV detects a defect inside the tunnel, the position information of the UAV is obtained through the fusion positioning between the relative pose sensor carried by the UAV itself and the inertial sensor, lidar, and wheel odometer sensor carried on the inspection vehicle.

[0007] Further, in S5, the process of the UAV obtaining the position information through the inspection vehicle in the target tracking mode is as follows: In the target tracking mode, the inspection vehicle inspects the inside of the tunnel in the extension direction of the target line. When the UAV detects a defect inside the tunnel, the forward distance of the inspection vehicle is obtained through the wheel odometer carried on the inspection vehicle, and the position information of the UAV is obtained according to the forward distance of the inspection vehicle.

[0008] Further, the task volume includes the tunnel inspection range. During the current inspection of the tunnel by the inspection vehicle, if the navigation state of the current UAV is the global positioning mode, the UAV will independently perform path planning according to the tunnel inspection range to obtain the first target path of the UAV, and use the first target path as the navigation for flight.

[0009] Further, the first flight distance is the flight distance generated when the UAV uses the first target path as the navigation for flight in the global positioning mode, and the first power consumption is the power drop generated when the UAV uses the first target path as the navigation for flight in the global positioning mode.

[0010] Further, in S4, the process of determining in real time whether the current UAV can complete the task volume in the global positioning mode according to the first flight distance and the first power consumption is specifically as follows: Calculate the remaining first target path of the current UAV based on the first target path corresponding to the first flown distance and the first power consumption, and fly with the remaining first target path as the navigation. Estimate the power consumption drop when the UAV flies with the remaining first target path as the navigation in the global positioning mode, and compare the power consumption drop with the remaining power of the current UAV. If the power consumption drop is less than the remaining power of the current UAV, it is determined that the current UAV can complete the task volume in the global positioning mode, and vice versa.

[0011] Further, it further includes step S6, and the step S6 is: S6. When the navigation state of the current UAV is the target tracking mode, obtain the second flown distance and the second power consumption generated by the UAV during the current inspection vehicle's inspection of the tunnel. According to the second flown distance and the second power consumption, determine in real time whether the current UAV can complete the task volume in the target tracking mode. If not, control the UAV to navigate back to the UAV chassis.

[0012] Further, the task volume includes the tunnel inspection range. During the current inspection vehicle's inspection of the tunnel, if the navigation state of the current UAV is the target tracking mode, the UAV will plan a path according to the tunnel inspection range and the target line to obtain the second target path of the UAV. The second target path coincides with the target line, and the second target path is used as the navigation for flight.

[0013] Further, in S6, the process of determining in real time whether the current UAV can complete the task volume in the target tracking mode according to the second flown distance and the second power consumption is specifically: Calculate the remaining second target path of the current UAV based on the second target path corresponding to the second flown distance and the second power consumption, and fly with the remaining second target path as the navigation. Estimate the power consumption drop when the UAV flies with the remaining second target path as the navigation in the target tracking mode, and compare the power consumption drop with the remaining power of the current UAV. If the power consumption drop is less than the remaining power of the current UAV, it is determined that the current UAV can complete the task volume in the target tracking mode, and vice versa.

[0014] A UAV navigation dynamic programming system based on an inspection vehicle, applying the described UAV navigation dynamic programming method based on an inspection vehicle, includes: An inspection task volume acquisition module: Obtain the task volume of the current inspection vehicle for inspecting the inside of the tunnel; UAV Global Positioning Module: During the current inspection of the tunnel by the inspection vehicle, match the navigation state of the current UAV. If the navigation state of the current UAV is the global positioning mode, the UAV obtains position information through multi-sensor fusion in the global positioning mode; Battery Detection Module: Detect whether the remaining battery power of the current UAV drops to a preset threshold; UAV Navigation Judgment Module: Obtain the first flight distance and the first power consumption generated during the current inspection of the tunnel by the inspection vehicle, and judge in real time whether the current UAV can complete the task volume in the global positioning mode according to the first flight distance and the first power consumption; UAV Navigation Switching Module: According to the target line obtained inside the tunnel and extending along the tunnel inside the tunnel, the inspection vehicle inspects the inside of the tunnel in the extension direction of the target line, then change the navigation state of the current UAV to the target tracking mode, and the UAV obtains position information through the inspection vehicle in the target tracking mode.

[0015] Advantages of the present invention: The present invention provides a dynamic path planning method and system for UAV navigation based on an inspection vehicle, which is mainly applied to tunnel inspection. The inspection vehicle and the UAV are used to detect defects inside the tunnel. The inspection vehicle and the UAV can communicate with each other to achieve cooperative positioning, facilitating the accurate location of defects and improving the inspection efficiency. However, in this scenario, UAVs are usually limited by aircraft size and battery technology, resulting in a general flight duration of 15 - 30 minutes, which is difficult to meet the needs of autonomous inspection of long tunnels. Therefore, the present invention focuses on setting different navigation methods for the inspection vehicle and the UAV, which can not only enable the UAV to perform autonomous path planning and flight, but also enable the UAV to automatically select and switch according to the energy consumption situation, meeting the needs of autonomous inspection of long tunnels. Description of the Drawings

[0016] Figure 1 It is a schematic diagram of the steps of the dynamic path planning method in Embodiment 1 of the present invention; Figure 2 It is a schematic diagram of the process of integrating the APF and BRRT* algorithms in Embodiment 1 of the present invention; Figure 3 It is a schematic diagram of the process of flying along a fixed tracking object in Embodiment 1 of the present invention; Figure 4 It is a schematic diagram of the process of flying along a fixed axis in Embodiment 1 of the present invention; Figure 5 It is a schematic diagram of the process of the UAV returning to the unmanned vehicle for charging and autonomous navigation in Embodiment 1 of the present invention; Figure 6 It is a schematic diagram of the overall process of the dynamic path planning method in Embodiment 1 of the present invention. Detailed implementation mode

[0017] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings 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 the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and in no way limits the present invention and its application or use. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work belong to the scope of protection of the present invention.

[0018] Unless otherwise specifically stated, the relative arrangements, numerical expressions and numerical values of the components and steps set forth in these embodiments do not limit the scope of the present invention.

[0019] At the same time, it should be understood that for the sake of convenience of description, the dimensions of each part shown in the drawings are not drawn according to the actual proportional relationship.

[0020] In addition, for the sake of clarity and conciseness, the descriptions of well-known structures, functions and configurations may be omitted. Those of ordinary skill in the art will recognize that various changes and modifications can be made to the examples described herein without departing from the spirit and scope of the present disclosure.

[0021] The techniques, methods and devices known to those of ordinary skill in the relevant fields may not be discussed in detail, but where appropriate, the said techniques, methods and devices should be regarded as part of the authorization specification.

[0022] In all the examples shown and discussed here, any specific value should be interpreted as merely exemplary, rather than as a limitation. Therefore, other examples of the exemplary embodiments may have different values.

[0023] The present invention will be described in detail below by referring to the accompanying drawings and in conjunction with the embodiments: Embodiment 1 Due to the structural characteristics of the tunnel and terrain limitations, the complexity and difficulty of its inspection are particularly prominent among the inspections of many transportation infrastructure. With the maturity of UAV technology, UAVs have been widely used in rescue, monitoring and other scenarios in recent years. UAVs have high flexibility and mobility, can easily cope with the complex environment in the tunnel, and have low costs, small and flexible equipment, and high operation convenience, showing obvious advantages in the tunnel inspection operation scenario. However, there is no GNSS signal in the tunnel, and the scene has a high degree of repetition and few geometric features of the moving direction, resulting in difficulty for traditional LiDAR SLAM algorithms to complete positioning. Therefore, at present, tunnel inspections generally need to be assisted by wheeled vehicles, odometers, RFID, track markers, etc. to complete positioning.

[0024] However, the above method can only achieve positioning inside the tunnel and cannot meet the actual intelligent requirements of tunnel inspection. The drone still needs to complete autonomous path planning and flight inside the tunnel, so as to achieve one-key inspection and improve the efficiency of tunnel inspection. Among them, drones that can fly stably in the tunnel are usually limited by the aircraft size and battery technology, and the flight duration is generally 15 - 30 minutes, which is difficult to meet the needs of long-tunnel autonomous inspection.

[0025] Therefore, in this embodiment, a dynamic planning method for drone navigation based on an inspection vehicle is proposed. The inspection vehicle inspects the inside of the tunnel, and an unmanned aircraft chassis and a drone for charging the drone are arranged on the inspection vehicle. As Figure 1 shown, the method includes the following steps: S1. Obtain the task volume of the current inspection vehicle for inspecting the inside of the tunnel; S2. During the process of the current inspection vehicle inspecting the tunnel, match the navigation state of the current drone. If the navigation state of the current drone is the global positioning mode, and the drone obtains position information through multi-sensor fusion in the global positioning mode, then go to step S3; S3. Detect whether the remaining power of the current drone drops to a preset threshold. If so, go to step S4; S4. Obtain the first flown distance and the first power consumption generated by the drone during the process of the current inspection vehicle inspecting the tunnel. According to the first flown distance and the first power consumption, judge in real time whether the current drone can complete the task volume in the global positioning mode. If not, go to step S5; S5. Obtain a target line located inside the tunnel and extending along the tunnel according to the inside of the tunnel. The inspection vehicle inspects the inside of the tunnel in the extending direction of the target line, then change the navigation state of the current drone to the target tracking mode, and the drone obtains position information through the inspection vehicle in the target tracking mode.

[0026] In one embodiment, in S2, the process of the drone obtaining position information through multi-sensor fusion in the global positioning mode is as follows: In the global positioning mode, when the drone detects a defect inside the tunnel, the position information of the drone is obtained through the fusion positioning between the relative pose sensor carried by the drone itself and the inertial sensor, lidar, and wheeled odometry sensor carried on the inspection vehicle.

[0027] The global positioning mode is an autonomous path planning method for the drone. In the global positioning mode, the drone can achieve positioning inside the tunnel through multi-sensor fusion technology, accurately master its position information inside the tunnel, and when detecting a defect during inspection, lock the defect position through the three-dimensional coordinates of the drone. Therefore, in this embodiment, the drone can first perform autonomous path planning in the global positioning mode. Here, the APF and BRRT* algorithms are fused, and an energy consumption optimization objective is introduced to achieve the collaborative optimization of path safety and energy efficiency. Specifically, as Figure 2 shown.

[0028] Step 1: Construct a potential field through APF. The potential field includes a gravitational field and a repulsive field. The gravitational field attracts the drone to move towards the target point, and the repulsive field repels the drone away from obstacles. Affected by gravity and repulsion, the drone is driven to move towards the target point and avoid obstacles.

[0029] Step 2: Optimize path generation through RT* bidirectional tree. Two random trees are generated simultaneously from the starting point and the ending point, and the tree expansion is guided by the gravitational field (starting point) and the repulsive field (ending point).

[0030] Step 3: Optimize energy consumption and collision risk. The fusion of APF and BRRT* algorithms performs well in avoiding obstacles and quickly generating a relatively optimal path, but does not consider the energy consumption optimization objective, and the path generated in a complex environment may not be smooth enough. Therefore, this step screens and smooths the above-generated path, and the multi-objective cost function uses the gradient descent function to solve each item as follows: ; Among them, is the multi-objective cost function, is the path length term, using the Euclidean distance or the reachable distance under dynamic constraints; is the energy consumption term, representing the energy consumption from the current node to the next node; is the safety term, representing the quantified collision risk, respectively represent the weights of each item.

[0031] Step 4: Apply constraints. To ensure the safe flight of the drone, parameters such as its heading angle, altitude, and flight distance can be constrained to avoid collisions with obstacles inside the tunnel or other situations. Among them, the heading angle constraint: the change in the heading angle between adjacent waypoints does not exceed , the fixed altitude flight altitude h or width w of the drone, and the safety distance constraint: the distance D from the tunnel facilities or obstacles > Dsafe. The above constraints can be adjusted accordingly according to the specific inspection scenario.

[0032] In one embodiment, in S5, the process by which the UAV obtains position information through the inspection vehicle in the target tracking mode is as follows: In the target tracking mode, the inspection vehicle inspects the interior of the tunnel in the extension direction of the target line. When the UAV detects a defect inside the tunnel during inspection, the forward distance of the inspection vehicle is obtained through the wheel odometer carried on the inspection vehicle, and the position information of the UAV is obtained based on the forward distance of the inspection vehicle.

[0033] The target tracking mode is another autonomous path planning method for the UAV. Specifically, the target tracking mode refers to a reactive flight mode. When the remaining battery power cannot complete the inspection task, it switches to a more power-saving reactive flight mode. The reactive flight mode does not require obtaining the positioning information of the UAV, and requires fewer types of sensors and computing resources. The inspection vehicle that cooperates with it for positioning is used to position the UAV. Specifically, the forward distance of the inspection vehicle on the target line can be obtained through the wheel mileage sensor of the inspection vehicle, and the forward distance of the inspection vehicle on the target line is directly used as the forward distance of the UAV on the target line to obtain the coordinate of the UAV on the z-axis, which can meet the basic inspection requirements.

[0034] In one embodiment, when the remaining battery power cannot complete the inspection task, it switches to a more power-saving reactive flight mode. In the reactive flight mode, the corresponding mode can be selected according to whether there are continuous markers inside the tunnel. Then, the target line in the reactive flight mode is determined according to the interior of the tunnel. When there are continuous markers inside the tunnel, for example, a road tunnel has road marking lines, a railway tunnel has tracks, and a municipal pipeline has a continuous pipe gallery, a virtual target line can be generated according to the continuous markers, and flying along the virtual target line is equivalent to flying along a fixed tracking object. When there are no continuous markers inside the tunnel, for example, a mine tunnel, a fire passage, etc., a fixed axis can be used as the target line, and the fixed axis can be the central axis of the mine tunnel, so that the UAV flies along the central axis.

[0035] Specifically, flying along a fixed tracking object is as Figure 3 shown. First, image preprocessing is performed on the images collected in the target tracking mode, including grayscale conversion, Gaussian filtering for denoising, and histogram equalization to enhance the contrast. Then, for edge processing, the Canny algorithm can be used to extract the edge of the lane line, and the edge position is determined through the gradient amplitude and direction. Then, through the Hough transform, the continuous marking lines in the image space are mapped to the parameter space to detect straight lines or curves. Finally, through the improved YOLOv4-tiny algorithm for real-time tracking detection, flying along a fixed tracking object can be achieved.

[0036] Specifically, flying along a fixed axis is as Figure 4As shown, first, lidar is used to obtain lidar point cloud data, then the centroid coordinates are calculated, and then the initial yaw angle is calculated. Through negative feedback calculation, the yaw angle at the nth point is obtained. , where the axial direction of the tunnel can be defined as the Z direction, that is, the forward direction of the UAV. At the same time, the horizontal left and right directions of the transverse section of the tunnel are defined as the Y direction. Taking the cross-sectional point cloud scanned by a single-line lidar as an example, the yaw angle is calculated.

[0037] In one embodiment, the task amount includes the tunnel inspection range. During the inspection of the tunnel by the current inspection vehicle, if the navigation state of the current UAV is the global positioning mode, the UAV will autonomously perform path planning according to the tunnel inspection range, obtain the first target path of the UAV, and use the first target path as navigation to fly. Specifically, the first target path refers to the virtual line formed by the forward flight distance of the UAV during the autonomous path planning of the UAV according to the tunnel inspection range. At this time, the UAV can plan the path of the UAV in real time according to the tunnel inspection range and the global positioning information of the UAV, so that the UAV flies along it to achieve UAV navigation.

[0038] In one embodiment, the first flight distance is the flight distance generated when the UAV uses the first target path as navigation in the global positioning mode, and the first power consumption is the power drop generated when the UAV uses the first target path as navigation in the global positioning mode.

[0039] In one embodiment, in S4, the process of determining in real time whether the current UAV can complete the task amount in the global positioning mode according to the first flight distance and the first power consumption is specifically as follows: Calculate the remaining first target path of the current UAV according to the first target path corresponding to the first flight distance and the first power consumption, use the remaining first target path as navigation to fly, estimate the power drop generated when the UAV uses the remaining first target path as navigation in the global positioning mode, and compare the power drop with the remaining power of the current UAV. If the power drop is less than the remaining power of the current UAV, it is determined that the current UAV can complete the task amount in the global positioning mode, and vice versa.

[0040] Specifically, in this embodiment, since the UAV on the inspection vehicle needs to complete the inspection task, the forward distance required for the UAV to complete the inspection task can be obtained in advance according to the inspection task. According to the first flight distance of the UAV and the forward distance required for the UAV to complete the inspection task, the remaining first target path of the current UAV, that is, the remaining forward distance, can be obtained.

[0041] In one embodiment, it further includes step S6, and the step S6 is as follows: S6. When the navigation state of the current drone is the target tracking mode, obtain the second flown distance and the second power consumption generated by the drone during the current inspection vehicle's inspection of the tunnel. Based on the second flown distance and the second power consumption, determine in real time whether the current drone can complete the task volume in the target tracking mode. If not, control the drone to navigate back to the drone chassis.

[0042] In one embodiment, the task volume includes the tunnel inspection range. During the current inspection vehicle's inspection of the tunnel, if the navigation state of the current drone is the target tracking mode, the drone will perform path planning based on the tunnel inspection range and the target line to obtain the second target path of the drone. The second target path coincides with the target line and is used as the navigation for flight.

[0043] Specifically, the second target path refers to the virtual line formed by the forward distance of the drone tracking and flying along the target line during the process of the drone performing path planning based on the tunnel inspection range and the target line. At this time, this virtual line is equivalent to a part of the target line, and the drone can be navigated based on the tunnel inspection range and the target line.

[0044] In one embodiment, in S6, the process of determining in real time whether the current drone can complete the task volume in the target tracking mode based on the second flown distance and the second power consumption is specifically as follows: Calculate the remaining second target path of the current drone based on the second target path corresponding to the second flown distance and the second power consumption, and use the remaining second target path as the navigation for flight. Estimate the power consumption drop generated when the drone uses the remaining second target path as the navigation for flight in the target tracking mode, and compare this power consumption drop with the remaining power of the current drone. If this power consumption drop is less than the remaining power of the current drone, it is determined that the current drone can complete the task volume in the target tracking mode, and vice versa.

[0045] Specifically, in this embodiment, the drone can obtain in advance the forward distance required for the drone to complete the inspection task according to the inspection task. Based on the second flown distance of the drone and the forward distance required for the drone to complete the inspection task, the remaining second target path of the current drone, that is, the remaining forward distance, can be obtained.

[0046] In one embodiment, during the process of determining in real time whether the current drone can complete the task volume in the global positioning mode based on the first flown distance and the first power consumption, and during the process of determining in real time whether the current drone can complete the task volume in the target tracking mode based on the second flown distance and the second power consumption, the wind resistance information collected during the flight process and the remaining path length can also be combined to determine whether the remaining power of the drone can complete the inspection task in this mode. And if it is determined that the remaining power can complete the flight task after switching the mode, the navigation mode is switched.

[0047] In one embodiment, since the drone needs to autonomously navigate back to the unmanned vehicle for charging. The drone and the unmanned vehicle cooperate to complete the inspection task in the tunnel. In the area where the unmanned vehicle can reach, the unmanned vehicle and the drone travel almost at the same pace. In the area where the unmanned vehicle cannot reach, the drone needs to fly independently. The path planning method for the drone to return to the unmanned vehicle needs to fully consider the above two scenarios. In addition, it should be noted that during the flight process, the drone and the unmanned vehicle always maintain communication, and the drone can obtain the positioning of the unmanned vehicle in real time. Specifically, the process of the drone returning to the unmanned vehicle for charging and autonomous navigation is as Figure 5 shown. First, it is judged whether the drone landing area is within the visual range of the drone. If so, it lands through MPC + visual servo, and then locks through magnetic attraction; if not, it locates the inspection vehicle in real time and performs dynamic path planning based on the positioning.

[0048] In summary, the present invention provides a dynamic planning method for drone navigation based on an inspection vehicle. The overall process of the dynamic planning method is as Figure 6 shown. First, autonomous path planning considering energy consumption is carried out in the global positioning mode, and then it is judged whether the remaining power of the drone in the global positioning mode can complete the task. If so, path planning continues in the global positioning mode; if not, it is identified whether there are continuous markers inside the tunnel. If so, it tracks and flies following the continuous markers. If not, it flies along the fixed axis. At the same time, the navigation mode is switched to the target tracking mode, and then it is judged whether the current remaining power of the drone in the target tracking mode can complete the inspection mode. If not, the drone returns to the drone box for charging.

[0049] Embodiment 2 A dynamic planning system for drone navigation based on an inspection vehicle, applying the dynamic planning method for drone navigation based on an inspection vehicle as described above, includes: An inspection task volume acquisition module: obtaining the task volume of the current inspection vehicle for inspecting the inside of the tunnel; A drone global positioning module: during the process of the current inspection vehicle inspecting the tunnel, matching the navigation state of the current drone. If the navigation state of the current drone is the global positioning mode, the drone obtains position information through multi-sensor fusion in the global positioning mode; Battery detection module: Detect whether the remaining battery power of the current drone has dropped to a preset threshold; Drone navigation judgment module: Obtain the first flight distance and the first power consumption generated during the current drone's inspection of the tunnel by the inspection vehicle, and determine in real time whether the current drone can complete the task volume in the global positioning mode according to the first flight distance and the first power consumption; Drone navigation switching module: According to the target line obtained inside the tunnel and extending along the tunnel inside the tunnel, when the inspection vehicle inspects the inside of the tunnel in the extension direction of the target line, change the navigation state of the current drone to the target tracking mode, and the drone obtains the position information through the inspection vehicle in the target tracking mode.

[0050] The above are only the preferred embodiments of the present invention, and do not impose any form of limitation on the present invention. Based on the technical essence of the present invention, any simple modifications, equivalent replacements, and improvements made to the above embodiments within the spirit and principles of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A dynamic programming method for UAV navigation based on an inspection vehicle, characterized in that, The inspection vehicle conducts inspections on the interior of the tunnel. An unmanned aerial vehicle (UAV) case for charging the UAV and the UAV are provided on the inspection vehicle. The method includes the following steps: S1. Obtain the task volume of the current inspection vehicle for inspecting the interior of the tunnel; S2. During the process of the current inspection vehicle inspecting the tunnel, match the navigation state of the current UAV. If the navigation state of the current UAV is the global positioning mode and the UAV obtains position information through multi-sensor fusion in the global positioning mode, then proceed to step S3; S3. Detect whether the remaining power of the current UAV has dropped to a preset threshold. If so, then proceed to step S4; S4. Obtain the first flight distance and the first power consumption generated by the UAV during the process of the current inspection vehicle inspecting the tunnel. Based on the first flight distance and the first power consumption, determine in real time whether the current UAV can complete the task volume in the global positioning mode. If not, then proceed to step S5; S5. Based on the interior of the tunnel, obtain a target line located inside the tunnel and extending along the tunnel. The inspection vehicle inspects the interior of the tunnel in the extending direction of the target line, and then change the navigation state of the current UAV to the target tracking mode. In the target tracking mode, the UAV obtains position information through the inspection vehicle.

2. The method for dynamic programming of UAV navigation based on an inspection vehicle according to claim 1, wherein, In S2, the process of the UAV obtaining position information through multi-sensor fusion in the global positioning mode is as follows: In the global positioning mode, when the UAV detects a defect inside the tunnel during inspection, the position information of the UAV is obtained through the fusion positioning between the relative pose sensor carried by the UAV itself and the inertial sensor, lidar, and wheel odometer sensor carried on the inspection vehicle.

3. A method for dynamic planning of UAV navigation based on an inspection vehicle according to claim 1, characterized in that, In S5, the process of the UAV obtaining position information through the inspection vehicle in the target tracking mode is as follows: In the target tracking mode, the inspection vehicle inspects the interior of the tunnel in the extending direction of the target line. When the UAV detects a defect inside the tunnel during inspection, the forward distance of the inspection vehicle is obtained through the wheel odometer carried on the inspection vehicle, and the position information of the UAV is obtained based on the forward distance of the inspection vehicle.

4. A dynamic programming method for UAV navigation based on an inspection vehicle according to claim 1, characterized in that, The task volume includes the tunnel inspection range. During the process of the current inspection vehicle inspecting the tunnel, if the navigation state of the current UAV is the global positioning mode, the UAV will independently perform path planning according to the tunnel inspection range to obtain the first target path of the UAV, and use the first target path as navigation for flight.

5. The method for dynamic planning of UAV navigation based on an inspection vehicle according to claim 4, wherein The first flight distance is the flight distance generated when the UAV uses the first target path as navigation in the global positioning mode, and the first power consumption is the power reduction amount generated when the UAV uses the first target path as navigation in the global positioning mode.

6. A method for dynamic planning of UAV navigation based on an inspection vehicle according to claim 5, characterized in that, In S4, the process of determining in real time whether the current UAV can complete the task volume in the global positioning mode based on the first flight distance and the first power consumption is specifically as follows: Calculate the remaining first target path of the current UAV based on the first target path corresponding to the first flown distance and the first power consumption, fly with the remaining first target path as the navigation, estimate the power drop when the UAV flies with the remaining first target path as the navigation in the global positioning mode, compare the power drop with the remaining power of the current UAV, if the power drop is less than the remaining power of the current UAV, it is determined that the current UAV can complete the task volume in the global positioning mode, and vice versa.

7. A method for dynamic planning of UAV navigation based on an inspection vehicle according to claim 1, characterized in that, It further includes step S6, and the step S6 is: S6. When the navigation state of the current UAV is the target tracking mode, obtain the second flown distance and the second power consumption generated by the UAV during the current inspection vehicle's inspection of the tunnel, and determine in real time whether the current UAV can complete the task volume in the target tracking mode according to the second flown distance and the second power consumption. If not, control the UAV to navigate back to the UAV chassis.

8. A method for dynamic programming of UAV navigation based on an inspection vehicle according to claim 7, characterized in that, The task volume includes the tunnel inspection range. During the current inspection vehicle's inspection of the tunnel, if the navigation state of the current UAV is the target tracking mode, the UAV will perform path planning according to the tunnel inspection range and the target line to obtain the second target path of the UAV. The second target path coincides with the target line, and the second target path is used as the navigation for flight.

9. A method for dynamic programming of UAV navigation based on an inspection vehicle according to claim 8, characterized in that, In S6, the process of determining in real time whether the current UAV can complete the task volume in the target tracking mode according to the second flown distance and the second power consumption is specifically: Calculate the remaining second target path of the current UAV based on the second target path corresponding to the second flown distance and the second power consumption, fly with the remaining second target path as the navigation, estimate the power drop when the UAV flies with the remaining second target path as the navigation in the target tracking mode, compare the power drop with the remaining power of the current UAV, if the power drop is less than the remaining power of the current UAV, it is determined that the current UAV can complete the task volume in the target tracking mode, and vice versa.

10. An unmanned aerial vehicle navigation dynamic programming system based on an inspection vehicle, characterized in that Applying a UAV navigation dynamic programming method based on an inspection vehicle as described in any one of claims 1-9, including: An inspection task volume acquisition module: obtain the inspection task volume of the current inspection vehicle for the interior of the tunnel; A UAV global positioning module: During the current inspection vehicle's inspection of the tunnel, match the navigation state of the current UAV. If the navigation state of the current UAV is the global positioning mode, the UAV obtains position information through multi-sensor fusion in the global positioning mode; A power detection module: detect whether the remaining power of the current UAV drops to a preset threshold; A UAV navigation judgment module: obtain the first flown distance and the first power consumption generated by the UAV during the current inspection vehicle's inspection of the tunnel, and determine in real time whether the current UAV can complete the task volume in the global positioning mode according to the first flown distance and the first power consumption; UAV Navigation Switching Module: Based on the information obtained inside the tunnel, a target line located inside the tunnel and extending along the tunnel is determined. The inspection vehicle conducts inspections inside the tunnel in the extension direction of the target line. Then, the navigation state of the current UAV is changed to the target tracking mode. In the target tracking mode, the UAV obtains position information through the inspection vehicle.

Images