Dynamic route planning method and system for power grid inspection

Abstract

The invention discloses a dynamic route planning method and system for power grid inspection, and relates to the technical field of intelligent inspection, and the technical scheme is characterized in that an airborne end identifies and positions real-time data to obtain a flight waypoint; the airborne end carries out waypoint filtering to obtain inspection waypoints; the airborne end is added to obtain a guide waypoint; and the airborne end uploads the inspection waypoint and the guide waypoint to the cloud end, and executes an instruction of tracking the inspection waypoint according to the real-time flight condition of the monitoring loading end. According to the invention, an inspection mode at the speed of 4-8m / s can be realized, only a pan-tilt camera needs to be carried, no requirement on a laser radar is needed, the cost is more advantageous, the onboard weight can be reduced, and the inspection endurance and the inspection efficiency can be remarkably improved for a small unmanned aerial vehicle.

Description

Technical Field

[0001] The invention patent relates to the field of intelligent inspection technology, and specifically, to a dynamic route planning method and system for power grid inspection. Background Art

[0002] Transmission lines are the backbone of the power grid. With the increase in the length of transmission lines, especially the rapid development of ultra-high voltage and UHV lines, the daily inspection of transmission lines has become more challenging. As a new and efficient inspection method, drones are used in daily inspections. However, during drone inspections, at least two staff members are usually required to work together to complete the inspection task. One operator is responsible for operating the drone platform, and the other operator is responsible for operating the airborne pod or gimbal to complete the observation and collection of target information. This inspection mode, on the one hand, requires close cooperation between the two operators, and has high requirements on the operator's operating skills and labor intensity; on the other hand, the control is affected by communication delays, and the effect of target information collection is often poor.

[0003] In order to reduce labor intensity and improve the effectiveness of collected information, intelligent collection systems and methods are urgently needed to realize the automatic collection of transmission line information. Therefore, automatic inspection by drones has the advantages of low technical requirements, easy operation and low labor costs, and has broad application prospects in the inspection field. The optoelectronic payloads carried by drones can provide real-time detection, pole tower tracking, real-time pole tower positioning and other services.

[0004] At present, common intelligent inspection methods include the following:

[0005] (1) Using laser radar to detect power lines and towers in real time, the drone is automatically controlled to fly along the lines. This method is better than manual inspection, but its disadvantages are low flight speed, low operating efficiency, and high equipment cost.

[0006] (2) For example, in a Chinese patent (CN107729808B), an intelligent image acquisition system and method for UAV inspection of power transmission lines are proposed. The system uses a visual algorithm to identify the direction of poles and wires and collects pictures for inspection. This solution has high requirements for camera configuration and requires multiple hovering during the operation. The disadvantages are low flight speed, low operation efficiency and high equipment cost.

[0007] At the same time, in the existing technology, the only method that can truly achieve uniform-speed intelligent inspection is the real-time point cloud inspection method of laser radar. However, this method currently cannot support inspections at speeds above 4m / s, and is relatively expensive. The onboard radar equipment also reduces the flight time of the UAV; therefore, the present invention aims to provide a dynamic route planning method and system for power grid inspection, which can achieve inspections at speeds of 4 to 8m / s. It only requires a gimbal camera and has no requirements for laser radar. It is not only more cost-effective, but also can reduce the onboard weight. For small UAVs, this can significantly improve the inspection flight time and efficiency. Summary of the invention

[0008] The purpose of the present invention is to provide a dynamic route planning method and system for power grid inspection. The present invention can realize inspection at a speed of 4 to 8 m / s. It only needs to be equipped with a gimbal camera and has no requirement for a laser radar. It is not only more cost-effective, but also can reduce the airborne weight. For small unmanned aerial vehicles, this can significantly improve the inspection flight time and efficiency.

[0009] The present invention is implemented as follows: a dynamic route planning method for power grid inspection, the method comprising:

[0010] The real-time data collected is identified and located by the airborne terminal to obtain the flight waypoints after identification and location;

[0011] The flight waypoints that have been identified and located are filtered by the airborne terminal to obtain the inspection waypoints in the actual inspection direction;

[0012] The airborne terminal continues to add waypoints along the actual inspection direction according to the inspection waypoints in the actual inspection direction, and obtains the guidance waypoints in the actual inspection direction;

[0013] The inspection waypoints in the actual inspection direction and the guidance waypoints in the actual inspection direction are uploaded to the cloud through the airborne terminal, and the real-time flight status of the airborne terminal is monitored. When the real-time flight status of the airborne terminal during its flight corresponds to a current inspection waypoint that is different from the published inspection waypoint, the airborne terminal executes an instruction to track the inspection waypoint.

[0014] Furthermore, the airborne terminal identifies and locates the collected real-time data to obtain the identified and located flight waypoints, including:

[0015] The airborne terminal refreshes the collected real-time data at a set frequency and performs positioning identification to obtain the flight waypoints after identification and positioning.

[0016] Furthermore, the airborne terminal filters the flight waypoints obtained after identification and positioning to obtain the inspection waypoints in the actual inspection direction, including:

[0017] The airborne terminal combines the identified and located flight waypoint with the position information of the airborne terminal to obtain the distance between the identified and located flight waypoint and the airborne terminal. The airborne terminal filters the identified and located flight waypoint according to the distance between the identified and located flight waypoint and the airborne terminal.

[0018] The airborne end filters the first flight waypoint after identification and positioning based on the distance between the flight waypoint after identification and positioning and the airborne end, and combines the second flight waypoint after identification and positioning with the airborne end according to the distance between the flight waypoint after identification and positioning and the airborne end to obtain the distance between the first flight waypoint and the second flight waypoint. The airborne end filters the flight waypoint after identification and positioning based on the distance between the first flight waypoint and the second flight waypoint.

[0019] Further, when the airborne end performs waypoint filtering on the flight waypoints obtained after identification and positioning, the airborne end calculates the angle between its yaw angle and the vector formed by the current inspection waypoint and the published inspection waypoint, and obtains the angle between the yaw angle calculated by the airborne end and the vector formed by the current inspection waypoint and the published inspection waypoint. The airborne end filters the flight waypoints obtained after identification and positioning according to the angle between its yaw angle and the vector formed by the current inspection waypoint and the published inspection waypoint.

[0020] Wherein, when the current inspection waypoint of the airborne terminal is an initial inspection waypoint, the airborne terminal uses its current position information as the current inspection waypoint information.

[0021] Furthermore, the method further comprises:

[0022] When the airborne end filters the flight waypoints after identification and positioning, the airborne end calculates the angle between the speed direction and the vector formed by the position information and the published inspection waypoint, and obtains the angle between the speed direction and the vector formed by the position information and the published inspection waypoint calculated by the airborne end. The airborne end filters the flight waypoints after identification and positioning according to the angle between the speed direction and the vector formed by the position information and the published inspection waypoint;

[0023] Wherein, when the airborne terminal is in a hovering state during its flight, the airborne terminal uses its current inspection waypoint as a published inspection waypoint.

[0024] Further, the airborne end adds a guide waypoint according to the inspection waypoints obtained in the actual inspection direction to obtain the guide waypoints in the actual inspection direction, including:

[0025] The airborne end obtains the guide waypoint based on the current inspection waypoint in the actual inspection direction and the extended set distance of the published inspection waypoint, the inspection waypoint includes the current inspection waypoint and the published inspection waypoint, and the airborne end calculates the position information of the guide waypoint based on the position information of the current inspection waypoint in the actual inspection direction and the position information of the published inspection waypoint.

[0026] Furthermore, the situation in which the current inspection waypoint used by the airborne terminal during its flight is different from the published inspection waypoint includes:

[0027] When the real-time flight status of the airborne terminal during its flight corresponds to the initial flight status, the airborne terminal executes the instruction of tracking and publishing the inspection waypoints;

[0028] When the real-time flight status of the airborne terminal during its flight corresponds to a hovering state, the airborne terminal executes an instruction to track and publish a patrol waypoint;

[0029] When the real-time flight status of the airborne terminal during its flight corresponds to the flight tracking state, the airborne terminal executes the instruction of tracking and publishing the inspection waypoint;

[0030] When the real-time flight status of the airborne terminal during its flight corresponds to the flight turning state, the airborne terminal will guide the waypoint to the published inspection waypoint, and the airborne terminal will point the published inspection waypoint to the guide waypoint, and the airborne terminal executes the instruction to track and publish the inspection waypoint.

[0031] The present invention also provides a dynamic route planning system for power grid inspection, the dynamic route planning system comprising:

[0032] The identification module is used to identify and locate the real-time data collected during the flight and obtain the flight waypoints after identification and positioning;

[0033] A filtering module is used to filter the flight waypoints that have been identified and located to obtain the inspection waypoints in the actual inspection direction;

[0034] A guidance module, used to add a guidance waypoint to the inspection waypoint in the actual inspection direction, so as to obtain the guidance waypoint in the actual inspection direction;

[0035] The tracking module is used to upload the inspection waypoints and the guidance waypoints in the actual inspection direction to the cloud, and monitor the real-time flight status of the airborne terminal during its flight. When the real-time flight status of the airborne terminal during its flight corresponds to a current inspection waypoint that is different from the published inspection waypoint, the airborne terminal executes the instruction to track the inspection waypoint.

[0036] As an optional implementation, in a dynamic route planning system for power grid inspection provided by the present invention, the transmission module identifies and locates the real-time data collected during the flight according to the airborne terminal to obtain the identified and located flight waypoints, including:

[0037] The airborne terminal updates the real-time data collected during the flight at a set frequency and performs positioning identification to obtain the flight waypoints after identification and positioning;

[0038] As an optional implementation, in a dynamic route planning system for power grid inspection provided by the present invention, the filtering module filters the flight waypoints identified and located by the airborne terminal to obtain the inspection waypoints in the actual inspection direction, including:

[0039] The airborne terminal combines the identified and located flight waypoint with the position information of the airborne terminal to obtain the distance between the identified and located flight waypoint and the airborne terminal. The airborne terminal filters the identified and located flight waypoint according to the distance between the identified and located flight waypoint and the airborne terminal.

[0040] The airborne end filters the first flight waypoint after identification and positioning based on the distance between the flight waypoint after identification and positioning and the airborne end, and combines the second flight waypoint after identification and positioning with the airborne end according to the distance between the flight waypoint after identification and positioning and the airborne end to obtain the distance between the first flight waypoint and the second flight waypoint. The airborne end filters the flight waypoint after identification and positioning based on the distance between the first flight waypoint and the second flight waypoint.

[0041] As an optional implementation, in a dynamic route planning system for power grid inspection provided by the present invention, when the filtering module performs waypoint filtering on the airborne end for the flight waypoints that have been identified and located, the airborne end calculates the angle between its yaw angle and the vector formed by the current inspection waypoint and the published inspection waypoint, and obtains the angle between the yaw angle calculated by the airborne end and the vector formed by the current inspection waypoint and the published inspection waypoint. The airborne end filters the flight waypoints that have been identified and located according to the angle between its yaw angle and the vector formed by the current inspection waypoint and the published inspection waypoint;

[0042] Wherein, when the current inspection waypoint of the airborne terminal is an initial inspection waypoint, the airborne terminal uses its position information as the current inspection waypoint information.

[0043] As an optional implementation, in a dynamic route planning system for power grid inspection provided by the present invention, when the filtering module filters the flight waypoints after identification and positioning at the airborne end, the airborne end calculates the angle between its speed direction and the vector formed by its position information and the published inspection waypoint, and obtains the angle between its speed direction and the vector formed by its position information and the published inspection waypoint calculated by the airborne end, and the airborne end filters the flight waypoints after identification and positioning according to the angle between its speed direction and the vector formed by its position information and the published inspection waypoint;

[0044] Wherein, when the airborne terminal is in a hovering state during its flight, the airborne terminal uses its current inspection waypoint as a published inspection waypoint.

[0045] As an optional implementation, in a dynamic route planning system for power grid inspection provided by the present invention, the airborne terminal adds a guide waypoint according to the inspection waypoints obtained in the actual inspection direction to obtain the guide waypoints in the actual inspection direction, including:

[0046] The airborne end obtains the guide waypoint based on the current inspection waypoint in the actual inspection direction and the extended set distance of the published inspection waypoint, the inspection waypoint includes the current inspection waypoint and the published inspection waypoint, and the airborne end calculates the position information of the guide waypoint based on the position information of the current inspection waypoint in the actual inspection direction and the position information of the published inspection waypoint.

[0047] As an optional implementation, in a dynamic route planning system for power grid inspection provided by the present invention, when the real-time flight status of the airborne terminal during its flight corresponds to a current inspection waypoint that is different from the published inspection waypoint, the airborne terminal executes an instruction to track the inspection waypoint, including:

[0048] When the real-time flight status of the airborne terminal during its flight corresponds to the current inspection waypoint as the initial waypoint, the airborne terminal executes the instruction of tracking and publishing the inspection waypoint;

[0049] The real-time flight status of the airborne terminal during its flight corresponds to a hovering state or when tracking a guiding waypoint, the airborne terminal executes an instruction to track and publish a patrol waypoint;

[0050] When the real-time flight status of the airborne terminal during its flight corresponds to the current inspection waypoint being a published inspection waypoint, the airborne terminal executes an instruction to track and publish the inspection waypoint;

[0051] When the real-time flight status of the airborne terminal during its flight corresponds to a guidance waypoint that is different from the published inspection waypoint, the airborne terminal points the guidance waypoint to the published inspection waypoint, and the airborne terminal points the published inspection waypoint to the guidance waypoint, and the airborne terminal executes the instruction to track the published inspection waypoint.

[0052] Furthermore, the dynamic route planning system further includes an airborne terminal, the airborne terminal includes a device body, and the airborne terminal further includes:

[0053] A memory storing executable program code;

[0054] a processor coupled to the memory;

[0055] The processor calls the executable program code stored in the memory to execute the steps performed by the airborne end in a dynamic route planning method for power grid inspection.

[0056] Furthermore, the dynamic route planning system further includes a cloud, the cloud includes a first end, and the first end includes:

[0057] A memory storing executable program code;

[0058] a processor coupled to the memory;

[0059] The processor calls the executable program code stored in the memory to execute the steps performed by the first end in a dynamic route planning method for power grid inspection.

[0060] Furthermore, the dynamic route planning system also includes a computer storage medium, wherein the computer storage medium stores computer instructions, and when the computer instructions are called, a dynamic route planning method for power grid inspection is executed.

[0061] Compared with the prior art, the present invention has the following beneficial effects:

[0062] 1. The present invention uses dynamic route planning to enable drones to conduct uniform and efficient inspections at a speed of 4 to 8 m / s, significantly improving the inspection speed and efficiency. Compared with traditional manual inspections and low-speed drone inspection methods, the inspection time is greatly reduced. At the same time, this method only requires a PTZ camera and does not require a laser radar, so it has a cost advantage. This reduces the weight of the aircraft, and for small drones, it can significantly improve the inspection flight time and inspection efficiency, while reducing equipment costs and maintenance costs;

[0063] 2. The present invention does not need to carry laser radar equipment, thereby reducing the load of the UAV, which is especially important for small UAVs, and can further increase the flight time and inspection range of the UAV. At the same time, the present invention can ensure that the UAV maintains the correct route during the inspection process through waypoint filtering and the increase of guided waypoints, reduce deviations caused by image recognition errors or positioning accuracy problems, and improve the precision and accuracy of power grid inspection data;

[0064] 3. The present invention can adapt to different flight speeds and environmental conditions, and dynamically adjust the waypoint update frequency and waypoint filtering logic, so that the UAV can flexibly cope with the complex and changeable power grid environment. At the same time, the method reduces the need for manual intervention through intelligent route planning, reduces the requirements for operator skills, reduces the possibility of human error, and also reduces the workload of operators;

[0065] 4. The present invention enables the drone to conduct autonomous inspections through dynamic route planning. At the same time, the route planning form adopted by the method provides adaptability for subsequent expansion work above the tower, so that more inspection tasks or services can be added on this basis, such as tower maintenance, fault diagnosis, etc.;

[0066] 5. The present invention can carry out stable power grid inspection work, is not limited by environmental factors, and ensures the continuity and reliability of the inspection work. At the same time, through dynamic route planning, the drone can get closer to the target and collect higher resolution image data, providing higher quality original data for subsequent data analysis and processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 It is a flow chart of a dynamic route planning method for power grid inspection provided by an embodiment of the present invention;

[0068] Figure 2 is a schematic diagram of waypoint filtering provided by an embodiment of the present invention;

[0069] Figure 3 is a schematic diagram of a route planning process provided by an embodiment of the present invention;

[0070] Figure 4 It is a data flow diagram of route planning provided by an embodiment of the present invention;

[0071] Figure 5 It is a structural schematic diagram of a dynamic route planning system for power grid inspection provided by an embodiment of the present invention;

[0072] Figure 6 is a structural schematic diagram of an airborne terminal provided by an embodiment of the present invention;

[0073] Figure 7It is a schematic diagram of a cloud structure provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0074] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0075] The implementation of the present invention is described in detail below in conjunction with specific embodiments.

[0076] The same or similar numbers in the drawings of this embodiment correspond to the same or similar parts; in the description of the present invention, it should be understood that if the terms "upper", "lower", "left", "right" and the like indicate directions or positional relationships based on the directions or positional relationships shown in the drawings, it is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction. Therefore, the terms describing the positional relationship in the drawings are only used for illustrative purposes and cannot be understood as limitations on this patent. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to specific circumstances.

[0077] Reference Figure 1-7 The figure shows a preferred embodiment of the present invention.

[0078] Example 1: A dynamic route planning method for power grid inspection

[0079] 1) The airborne end identifies and locates the collected real-time data to obtain the flight waypoints after identification and positioning.

[0080] In this embodiment, after receiving the take-off command, the flight control controls the drone to take off. After take-off, the waypoint update frequency will be adapted according to the flight speed:

[0081] When the flight speed is less than 6.5, the waypoints are published every 5 meters;

[0082] When the flight speed is greater than 6.5 and less than 8.5, it is released every 6 meters;

[0083] When the flight speed is greater than 8.5 and less than 10.5, it is released every 7 meters;

[0084] When the flight speed is greater than 10.5 meters, it is released every 9 meters.

[0085] In this embodiment, a more accurate flight speed measurement sensor is introduced, and a high-precision speed measurement module based on a laser radar is used, with a measurement accuracy of up to millimeter level, so that the flight speed of the UAV can be obtained more accurately in real time, thereby further refining the adaptation strategy of the waypoint update frequency. In actual applications, when the flight speed fluctuates around 6.5 m / s, it can be subdivided into intervals of 0.1 m / s according to its specific fluctuation range, and each interval corresponds to a more accurate waypoint update interval, such as every 4.8 meters at 6.4-6.5 m / s, every 5.2 meters at 6.5-6.6 m / s, and so on, so that the waypoint release is more in line with the actual flight status of the UAV.

[0086] 2) The airborne end will filter the flight waypoints after identification and positioning to obtain the inspection waypoints in the actual inspection direction.

[0087] After receiving the tower positioning, this embodiment initially performs waypoint filtering, and the filtering rules are as follows:

[0088] If the published waypoint is less than 5 meters or more than 100 meters away from the current position of the drone, the waypoint will be filtered;

[0089] To prevent the drone from tracking waypoints too frequently, if the published waypoint is less than 1 meter away from the last published waypoint, the waypoint will be filtered;

[0090] Initially filter waypoints. When a new waypoint is released (or the drone is about to fly to the current waypoint and needs to refresh the waypoint), the system will calculate the angle between the drone's yaw angle and the vector formed by the current waypoint (if it is the first time to release waypoint No. 11, use the drone's position as the current waypoint) and the new waypoint. If the angle is between 60 degrees and 300 degrees, it is considered that there is a large deviation between the new waypoint and the current route direction. In order to avoid the drone turning back for inspection, such waypoints need to be filtered out, such as Figure 2 shown.

[0091] In this embodiment, the tower positioning data is preprocessed by using a deep learning algorithm to train an image recognition model specifically for power grid towers. The model can automatically identify and eliminate data with large tower positioning errors caused by environmental interference (such as tree occlusion, light and shadow changes, etc.), thereby improving the accuracy of tower positioning before waypoint filtering. The model can identify positioning offsets caused by tower reflections under direct strong light and correct them, making the waypoints after preliminary filtering more accurate and reliable.

[0092] 3) The airborne end continues to add waypoints along the actual inspection direction according to the inspection waypoints in the actual inspection direction to obtain the guidance waypoints in the actual inspection direction.

[0093] In this embodiment, after waypoint filtering, the system will publish waypoint n pointing to guide point 199. Guide point 199 is a hovering waypoint. After the drone reaches point 199, it will hover there, just in front of the inspection direction. Waypoint 11 is the first waypoint, and its guide point is obtained by extending the drone position and waypoint 11 by 20 meters. The guide points of other waypoints are obtained by extending the current tracked waypoint and the previous waypoint by 20 meters.

[0094] In the guidance point release link, this embodiment uses advanced flight control algorithms, including a flight control strategy based on model predictive control (MPC). When the UAV flies to the guidance point 199 and hovers, it can control the hovering position and attitude more accurately. The MPC algorithm can predict the flight status of the UAV in the future in real time, and adjust the control instructions in advance accordingly, so that when the UAV reaches the guidance point 199, it can achieve precise hovering with a smaller error, providing a more stable starting state for subsequent waypoint tracking. When the UAV approaches the guidance point 199, the MPC algorithm will calculate the optimal deceleration and hovering control instructions in advance according to factors such as the current flight speed, acceleration, and distance from the guidance point, so as to reduce the hovering error from the original centimeter level to within the millimeter level.

[0095] After the waypoint is released, when it is about to be tracked, the waypoint filtering is performed again. The system will calculate the angle between the drone speed direction (if the drone is already in a hovering state, the vector direction formed by waypoint 11 and the drone's current position will be used) and the vector formed by the drone's current position and the waypoint to be tracked. If the angle is between 60 and 300 degrees, it is considered that the waypoint to be tracked is in the opposite direction of the current route. In order to avoid the drone turning back for inspection, such waypoints need to be filtered out.

[0096] 4) The airborne terminal will upload the inspection waypoints and guidance waypoints in the actual inspection direction to the cloud, and monitor the real-time flight status of the airborne terminal. When the current inspection waypoints based on the airborne terminal during its flight are different from the published inspection waypoints, the airborne terminal executes the instruction to track the inspection waypoints.

[0097] In this embodiment, after the secondary waypoint filtering, the system will issue waypoint tracking instructions according to the following situations:

[0098] If the current waypoint is 11, the system will track to waypoint 11;

[0099] When the drone is in hovering state, or when tracking 199 waypoints, a new waypoint is released, and the system will also track to the new waypoint.

[0100] When the waypoint tracked by the drone is the same as the currently published waypoint, it means that the currently tracked waypoint is being refreshed and the system will track the currently tracked waypoint.

[0101] When the waypoint m tracked by the drone is different from the currently published waypoint n, it means that the waypoint published at this time is the next tower, not the tower corresponding to the waypoint currently being tracked. The system points waypoint m to waypoint n, and waypoint n to waypoint 199. At the same time, it will track waypoint m and refresh the status of waypoint m for the flight control.

[0102] In the continuous waypoint tracking link, the data transmission speed and stability between the airborne end and the cloud will be greatly improved. In this embodiment, the low latency and high bandwidth characteristics of the 5G network are utilized, and the airborne end can upload the collected tower positioning data and its own flight status data to the cloud in real time and at high speed, and quickly receive the waypoint update instructions issued by the cloud. In this way, during the inspection process, even if a complex power grid environment or sudden changes in the tower are encountered, the system can respond quickly and adjust the waypoint planning in time to ensure that the drone always inspects according to the optimal route. If a new obstacle is found near a certain section of the power grid line during the inspection, the airborne end can immediately upload the relevant information to the cloud. The cloud completes the waypoint re-planning within a few milliseconds and sends the new waypoint instruction to the airborne end. The drone then continues to inspect along the new route to bypass the obstacle. The whole process is seamless, which greatly improves the flexibility and safety of the inspection.

[0103] Subsequently, the positioning of the tower is continuously received, and the above preliminary waypoint filtering is repeated to track the waypoints in the waypoint tracking logic step, and finally the inspection route is planned.

[0104] The route planning schematic diagram in this embodiment is as follows Figure 3 As shown, by taking the present invention as an example, in an inspection process, the final published route and flight trajectory, the inspection speed is 4m / s, the white circle represents the real position of the tower, the blue circle represents the waypoint tracked during the inspection, the deviation along the waypoint is within 5 meters, the UAV gradually inspects the tower in front according to the waypoint guidance, and hovers at the 199 waypoint after the mission is completed, and the inspection of the tower can be realized; continue to use the inspection route of the inspection test with a speed of 8m / s, the UAV gradually inspects the tower in front according to the waypoint guidance, and still hovers at the 199 waypoint after the mission is completed.

[0105] This embodiment uses dynamic route planning to enable the drone to perform uniform and efficient inspections at a speed of 4 to 8 m / s, significantly improving the inspection speed and efficiency. Compared with traditional manual inspections and low-speed drone inspection methods, the inspection time is greatly reduced. At the same time, this method only requires a gimbal camera and does not require a laser radar, so it has a cost advantage. This reduces the weight of the aircraft, and for small drones, it can significantly improve the inspection flight time and inspection efficiency, while reducing equipment costs and maintenance costs.

[0106] Example 2: A dynamic route planning system for power grid inspection

[0107] In this embodiment, the dynamic route planning system includes an airborne terminal, which includes:

[0108] The identification module is used to identify and locate the real-time data collected during the flight and obtain the flight waypoints after identification and positioning;

[0109] A filtering module is used to filter the flight waypoints that have been identified and located to obtain the inspection waypoints in the actual inspection direction;

[0110] A guidance module, used to add a guidance waypoint to the inspection waypoint in the actual inspection direction, so as to obtain the guidance waypoint in the actual inspection direction;

[0111] The tracking module is used to upload the inspection waypoints and the guidance waypoints in the actual inspection direction to the cloud, and monitor the real-time flight status of the airborne terminal during its flight. When the real-time flight status of the airborne terminal during its flight corresponds to the current inspection waypoints and the published inspection waypoints, the airborne terminal executes the instruction to track the inspection waypoints.

[0112] The dynamic route planning system of this embodiment is as follows:

[0113] In the dynamic waypoint refresh process in this embodiment, the airborne end is responsible for receiving the positioning results, and the waypoint planning system refreshes the waypoints at a frequency of approximately 1 Hz. At the same time, by combining with a satellite navigation enhancement system (such as the satellite-based augmentation service of the Beidou satellite navigation system), the accuracy and reliability of waypoint planning are further improved. The satellite navigation enhancement system can provide the UAV with more accurate positioning information, so that the waypoint planning system can determine the position and flight trajectory of the UAV with centimeter or even millimeter level accuracy when refreshing the waypoints. In complex terrain environments such as mountainous areas, traditional satellite navigation signals may be blocked and interfered, resulting in reduced positioning accuracy. With the help of the satellite navigation enhancement system, even in these environments, the UAV can obtain accurate positioning data, thereby ensuring the accuracy of waypoint planning, allowing the UAV to fly accurately along the predetermined route and successfully complete the inspection mission.

[0114] In the inspection waypoint filtering process of this embodiment, in order to prevent the ID from changing or identifying a distant tower due to image recognition, thereby interfering with the inspection route, if the published waypoint is less than 5 meters or greater than 100 meters from the current position of the drone, the waypoint will be filtered; to prevent the drone from tracking waypoints too frequently, if the published waypoint is less than 1 meter away from the last published waypoint, the waypoint will be filtered; preliminarily filter the waypoints, when a new waypoint is published (or the drone is about to fly to the current waypoint and needs to refresh the waypoint), the system will calculate the angle between the drone's yaw angle and the vector formed by the current waypoint (if it is the first time to publish waypoint No. 11, the drone position is used as the current waypoint) and the new waypoint. If the angle is between 60 degrees and 300 degrees, it is considered that there is a large deviation between the new waypoint and the current route direction. In order to avoid the drone turning back for inspection, such waypoints need to be filtered out.

[0115] The inspection waypoint filtering process in this embodiment also includes a secondary waypoint filtering link. By introducing an anomaly detection algorithm to filter the waypoint data more intelligently, it is possible to learn a large amount of normal inspection waypoint data. The anomaly detection algorithm can establish a normal distribution model of waypoint data. In the actual inspection process, when new waypoint data is generated, the algorithm will compare it with the normal distribution model, quickly identify abnormal waypoints caused by image recognition errors, sensor failures, etc., and filter them out. If the coordinate value of a certain waypoint deviates far from the normal distribution range, or the distance, angle and other characteristics between it and the adjacent waypoints are too different from the characteristics of the normal waypoints, , the anomaly detection algorithm will determine it as an abnormal waypoint and filter it, thereby effectively preventing the drone from deviating from the normal inspection route due to tracking abnormal waypoints, improving the stability and safety of the inspection. At the same time, before issuing each track command, the system will calculate the angle between the drone speed direction (if the drone is already in a hovering state, the vector direction formed by waypoint 11 and the current position of the drone will be used) and the vector formed by the current position of the drone and the waypoint to be tracked. If the angle is between 60 and 300 degrees, it is considered that the waypoint to be tracked is in the opposite direction of the current route. In order to avoid the drone returning for inspection, such waypoints need to be filtered out.

[0116] In the guidance waypoint link, this embodiment adopts a guidance waypoint generation strategy based on reinforcement learning. The reinforcement learning algorithm will automatically learn and generate the optimal guidance waypoint according to the flight status, environmental information and inspection mission objectives of the UAV. During the inspection process, the algorithm will adjust the position and direction of the guidance waypoint in real time according to the current speed, altitude, posture of the UAV and the position and distance of the front tower and other factors, so that the UAV can fly to the next inspection target in the most energy-saving and efficient way. At the same time, the reinforcement learning algorithm can also continuously optimize the guidance waypoint generation strategy. As the inspection mission continues, the generated guidance waypoints will increasingly meet the actual flight requirements, further improving the inspection efficiency and quality of the UAV.

[0117] The guide point is obtained by extending the last published waypoint and the currently published waypoint by 20 meters. For the first waypoint, the current position of the drone (x1, y1) and the currently published waypoint (x2, y2) are used to extend 20 meters.

[0118] First, calculate the equation of the line from (x1, y1) and (x2, y2):

[0119]

[0120] b=y1-k×x1

[0121] Then calculate the projection of the extension length L on the x-axis:

[0122]

[0123] Then calculate the coordinates of the guide point:

[0124] x 199 =x2+x L

[0125] y 199 =k×x 199 +b

[0126] In the waypoint tracking link of this embodiment, the waypoint tracking instruction is issued according to the following situations:

[0127] If the current waypoint is 11, the system will track to waypoint 11;

[0128] When the drone is in hovering state or tracking 199 waypoints, a new waypoint is released and the system will also track to the new waypoint;

[0129] When the waypoint tracked by the drone is the same as the currently published waypoint, it means that the currently tracked waypoint is being refreshed, and the system will track the currently tracked waypoint;

[0130] When the waypoint m tracked by the drone is different from the currently published waypoint n, it means that the waypoint published at this time is the next tower, not the tower corresponding to the waypoint currently being tracked. The system points waypoint m to waypoint n, and waypoint n to waypoint 199. At the same time, it will track waypoint m and refresh the status of waypoint m for the flight control.

[0131] At the same time, in the waypoint tracking link, this embodiment uses virtual reality (VR) and augmented reality (AR) technology to provide drone operators with more intuitive and real-time flight monitoring and waypoint tracking assistance. Operators can use VR equipment to immersively observe the flight environment and inspection status of the drone. At the same time, AR technology can overlay waypoint information, flight trajectory, obstacle warning and other data in the operator's field of view in real time. When the drone approaches a tower, the AR system will highlight the waypoint corresponding to the tower in the operator's field of view, and use virtual lines to clearly indicate the flight direction and distance of the drone, helping the operator to more accurately judge the flight status of the drone, and promptly discover and deal with possible waypoint tracking deviation problems, thereby improving the convenience and accuracy of operation, and also providing more powerful manual intervention support for the autonomous flight of the drone.

[0132] The route planning flow chart in this embodiment is as follows Figure 4 As shown, the present invention is suitable for high-speed power grid inspection technology. During the flight of the drone, the pole tower waypoints can be continuously adjusted to achieve dynamic planning. Compared with previous technologies, the present method has the advantages of low cost and high inspection efficiency. Under the condition of stable visual inspection, the inspection speed can be higher than 4 meters / s. In addition, the present method adopts route planning for inspection, which also provides adaptability for subsequent expansion work above the tower.

[0133] Embodiment 3: An airborne terminal includes a device body, and the airborne terminal further includes:

[0134] A memory storing executable program code;

[0135] a processor coupled to the memory;

[0136] The processor calls the executable program code stored in the memory to execute the steps executed by the airborne end in a dynamic route planning method for power grid inspection in Embodiment 1 of the present invention, the dynamic route planning method comprising: the airborne end identifies and locates the real-time data collected during its flight to obtain flight waypoints after identification and positioning; the airborne end filters the flight waypoints after identification and positioning to obtain inspection waypoints in the actual inspection direction; the airborne end adds guide waypoints according to the inspection waypoints in the actual inspection direction to obtain guide waypoints in the actual inspection direction;

[0137] The airborne terminal will upload the inspection waypoints and guidance waypoints in the actual inspection direction to the cloud, and monitor the real-time flight status of the airborne terminal during its flight. When the real-time flight status of the airborne terminal during its flight corresponds to the current inspection waypoints and the published inspection waypoints, the airborne terminal executes the instruction to track the inspection waypoints.

[0138] In addition to storing executable program codes, the onboard memory of this embodiment is also equipped with a large-capacity non-volatile storage chip for storing a large amount of inspection data collected by the drone during flight, such as high-definition images, videos, sensor data, etc. In addition, a new type of phase change memory or magnetic memory can be used. These storage chips will have higher storage density, faster read and write speeds, and longer service life. After the inspection is completed, the onboard terminal can quickly transmit this data to the cloud for further analysis and processing, providing more comprehensive and detailed data support for the maintenance and management of the power grid.

[0139] In terms of processors, a multi-core heterogeneous processor architecture is adopted, combined with a dedicated flight control chip and an artificial intelligence acceleration chip. The flight control chip is responsible for the basic flight control tasks of the UAV, such as attitude adjustment, speed control, etc., while the artificial intelligence acceleration chip is specifically used to process complex computing tasks such as image recognition, waypoint filtering, and path planning. This multi-core heterogeneous processor architecture can give full play to the advantages of each chip and improve the overall computing performance and energy efficiency of the airborne end.

[0140] Embodiment 4: A cloud comprising:

[0141] A memory storing executable program code;

[0142] a processor coupled to the memory;

[0143] The processor calls the executable program code stored in the memory to execute the steps executed by the cloud in a dynamic route planning method for power grid inspection in Embodiment 1 of the present invention, wherein the dynamic route planning method comprises: the airborne end identifies and locates the real-time data collected during its flight to obtain the identified and located flight waypoints; the airborne end filters the identified and located flight waypoints to obtain the inspection waypoints in the real inspection direction; the airborne end adds guide waypoints according to the inspection waypoints in the real inspection direction to obtain the guide waypoints in the real inspection direction;

[0144] The airborne terminal will upload the inspection waypoints and guidance waypoints in the actual inspection direction to the cloud, and monitor the real-time flight status of the airborne terminal during its flight. When the real-time flight status of the airborne terminal during its flight corresponds to the current inspection waypoints and the published inspection waypoints, the airborne terminal executes the instruction to track the inspection waypoints.

[0145] Embodiment 5: A computer storage medium stores computer instructions. When the computer instructions are called, a dynamic route planning method for power grid inspection in Embodiment 1 of the present invention is executed. The dynamic route planning method includes: the airborne end identifies and locates the real-time data collected during its flight to obtain flight waypoints after identification and location; the airborne end filters the flight waypoints after identification and location to obtain inspection waypoints in the actual inspection direction; the airborne end adds guide waypoints according to the inspection waypoints in the actual inspection direction to obtain guide waypoints in the actual inspection direction;

[0146] The airborne terminal will upload the inspection waypoints and guidance waypoints in the actual inspection direction to the cloud, and monitor the real-time flight status of the airborne terminal during its flight. When the real-time flight status of the airborne terminal during its flight corresponds to the current inspection waypoints and the published inspection waypoints, the airborne terminal executes the instruction to track the inspection waypoints.

[0147] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A dynamic route planning method for power grid inspection, characterized in that: The method comprises: The collected real-time data is identified and located by the airborne terminal to obtain the flight waypoints after identification and location; The flight waypoints that have been identified and located are filtered by the airborne terminal to obtain the inspection waypoints in the actual inspection direction; The airborne terminal continues to add waypoints along the actual inspection direction according to the inspection waypoints in the actual inspection direction, and obtains the guidance waypoints in the actual inspection direction; The inspection waypoints in the actual inspection direction and the guidance waypoints in the actual inspection direction are uploaded to the cloud through the airborne terminal, and the real-time flight status of the airborne terminal is monitored. When the current inspection waypoints based on the airborne terminal during its flight are different from the published inspection waypoints, the airborne terminal executes the instruction to track the inspection waypoints.

2. A dynamic route planning method for power grid inspection according to claim 1, characterized in that: The airborne terminal identifies and locates the collected real-time data to obtain the identified and located flight waypoints, including: The airborne terminal refreshes the collected real-time data at a set frequency and performs positioning identification to obtain the flight waypoints after identification and positioning.

3. A dynamic route planning method for power grid inspection according to claim 1, characterized in that: The airborne terminal filters the flight waypoints obtained after identification and positioning to obtain the inspection waypoints in the actual inspection direction, including: The airborne terminal combines the identified and located flight waypoint with the position information of the airborne terminal to obtain the distance between the identified and located flight waypoint and the airborne terminal. The airborne terminal filters the identified and located flight waypoint according to the distance between the identified and located flight waypoint and the airborne terminal. The airborne end filters the first flight waypoint after identification and positioning based on the distance between the flight waypoint after identification and positioning and the airborne end, and combines the second flight waypoint after identification and positioning with the airborne end according to the distance between the flight waypoint after identification and positioning and the airborne end to obtain the distance between the first flight waypoint and the second flight waypoint. The airborne end filters the flight waypoint after identification and positioning based on the distance between the first flight waypoint and the second flight waypoint.

4. A dynamic route planning method for power grid inspection according to claim 1, characterized in that: The method further comprises: When the airborne end performs waypoint filtering on the flight waypoints obtained after identification and positioning, the airborne end calculates the angle between its yaw angle and the vector formed by the current inspection waypoint and the published inspection waypoint, and obtains the angle between the yaw angle calculated by the airborne end and the vector formed by the current inspection waypoint and the published inspection waypoint. The airborne end filters the flight waypoints obtained after identification and positioning according to the angle between its yaw angle and the vector formed by the current inspection waypoint and the published inspection waypoint; Wherein, when the current inspection waypoint of the airborne terminal is an initial inspection waypoint, the airborne terminal uses its current position information as the current inspection waypoint information.

5. A dynamic route planning method for power grid inspection according to claim 1, characterized in that: The method further comprises: When the airborne end filters the flight waypoints after identification and positioning, the airborne end calculates the angle between the speed direction and the vector formed by the position information and the published inspection waypoint, and obtains the angle between the speed direction and the vector formed by the position information and the published inspection waypoint calculated by the airborne end. The airborne end filters the flight waypoints after identification and positioning according to the angle between the speed direction and the vector formed by the position information and the published inspection waypoint; Wherein, when the airborne terminal is in a hovering state during its flight, the airborne terminal uses its current inspection waypoint as a published inspection waypoint.

6. A dynamic route planning method for power grid inspection according to claim 1, characterized in that: The airborne end adds a guide waypoint according to the inspection waypoints obtained in the actual inspection direction to obtain the guide waypoints in the actual inspection direction, including: The airborne end obtains the guide waypoint based on the current inspection waypoint in the actual inspection direction and the extended set distance of the published inspection waypoint, the inspection waypoint includes the current inspection waypoint and the published inspection waypoint, and the airborne end calculates the position information of the guide waypoint based on the position information of the current inspection waypoint in the actual inspection direction and the position information of the published inspection waypoint.

7. A dynamic route planning method for power grid inspection according to claim 1, characterized in that: The situation where the current inspection waypoint used by the airborne terminal during its flight is different from the published inspection waypoint includes: When the real-time flight status of the airborne terminal during its flight corresponds to the initial flight status, the airborne terminal executes the instruction of tracking and publishing the inspection waypoints; When the real-time flight status of the airborne terminal during its flight corresponds to a hovering state, the airborne terminal executes an instruction to track and publish a patrol waypoint; When the real-time flight status of the airborne terminal during its flight corresponds to the flight tracking state, the airborne terminal executes the instruction of tracking and publishing the inspection waypoint; When the real-time flight status of the airborne terminal during its flight corresponds to the flight turning state, the airborne terminal will guide the waypoint to the published inspection waypoint, and the airborne terminal will point the published inspection waypoint to the guide waypoint, and the airborne terminal executes the instruction to track and publish the inspection waypoint.

8. A dynamic route planning system for power grid inspection, characterized in that: The dynamic route planning system comprises: The identification module is used to identify and locate the real-time data collected during the flight and obtain the flight waypoints after identification and positioning; A filtering module is used to filter the flight waypoints that have been identified and located to obtain the inspection waypoints in the actual inspection direction; A guidance module, used to add a guidance waypoint to the inspection waypoint in the actual inspection direction, so as to obtain the guidance waypoint in the actual inspection direction; The tracking module is used to upload the inspection waypoints and the guidance waypoints in the actual inspection direction to the cloud, and monitor the real-time flight status of the airborne terminal during its flight. When the real-time flight status of the airborne terminal during its flight corresponds to a current inspection waypoint that is different from the published inspection waypoint, the airborne terminal executes the instruction to track the inspection waypoint.

9. A dynamic route planning system for power grid inspection according to claim 8, characterized in that: The dynamic route planning system further includes an airborne terminal, the airborne terminal includes a device body, and the airborne terminal further includes: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the steps performed by the airborne end in the dynamic route planning method for power grid inspection as described in any one of claims 1-7.

10. A dynamic route planning system for power grid inspection according to claim 8, characterized in that: The dynamic route planning system further includes a cloud, the cloud includes a first end, and the first end includes: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the steps performed by the first end in the dynamic route planning method for power grid inspection as described in any one of claims 1-7.

11. A dynamic route planning system for power grid inspection according to claim 8, characterized in that: The dynamic route planning system also includes a computer storage medium, which stores computer instructions. When the computer instructions are called, a dynamic route planning method for power grid inspection as described in any one of claims 1 to 7 is executed.

Images