A medium and large UAV spraying system

Abstract

The present application discloses a medium-to-large-sized UAV spraying system, and the technical field to which it belongs is intelligent manufacturing and UAV spraying technology. The medium-to-large-sized UAV spraying system includes a ceiling rail suspension structure, a ceiling rail deformation monitoring device, and a robot posture optimization module. The ceiling rail suspension structure does not occupy ground space and can realize three-dimensional full coverage movement of the spraying robot in the working area. The ceiling rail deformation monitoring device can accurately monitor the deformation of the ceiling rail suspension structure. The robot posture optimization module can adjust the posture of the spraying robot, thereby reducing the spraying error caused by the deformation of the ceiling rail suspension structure and the abnormal posture of the UAV. The medium-to-large-sized UAV spraying system provided by the present application can realize efficient and accurate automated spraying of medium-to-large-sized UAVs while reducing the ground space occupied.

Description

Technical Field

[0001] The present application relates to the fields of intelligent manufacturing and drone spraying technology, and in particular to a medium to large drone spraying system. Background Art

[0002] Unmanned aerial vehicle, also known as UAV, is an unmanned aerial vehicle that is controlled by radio remote control equipment and its own program control device.

[0003] In order to improve the performance and service life of drones, drones often need to be sprayed. Medium and large drones usually have large sizes (e.g., wingspan not less than 20 meters, fuselage length not less than 11 meters, height not less than 4.1 meters), and complex surface curvature changes, which makes the spraying process of medium and large drones complex and difficult to achieve.

[0004] In the related art, medium and large UAVs are usually sprayed by manual spraying or robot automatic spraying.

[0005] The manual spraying method mainly relies on the operator's experience, so the manual spraying method has poor stability, high work intensity, harsh environment, low efficiency and unstable quality.

[0006] The robot automatic spraying method mainly uses ground rails, gantry rails, AGV (automatic guided vehicle) and turntables to drive the spraying robot to move along the spraying trajectory. The devices relied on by the above schemes need to lay tracks on the ground, install brackets or provide mobile platforms, which means that the running path must be reserved in the working area and ensure sufficient safety distance around. Due to the special structures of the back, belly and upper and lower surfaces of the wings of medium and large drones, the space for drone spraying needs to meet the requirements of unobstructed, easy maintenance and full coverage. However, the above-mentioned robot automatic spraying method not only occupies a large space and has low space utilization, but also has poor maintainability and poor spray accessibility.

[0007] Therefore, how to achieve efficient and accurate automated spraying for medium and large UAVs while reducing the ground space occupied is a technical problem that technicians in this field currently need to solve. Summary of the invention

[0008] The purpose of this application is to provide a medium-to-large UAV spraying system that can achieve efficient and accurate automated spraying of medium-to-large UAVs while reducing ground space occupation.

[0009] In order to solve the above technical problems, the present application provides a medium-to-large UAV spraying system, including: a ceiling rail suspension structure, a ceiling rail deformation monitoring device and a robot posture optimization module;

[0010] The overhead rail suspension structure includes an X-axis track, a Y-axis track, a Z-axis track and a spraying robot, wherein the X-axis track is fixedly installed on the top of the robot working area, the Y-axis track is installed on the X-axis track, the Z-axis track is suspended and installed on the Y-axis track, and the spraying robot is installed on the Z-axis track; the Y-axis track is used to slide along the X-axis track, the Z-axis track is used to slide along the Y-axis track, the spraying robot is used to slide along the Z-axis track, and the spraying robot is also used to spray the drone;

[0011] The overhead rail deformation monitoring device includes a sensor and a data processing module; the sensor is arranged on any one of the X-axis track, the Y-axis track and the Z-axis track or a combination of any several of the tracks; the data processing module is used to process the data collected by the sensor to obtain the deformation of the overhead rail suspension structure; wherein the sensor includes any one of a laser rangefinder, a strain sensor and a vibration sensor or a combination of any several of the above;

[0012] The robot posture optimization module is used to adjust the posture of the spraying robot according to the deformation of the overhead rail suspension structure and / or the posture error of the drone; wherein the posture error of the drone is determined based on the actual posture and standard posture of the drone.

[0013] Optionally, the Y-axis track includes a Y-axis track body, an X-axis slide and an X-axis driving assembly, the Y-axis track body is installed on the X-axis track through the X-axis slide, and the X-axis driving assembly is used to drive the Y-axis track body to move along the X-axis through the X-axis slide;

[0014] The Z-axis track includes a Z-axis track body, a Y-axis skateboard, a Y-axis drive assembly, a Z-axis skateboard and a Z-axis drive assembly. The Z-axis track body is suspended on the Y-axis track body through the Y-axis skateboard, and the Y-axis drive assembly is used to drive the Z-axis track body to move along the Y-axis through the Y-axis skateboard; the spraying robot is installed on the Z-axis track body through the Z-axis skateboard, and the Z-axis drive assembly is used to drive the spraying robot to move along the Z-axis through the Z-axis skateboard.

[0015] Optionally, a paint supply device is provided on the Z-axis track, and the paint supply device is connected to the spraying robot.

[0016] Optionally, if the sensor includes the laser rangefinder, the laser rangefinder is arranged at the midpoint of the track of the overhead rail type suspension structure, and / or at a point where the track is divided into three equal parts, and / or at a point where the track is divided into four equal parts;

[0017] If the sensor comprises a strain sensor, the strain sensor is arranged at the load-bearing part and / or the connecting part of the overhead rail type suspension structure;

[0018] If the sensor comprises a vibration sensor, the vibration sensor is arranged at a supporting portion and / or a connecting portion of the overhead rail type suspension structure.

[0019] Optionally, the overhead rail deformation monitoring device further includes an early warning module, which is used to control the overhead rail suspension structure to shut down and output early warning information when the deformation of the overhead rail suspension structure is greater than a threshold value.

[0020] Optionally, the data processing module includes:

[0021] A preprocessing unit, used for performing Kalman filtering on the data collected by the sensor to obtain filtered data;

[0022] The deformation analysis unit is used to process the filtered data using a long short-term memory artificial neural network to obtain deformation information; wherein the deformation information includes the current deformation amount and deformation trend of the overhead rail suspension structure.

[0023] Optionally, the data processing module includes:

[0024] The auxiliary adjustment unit is used to determine the deformation cause according to the deformation information and output a deformation adjustment strategy corresponding to the deformation cause.

[0025] Optionally, also include:

[0026] A visual detection device is arranged in the working area of ​​the robot, and is used to detect the actual position and posture of the drone.

[0027] Optionally, the robot posture optimization module includes:

[0028] An error interval determination unit, used to set a reference error interval according to the deformation of the overhead rail type suspension structure before starting the operation;

[0029] The error compensation unit is used to calculate the deviation matrix between the actual posture and the standard posture of the UAV when the overhead rail suspension structure is not in an operating state and the posture error of the UAV is within the reference error interval, and use the rotation parameters and translation parameters in the deviation matrix to adjust the posture of the spray robot.

[0030] Optionally, the robot posture optimization module further includes:

[0031] The error judgment unit is used to determine that the spray robot meets the prerequisites for spraying operations if the UAV posture error is less than the lower limit of the reference error interval; the error judgment unit is also used to determine that the spray robot does not meet the prerequisites for spraying operations if the UAV posture error is greater than the upper limit of the reference error interval, and generate a warning message.

[0032] Optionally, the robot posture optimization module includes:

[0033] The online compensation unit is used to adjust the posture of the spraying robot according to the deformation of the overhead rail type suspension structure at a current moment when the overhead rail type suspension structure is in an operating state.

[0034] The present application provides a medium-to-large UAV spraying system, including: an overhead rail suspension structure, an overhead rail deformation monitoring device and a robot posture optimization module. The above-mentioned overhead rail suspension structure includes an X-axis track, a Y-axis track, a Z-axis track and a spraying robot arranged at the top of the robot's working area. The overhead rail suspension structure does not occupy ground space and can realize three-dimensional full coverage movement of the spraying robot in the working area. The above-mentioned overhead rail deformation monitoring device can accurately monitor the deformation of the overhead rail suspension structure, and the robot posture optimization module can adjust the posture of the spraying robot according to the deformation of the overhead rail suspension structure and / or the UAV posture error, thereby reducing the spraying error caused by the deformation of the overhead rail suspension structure and the abnormal posture of the UAV. The present application can realize efficient and accurate automated spraying of medium-to-large UAVs while reducing the ground space occupied. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0036] Figure 1 A schematic diagram of an overhead rail suspension structure provided in an embodiment of the present application;

[0037] Figure 2 A schematic diagram of the structure of a Z-axis track of an overhead rail type suspension structure provided in an embodiment of the present application;

[0038] Figure 3 A schematic diagram of the layout of a sensor on a ceiling rail suspension structure provided in an embodiment of the present application;

[0039] Figure 4 A flow chart of a deformation monitoring and early warning method for an overhead rail type suspension structure provided in an embodiment of the present application;

[0040] Figure 5 A workflow diagram of a robot posture optimization module provided in an embodiment of the present application;

[0041] Figure 6 A flowchart of a robot posture optimization method with hierarchical compensation of multi-source errors provided in an embodiment of the present application. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0043] The drones in this article refer to medium and large drones, i.e. medium drones or large drones. Medium and large drones refer to drones with a wingspan of not less than 20 meters, a fuselage length of not less than 11 meters, and a height of not less than 4.1 meters. Due to the large size of medium and large drones, the requirements for the working area during the spraying process are higher. When there are many tracks, brackets, and mobile platforms on the ground in the working area, efficient and accurate spraying cannot be achieved.

[0044] In order to achieve efficient and accurate automated spraying of medium and large UAVs while reducing the ground space occupied, an embodiment of the present application provides a medium and large UAV spraying system. The above-mentioned medium and large UAV spraying system includes: a ceiling rail suspension structure, a ceiling rail deformation monitoring device and a robot posture optimization module.

[0045] The introduction of the overhead rail suspension structure in the spraying system of medium and large UAVs is as follows:

[0046] See also Figure 1 , Figure 1 This is a schematic diagram of an overhead rail suspension structure provided in an embodiment of the present application, in which 101 represents the X-axis rail, 102 represents the Y-axis rail, 103 represents the Z-axis rail, and 104 and 105 represent spraying robots. Figure 1The working area shown can also be provided with a robot cleaning room, a monitoring room and a matching air supply device. The above-mentioned overhead rail suspension structure includes an X-axis track, a Y-axis track, a Z-axis track, at least one spraying robot and corresponding connectors. The X-axis track is fixedly installed on the top of the robot working area, the Y-axis track is installed on the X-axis track, the Z-axis track is suspended and installed on the Y-axis track, and the spraying robot is installed on the Z-axis track; the Y-axis track is used to slide along the X-axis track, the Z-axis track is used to slide along the Y-axis track, the spraying robot is used to slide along the Z-axis track, and the spraying robot is also used to spray the drone. The above-mentioned overhead rail suspension structure includes an X-axis track, a Y-axis track, a Z-axis track and a spraying robot arranged at the top of the robot working area. This scheme adopts an overhead rail suspension structure for the layout of the spraying robot. Compared with the traditional ground guide rail type, gantry type and other layouts, the overhead rail suspension structure in this embodiment does not occupy the ground space and can realize the three-dimensional full coverage movement of the spraying robot in the working area, which meets the spraying requirements of complex curved surfaces of medium and large drones.

[0047] Specifically, the X-axis track is installed at the top of the robot's working area (such as a factory building) and is responsible for the lateral movement of the spray robot to ensure full coverage of the width of the spray area. The X-axis track can be equipped with welded steel components, drag chain components, square guide rails, and racks. The X-axis track is a fixed track with flat rails and racks. The X-axis track is connected to the pulley on the X-axis slide. The X-drive assembly drives the X-axis slide to move along the X-axis through the gear rack.

[0048] The Y-axis track is installed on the X-axis track through the X-axis slide and drive assembly, which is responsible for the longitudinal movement of the robot and covers the front and rear range of the spraying area. The Y-axis track is a mobile track, and its two ends are fixed on the X-axis slide. The Y-axis track is provided with a flat rail and a rack. Under the drive of the X drive assembly, the Y-axis track can move along the X-axis as a whole.

[0049] The Z-axis track is suspended under the Y-axis slide to control the vertical movement of the spray robot and provide the spray robot with the ability to move up and down. By adjusting the Z-axis displacement, it can ensure that the spray robot's nozzle maintains the best spraying distance with the target surface of the spraying area, thereby improving the spraying quality and efficiency.

[0050] The Z-axis track includes the Y-axis slide, drive assembly, drag chain assembly, linear rail and other parts. The drag chain assembly is responsible for combing and guiding the cables and pipelines inside the system to ensure that the lines are smooth and not entangled or damaged when the spray robot is running. The linear rail provides a high-precision linear motion path for the Z-axis to ensure that the Z-axis movement of the spray robot is accurate and smooth. The spray robot is firmly fixed on the Z-axis slide through the flange connecting plate. The connecting plate is precisely matched with the mounting hole at the bottom of the spray robot and is firmly connected by bolts to ensure the stability of the spraying operation.

[0051] The overhead rail suspension structure in this embodiment fixes the X-axis track on the top of the robot's working area. This installation method ensures that the X-axis track will not be displaced or vibrated unnecessarily, thereby enhancing the overall stability of the overhead rail suspension structure. In addition, the design of installing the X-axis track on the top makes full use of the vertical space of the working area, avoids occupying the ground space, and improves the space utilization of the working area and the flexibility of the spray robot. It can be seen that compared with the traditional ground installation method, the overhead rail suspension structure improves the space utilization of the spray workshop and the convenience of cleaning and maintenance.

[0052] The introduction of the sky rail deformation monitoring device in the medium and large UAV spraying system is as follows:

[0053] The overhead rail deformation monitoring device includes a sensor and a data processing module; the sensor is arranged on any one of the X-axis track, the Y-axis track and the Z-axis track or a combination of any several tracks. The sensor is connected to the data processing module and transmits the collected data to the data processing module. The above-mentioned data processing module is used to process the data collected by the sensor to obtain the deformation of the overhead rail suspension structure. Among them, the sensor includes any one or a combination of any several of a laser rangefinder, a strain sensor and a vibration sensor.

[0054] The laser rangefinder is used to measure the distance between itself and the reference object. The displacement of the overhead rail suspension structure can be determined based on the data collected by the laser rangefinder. The strain sensor is used to measure the surface strain of the object. The strain amount of the overhead rail suspension structure can be determined based on the data collected by the strain sensor. The vibration sensor is used to measure the vibration parameters. The vibration signal of the overhead rail suspension structure can be determined based on the data collected by the vibration sensor.

[0055] The data processing module can pre-process the data collected by the sensor, and use the machine learning model to calculate the pre-processed data to obtain the deformation of the overhead rail suspension structure. The pre-processing can include data cleaning, signal filtering, feature extraction, normalization and other operations. In this embodiment, an appropriate machine learning model can be selected according to the task requirements, such as a long short-term memory network (LSTM) or a convolutional neural network (CNN).

[0056] The introduction of the robot posture optimization module in the medium and large UAV spraying system is as follows:

[0057] The robot posture optimization module in the above-mentioned medium and large UAV spraying system is used to adjust the posture of the spraying robot so that the spraying robot is aligned with the medium and large UAV. Specifically, the robot posture optimization module can obtain the deformation of the overhead rail suspension structure from the data processing module, and can also determine the posture error of the medium and large UAV, and then adjust the posture of the spraying robot according to the deformation of the overhead rail suspension structure and / or the posture error of the UAV. The above-mentioned UAV posture error is determined based on the actual posture and standard posture of the UAV. The robot posture optimization module can adjust the posture of the spraying robot according to the deformation of the overhead rail suspension structure, can adjust the posture of the spraying robot according to the posture error of the UAV, and can also adjust the posture of the spraying robot according to the deformation of the overhead rail suspension structure and the posture error of the UAV.

[0058] The above-mentioned overhead rail deformation monitoring device can accurately monitor the deformation of the overhead rail suspension structure, and the robot posture optimization module can adjust the posture of the spraying robot according to the deformation of the overhead rail suspension structure and / or the posture error of the drone, thereby reducing the spraying error caused by the deformation of the overhead rail suspension structure and the abnormal posture of the drone. This embodiment can achieve efficient and accurate automated spraying for medium and large drones while reducing the ground space occupied.

[0059] The following is a further description of the overhead rail suspension structure in the above-mentioned large UAV spraying system:

[0060] The X-axis track in the overhead rail suspension structure includes an X-axis track body; the Y-axis track includes a Y-axis track body, an X-axis slide and an X-axis drive assembly, and the Y-axis track body is installed on the X-axis track body through the X-axis slide; the X-axis slide is used to connect and support the Y-axis track body; the X-axis drive assembly is used to drive the Y-axis track body to move along the X-axis through the X-axis slide.

[0061] The Z-axis track includes a Z-axis track body, a Y-axis slide, a Y-axis drive assembly, a Z-axis slide and a Z-axis drive assembly. The Z-axis track body is suspended and installed on the Y-axis track body through the Y-axis slide. The Y-axis slide is used to connect and support the Z-axis track body. The Y-axis drive assembly is used to drive the Z-axis track body to move along the Y-axis through the Y-axis slide. The spraying robot is installed on the Z-axis track body through the Z-axis slide. The Z-axis slide is used to connect and support the spraying robot. The Z-axis drive assembly is used to drive the spraying robot to move along the Z-axis through the Z-axis slide.

[0062] See also Figure 2 , Figure 2A schematic diagram of the structure of a Z-axis track of an overhead rail type suspension structure provided in an embodiment of the present application, 201 represents a Y-axis skateboard, 202 represents a Y-axis drive assembly, 203 represents a drag chain assembly, 204 represents a Z-axis linear rail, 205 represents a painting robot, and 206 represents a Z-axis drive assembly.

[0063] In the above embodiment, the X-axis drive assembly can control the spraying robot to adjust its position along the X-axis direction, the Y-axis drive assembly can control the spraying robot to adjust its position along the Y-axis direction, and the Z-axis drive assembly can control the spraying robot to adjust its position along the Z-axis direction. Through the coordinated work of the above-mentioned X-axis drive assembly, Y-axis drive assembly and Z-axis drive assembly, the overhead rail suspension structure can control the spraying robot to move in the lateral, longitudinal and vertical directions in three-dimensional space, ensuring that the surface of medium and large drones can be fully covered with spray. The above-mentioned drive assembly may include a motor, a reducer, and a gear.

[0064] As a further introduction to the above-mentioned medium and large drone spraying system, a paint supply device is also provided on the above-mentioned Z-axis track, and the paint supply device is connected to the spraying robot. Specifically, the paint supply device can be connected to the feed port of the spraying robot through a pipeline.

[0065] The following is a further description of the sky rail deformation monitoring device in the above-mentioned large UAV spraying system:

[0066] During the long-term operation of the spray robot, the track of the overhead suspension structure may be deformed due to external force or equipment weight, which will directly affect the positioning accuracy of the spray robot. Therefore, timely monitoring and early warning of the deformation of the overhead suspension structure is crucial.

[0067] In order to ensure that the spraying robot always maintains high precision during long-term operation and prevent the deformation of the track or accidental collision from causing a decrease in spraying accuracy and damage to the equipment, this solution can set up a variety of sensors (such as laser rangefinders, strain sensors, vibration sensors) in the overhead rail deformation monitoring device to determine the deformation of the overhead rail suspension structure based on multimodal monitoring.

[0068] If the sensor in the overhead rail deformation monitoring device includes the laser rangefinder, the laser rangefinder is set at the midpoint of the track of the overhead rail suspension structure, and / or the track trisection point, and / or the track quarter point. Specifically, the laser rangefinder can record the slight deviation of the track relative to its designed standard position (i.e., the reference point) in real time. Once the displacement of the track exceeds the preset threshold, this solution can immediately issue an early warning and guide maintenance through the scheduling system to avoid affecting the operation accuracy.

[0069] If the sensor in the overhead rail deformation monitoring device includes a strain sensor, the strain sensor is arranged at the load-bearing part and / or the connection part of the overhead rail suspension structure to monitor the tension, compression and torsional deformation of the rail in real time. The strain sensor is mainly used to detect the structural changes of the rail when it is subjected to the weight of the equipment or external force, so as to ensure that the rail remains stable under the action of stress. If the strain sensor detects abnormal strain, this solution can judge the risk level according to the strain size and issue an early warning to prevent further structural damage.

[0070] If the sensor in the overhead rail deformation monitoring device includes a vibration sensor, the vibration sensor is arranged at the supporting part and / or the connecting part of the overhead rail suspension structure. The vibration sensor is used to monitor the vibration generated by each axis track during the operation of the spray robot, especially the deformation caused by the instantaneous impact when the spray robot moves at high speed. The vibration sensor can monitor high-frequency vibration signals and, combined with the data of other sensors, analyze whether the track is subjected to excessive dynamic stress under high load conditions, thereby ensuring the safety of the track structure.

[0071] See also Figure 3 , Figure 3 This is a schematic diagram of the layout of a sensor on an overhead rail suspension structure provided in an embodiment of the present application, in which 101 represents the X-axis track, 102 represents the Y-axis track, 103 represents the Z-axis track, 301 represents the laser rangefinder, 302 represents the strain sensor, and 303 represents the vibration sensor.

[0072] The data processing module in the above-mentioned overhead rail deformation monitoring device can receive real-time data collected by the laser rangefinder, strain sensor and vibration sensor through wireless transmission, and can also combine the deep learning algorithm to comprehensively analyze and process the collected data, so as to accurately monitor the deformation of the overhead rail suspension structure, thereby ensuring the continuity and stability of the spraying operation.

[0073] Specifically, the data processing module in the above-mentioned overhead rail deformation monitoring device includes a preprocessing unit and a deformation analysis unit.

[0074] In order to ensure the stability and accuracy of the raw data collected by the sensor, the preprocessing unit can preprocess the raw data collected by the sensor. The preprocessing unit is used to perform Kalman filtering on the data collected by the sensor to obtain filtered data. The above process preprocesses the raw data collected by the sensor by applying the Kalman Filter algorithm to reduce the noise measured by the sensor. As a recursive optimization algorithm, Kalman filtering can dynamically estimate the current measurement value based on the historical data of the sensor, thereby extracting high-precision displacement, strain and vibration information, and can provide reliable data support even in the case of large environmental noise.

[0075] The deformation analysis unit is used to process the filtered data using a long short-term memory artificial neural network (LSTM) to obtain deformation information; wherein the deformation information includes the current deformation and deformation trend of the overhead rail suspension structure. The above-mentioned current deformation includes: the deformation of a single track in the overhead rail suspension structure and the deformation of the entire overhead rail suspension structure at the current moment. The above-mentioned deformation trend includes: the deformation of a single track in the overhead rail suspension structure and the deformation of the entire overhead rail suspension structure at the future moment. The long short-term memory artificial neural network is a recurrent neural network with a long-term memory function, which can effectively capture the temporal dependency in the track deformation data. Through the long short-term memory network model, the deformation analysis unit can analyze the change trend of the sensor historical data, obtain the current track deformation, and accurately predict the future track deformation. The forget gate, input gate and output gate mechanism of the long short-term memory network model help the model understand the track deformation dynamics over a long time span, thereby achieving more accurate long-term prediction.

[0076] The data processing module may further include: an auxiliary adjustment unit, configured to determine a deformation cause according to the deformation information, and output a deformation adjustment strategy corresponding to the deformation cause.

[0077] As a further introduction to the above-mentioned medium and large UAV spraying system, the overhead rail deformation monitoring device also includes an early warning module, which is used to control the overhead rail suspension structure to shut down and output early warning information when the deformation of the overhead rail suspension structure is greater than a threshold value.

[0078] On the basis of obtaining the current deformation amount and deformation trend of the overhead rail suspension structure, the above-mentioned warning module can determine whether the current deformation amount and / or deformation trend is greater than the deformation threshold; if so, the overhead rail suspension structure is controlled to shut down and output warning information; if not, the data collected by the sensor can be obtained again to use the long short-term memory artificial neural network to predict the new current deformation amount and deformation trend. It can be seen that in this scheme, if the current deformation amount is greater than the deformation threshold, the system will shut down and issue a warning; if the deformation trend is greater than the deformation threshold, the system will shut down and issue a warning; if the current deformation amount and deformation trend are present, the system will shut down and issue a warning; specifically, the system can shut down and issue a warning when at least one of the following situations occurs:

[0079] Case 1: At the current moment, the deformation of a certain track is greater than threshold 1;

[0080] Case 2: At the current moment, the overall deformation of the overhead suspension structure is greater than threshold 2;

[0081] Case 3: It is predicted that the deformation of a certain track at a future time is greater than threshold 3;

[0082] Case 4: It is predicted that the overall deformation of the overhead suspension structure will be greater than threshold 4 at a future moment.

[0083] The above data processing module integrates Kalman filtering and long short-term memory network models to process sensor data and predict trends. Kalman filtering removes noise, and long short-term memory network models are used to predict the deformation trend of the track, achieving more accurate long-term predictions and early warnings, effectively reducing the risk of equipment failure and maintenance costs, and improving system reliability and spraying accuracy.

[0084] This solution can install laser rangefinders, strain sensors and vibration sensors at key nodes of the X-axis track, Y-axis track and Z-axis track to monitor the displacement, strain and vibration deformation of each track in real time. After obtaining the data collected by the above sensors, this solution can use data fusion analysis and intelligent algorithm processing to comprehensively monitor the deformation of the track and achieve comprehensive monitoring and early warning of track deformation. The above method can ensure the stability of the overhead rail suspension structure, provide a safer and more efficient operating environment for spraying operations, and improve production efficiency and spraying quality. The above monitoring and early warning mechanism not only improves the operating accuracy of the spraying robot, enhances the safety and stability of the spraying operation, but also greatly improves the overall reliability of the system and reduces the risk of equipment failure and maintenance costs.

[0085] See also Figure 4 , Figure 4 This is a flow chart of a deformation monitoring and early warning method for an overhead rail suspension structure provided in an embodiment of the present application. This solution adopts multimodal monitoring technology to monitor the displacement, strain and vibration deformation of each track in the overhead rail suspension structure in real time. Through data fusion and intelligent algorithms, the deformation of the track is fully monitored and early warning is achieved. Specifically, the following steps are included:

[0086] S401: Collect data using a laser rangefinder, a strain sensor, and a vibration sensor.

[0087] Specifically, this solution uses laser rangefinders, strain sensors and vibration sensors to collect real-time displacement, strain and vibration data of the sky rail on the X-axis track, Y-axis track and Z-axis track, so as to use these data for sky rail deformation monitoring and early warning.

[0088] S402: Process the received sensor data using a Kalman filter algorithm to eliminate noise and improve data accuracy.

[0089] S403: Based on the long short-term memory artificial neural network and historical data, intelligent data analysis is performed to obtain the current deformation amount and deformation trend of the track.

[0090] Specifically, this embodiment can combine historical data and use a long short-term memory artificial neural network to perform intelligent analysis on the filtered sensor data to evaluate the current deformation of each axis of the overhead rail suspension structure and predict future deformation trends, and output the specific deformation position and deformation amount.

[0091] S404: Determine whether the deformation amount (current deformation amount or deformation trend) exceeds a preset threshold.

[0092] If the deformation value does not exceed the preset threshold, the process proceeds to step S409 to perform the spraying operation normally. If the deformation value exceeds the preset threshold, the process proceeds to step S405.

[0093] S405: Control the spraying robot to stop and return to its original position, and issue a warning prompt message.

[0094] This step is based on the fact that the deformation of the overhead rail exceeds a preset threshold. At this time, the spray robot can be controlled to stop and instructed to return to its original position, while issuing a warning prompt message.

[0095] S406: Outputting a deformation adjustment strategy.

[0096] Specifically, this embodiment can analyze the deformation cause according to the deformation result and output a corresponding deformation adjustment strategy.

[0097] S407: Correct the track by manual adjustment.

[0098] Specifically, the operator can perform track correction according to the deformation adjustment strategy.

[0099] S408: Confirm that the track has been correctly corrected using a laser rangefinder.

[0100] S409: Control the spraying robot to spray the drone.

[0101] In the process of performing the above operations, the deformation monitoring data can be archived to facilitate subsequent analysis and tracking.

[0102] This embodiment realizes high-precision and immediate error detection and adjustment through real-time data collection of multi-source sensors, combined with deep learning and filtering technology. This embodiment combines the data processed by Kalman filtering with the long short-term memory network model to comprehensively analyze the time series changes of multimodal data such as displacement, strain and vibration of the track. Through the sensitive recognition ability of the long short-term memory network model to small changes in the track, this embodiment can issue an early warning when the deformation of the overhead rail suspension structure exceeds the safety threshold. Based on the analysis results of the deep learning model, maintenance and adjustment suggestions can be provided to operators to prevent the occurrence of potential failures.

[0103] The following is a further description of the robot posture optimization module in the above-mentioned large-scale UAV spraying system:

[0104] The above-mentioned robot posture optimization module includes an error interval determination unit, an error compensation unit, an error judgment unit and an online compensation unit.

[0105] The error interval determination unit is used to set a reference error interval according to the deformation of the overhead rail type suspension structure before starting the operation.

[0106] The error compensation unit is used to calculate the deviation matrix between the actual posture and the standard posture of the UAV when the overhead rail suspension structure is not in an operating state and the posture error of the UAV is within the reference error interval, and use the rotation parameters and translation parameters in the deviation matrix to adjust the posture of the spray robot. Specifically, the above process can calculate the deviation matrix by the method of maximum characteristic fitting to obtain the optimal rotation parameters and translation parameters. In this way, the posture of the spray robot can be accurately adjusted to ensure that it can accurately adapt to the actual position and posture of medium and large UAVs, thereby improving the operation accuracy and reliability.

[0107] The error judgment unit is used to determine that the spray robot meets the prerequisites for spraying operations if the UAV posture error is less than the lower limit of the reference error interval; the error judgment unit is also used to determine that the spray robot does not meet the prerequisites for spraying operations if the UAV posture error is greater than the upper limit of the reference error interval, and generate a warning message.

[0108] The online compensation unit is used to adjust the posture of the spray robot according to the deformation of the overhead rail suspension structure at the current moment when the overhead rail suspension structure is in operation. During the operation of the spray robot on medium and large drones, the deformation of the overhead rail suspension structure will change in real time. The online compensation unit can calculate the required correction parameters according to the influence of the current deformation on the posture of the spray robot, and then make real-time corrections to the posture of the spray robot based on the correction parameters. The above operation can dynamically compensate for the deformation of the overhead rail suspension structure during the operation process, ensuring that the spray robot always operates according to the most optimized path and posture.

[0109] The above-mentioned medium and large UAV spraying system also includes: a visual detection device arranged in the working area of ​​the robot, which is used to detect the actual posture of the UAV.

[0110] See also Figure 5 , Figure 5 A workflow diagram of a robot posture optimization module provided in an embodiment of the present application specifically includes the following steps:

[0111] S501: Determine the posture error of the UAV according to the actual posture and preset posture of the medium and large UAV;

[0112] The preset posture refers to the ideal position and posture of a medium or large UAV that is pre-set according to the design or planning. The default operation path of the spraying robot is set based on the above preset posture to ensure that the spraying robot can perform the spraying task in the expected way.

[0113] The actual posture is the actual position and posture of the spraying robot; specifically, in this embodiment, the medium-to-large UAV can be scanned by a visual detection device to obtain the actual posture of the medium-to-large UAV. The above-mentioned preset posture and actual posture are both postures in the world coordinate system.

[0114] By comparing the actual position and attitude with the preset position and attitude and calculating the difference between the two, the UAV position and attitude error can be obtained. The above UAV position and attitude error is used to describe the difference between the actual position and attitude of medium and large UAVs and their preset ideal position and attitude.

[0115] S502: Acquire an initial deformation of the overhead rail suspension structure, and set a reference error interval according to the initial deformation;

[0116] Among them, the above-mentioned initial deformation is the deformation of the overhead rail suspension structure before the operation begins. The deformation of the overhead rail suspension structure and the posture error of the drone will affect the positioning accuracy of the spray robot. This embodiment sets the reference error interval according to the initial deformation. When the posture error of the drone is within the reference error interval, the error can be eliminated by adjusting the posture and operation path of the spray robot. When the posture error of the object is greater than the upper limit of the reference error interval, an early warning is triggered for manual intervention.

[0117] As a feasible implementation, [T1, T2] may be used to represent the reference error interval, the initial deformation amount is negatively correlated with T1, and the initial deformation amount is negatively correlated with T2. T=T2-T1, in this embodiment, the initial deformation can be set The value of T is negatively correlated, and it can also be The value of T is set to a fixed value.

[0118] S503: If the drone posture error is within the reference error range, adjusting the initial working posture of the spraying robot according to the drone posture error to obtain a target working posture, and adjusting a default working path according to the target working posture to obtain a target working path;

[0119] Among them, this step is based on the fact that the object posture error is within the reference error range. Due to the existence of the object posture error, this embodiment adjusts the posture of the spraying robot when it starts working so that the spraying robot is aligned with the medium and large UAV.

[0120] The initial working posture is the posture of the spraying robot when it starts working according to the default working path. In this embodiment, the target working posture is obtained by adjusting the initial working posture according to the posture error of the drone; in this embodiment, the default working path can be replanned and adjusted according to the newly obtained target working posture to obtain the target working path. The target working posture is the posture of the spraying robot when it starts working according to the target working path.

[0121] Through the above method, this embodiment can automatically compensate for posture differences within the allowable error range, ensuring that the spraying robot always operates according to the most optimized path, thereby improving operation accuracy and efficiency.

[0122] After the above operations, the medium and large UAV spraying system can control the overhead rail suspension structure based on the target working posture and the target working path to make the spraying robot in the target working posture, and then spray the medium and large UAV according to the target working path.

[0123] This embodiment obtains the initial deformation of the overhead rail suspension structure before the spray robot starts working, and sets a reference error interval according to the initial deformation. This embodiment determines the object posture error according to the actual object posture of the medium and large unmanned aerial vehicle and the preset object posture. If the object posture error is within the reference error interval, the initial working posture of the spray robot is adjusted to obtain the target working posture, and the default working path is adjusted according to the target working posture to obtain the target working path, so as to control the overhead rail suspension structure based on the adjusted target working posture and target working path. When the initial deformation is large, the object posture error remains small to ensure the working accuracy through error compensation. When the initial deformation is small, the allowable object posture error can be appropriately increased. This embodiment sets the reference error interval according to the initial deformation of the overhead rail suspension structure, which can effectively determine whether the working posture and working path can be adjusted under different deformation conditions. The above process sets the error allowable range according to the initial deformation of the overhead rail suspension structure, and automatically adjusts the working path of the spray robot in combination with the error between the actual object posture and the preset object posture, which can compensate for the error caused by the inconsistency between the simulation environment and the actual environment. Therefore, this embodiment can automatically adjust the working path of the spraying robot and improve the working accuracy.

[0124] As a feasible implementation manner, when setting the reference error interval, corresponding error intervals may be set for the position and the attitude respectively, that is, the position error interval and the attitude error interval are set according to the initial deformation amount.

[0125] On the basis of determining the position error interval and the attitude error interval, this embodiment can determine the drone position error and the drone attitude error according to the drone posture error; compare the drone position error with the position error interval; and compare the drone attitude error with the attitude error interval.

[0126] If the position error of the drone is greater than the upper limit of the position error interval, or the attitude error of the drone is greater than the upper limit of the attitude error interval, the overhead rail suspension structure is controlled to shut down and output a warning message. Specifically, if the position error of the drone is greater than the upper limit of the position error interval, or the attitude error of the drone is greater than the upper limit of the attitude error interval, it means that the position error of the drone is large and manual intervention is required.

[0127] If the drone position error is less than the lower limit of the position error interval, and the drone attitude error is less than the lower limit of the attitude error interval, the overhead rail suspension structure is controlled based on the initial operation posture and the default operation path, so that the spray robot can operate on medium and large drones. Specifically, if the drone position error is less than the lower limit of the position error interval, and the drone attitude error is less than the lower limit of the attitude error interval, it means that the drone attitude error is small, and the attitude and path can be adjusted without adjusting the attitude and path, and the drone can be directly controlled based on the initial operation posture and the default operation path.

[0128] The above content provides the processing methods when the drone posture error is large and the drone posture error is small. If the drone posture error is within the reference error interval, the relevant operations of S503 can be performed. The following lists the scenarios where the drone posture error is within the reference error interval: (1) The drone position error is within the position error interval, and the drone attitude error is within the attitude error interval; (2) The drone position error is within the position error interval, and the drone attitude error is less than the lower limit of the attitude error interval; (3) The drone position error is less than the lower limit of the position error interval, and the drone attitude error is within the attitude error interval.

[0129] The control method of the above-mentioned overhead rail type suspension structure is described below with reference to an embodiment in practical application.

[0130] Before the spraying robot sprays the surface of the drone, it needs to align its position. Since the posture of the drone in the simulation environment is difficult to be completely consistent with the actual working environment, and there is a certain deformation error in the overhead rail suspension structure, position and rotation errors may occur during the actual positioning process. The structural characteristics of the spraying robot itself make this error particularly obvious when crossing the singularity point, which can easily lead to inconsistency between the actual motion posture of each axis and the simulated posture. In severe cases, it may even cause an accident in which the joint arm of the spraying robot collides with the drone. The above situation will not only interrupt the spraying operation, but may also cause damage to the equipment.

[0131] In view of the above-mentioned related situations, the traditional processing method is to import the actual posture data of the UAV into the path planning software to re-plan the path and simulate and confirm it, and then conduct process approval. The above process is cumbersome and lacks real-time performance. Each spraying operation needs to be repeated, which is difficult to meet the efficient demand of one-time planning and multiple reuse of the same model.

[0132] In order to improve the spraying quality, work efficiency and safety, this application proposes a robot posture optimization scheme based on multi-source error hierarchical compensation. This scheme compensates for the UAV posture error and the overhead rail deformation error, corrects the error in the actual environment, and adjusts the posture of the spraying robot to ensure that the spraying robot always maintains an accurate posture.

[0133] The above-mentioned robot posture optimization scheme with multi-source error hierarchical compensation is based on three links: multi-source data acquisition, error evaluation and intelligent adjustment. It realizes high-precision control and posture optimization of the spray robot in the actual environment, and effectively solves the error problem caused by the inconsistency between the simulation environment and the actual environment.

[0134] In the multi-source data acquisition phase, this solution uses data from multiple sources to optimize the robot's posture. The data sources used include: visual calibration data, track deformation monitoring data, and theoretical benchmark data. The actual posture of the drone can be determined based on the visual calibration data, the deformation of the overhead rail suspension structure can be determined based on the track deformation monitoring data, and the posture that the drone should theoretically achieve can be determined based on the theoretical benchmark data.

[0135] Specifically, in this embodiment, the feature points of the drone can be acquired through a visual detection system, and the pose calculation can be performed to obtain the actual pose of the drone.

[0136] Before the spraying robot starts working, the present embodiment can use a laser rangefinder, a strain sensor and a vibration sensor to respectively capture the micro displacement, structural strain and vibration deformation of the overhead rail suspension structure, thereby determining the deformation amount of the overhead rail suspension structure, i.e., the initial deformation amount. By detecting the deformation amount of the overhead rail suspension structure, the actual position and attitude information of the drone in the working area can be obtained.

[0137] In the error evaluation phase, this scheme can perform error calculation based on the acquired multi-source data, and compare the actual posture of the UAV with the theoretical posture to evaluate the posture error of the UAV.

[0138] The intelligent adjustment algorithm of this embodiment can adopt a hierarchical compensation strategy to compensate for the UAV posture error and the overhead rail deformation error, and correct the posture and operation path of the spray robot. Specifically, this embodiment can use multi-level thresholds (such as T1, T2) to classify the degree of deviation, and the above multi-level thresholds can be determined according to the initial deformation of the overhead rail suspension structure.

[0139] If the UAV posture error is less than T1, it means that the UAV posture error is within an acceptable range, and the spraying robot can directly enter the spraying operation.

[0140] If the posture error of the drone is greater than or equal to T1 and less than or equal to T2, this embodiment can start the intelligent adjustment program to correct the posture and operation path of the spray robot so that the precision spray robot can align the spray path and the target surface. In the intelligent adjustment program, in order to further optimize the error processing, this solution can use the characteristic maximum fitting method to calculate the deviation matrix (also known as the posture compensation matrix). The optimal rotation parameters and translation parameters can be obtained by the calculation of the maximum fitting method, so that the actual posture of the drone in the coordinate system of the spray robot is consistent with the theoretical reference, and accurate error compensation is achieved. Through real-time data analysis and intelligent adjustment, this solution can quickly respond to environmental changes, avoid time delays in manual adjustments, and greatly improve the spraying efficiency. This solution realizes automated error detection and optimization adjustment, so that similar models can be reused multiple times through one path planning, saving time for re-path planning and process approval, reducing the need for manual operation, and significantly improving the efficiency of spraying operations.

[0141] If the drone’s posture error is greater than T2, an early warning will be issued and manual intervention will be prompted so that the spraying operation can be readjusted. Through the above early warning mechanism, this solution can avoid the risks caused by singular point collisions and excessive errors, and significantly improve the safety of the operation.

[0142] The thresholds T1 and T2 can be determined according to the deformation of the overhead rail suspension structure. The specific calculation formula is as follows:

[0143] ;

[0144] ;

[0145] Indicates the deformation of the overhead suspension structure, which may include displacement, strain or a combination thereof. k1, k2, b1 and b2 are empirical parameters determined by experiments or simulations, and T1 is less than T2.

[0146] The above-mentioned hierarchical compensation strategy of multi-source data provides reliable data support for a comprehensive understanding of the current working environment, enabling the spraying robot to achieve high-precision operation in a changing environment.

[0147] In the intelligent adjustment phase, this solution can implement a hierarchical adjustment strategy for different error conditions to ensure that the robot can operate accurately in complex environments. After the intelligent adjustment is completed, this solution can generate a corrected positioning coordinate system and import it into the spraying path planning system to generate a spraying path (i.e., spraying trajectory) that adapts to the current environment. In addition, real-time track correction can be performed during the spraying process, and the robot position can be dynamically compensated through the deformation monitoring data of the overhead rail to ensure the consistency of the spraying accuracy. This real-time track correction mechanism ensures the robot's continued high-precision operation in the case of track deformation.

[0148] With the support of multi-source data, this solution can intelligently adjust the spray robot posture and realize real-time optimization of the spray path. This solution can use the corrected spray robot posture as a positioning reference to generate a new spray path. The above process ensures that the spray robot maintains operating accuracy and stability in a dynamic environment, thereby improving the spray quality. The above intelligent adjustment mechanism not only realizes automated error detection and adjustment, but also reduces dependence on manual intervention, significantly improving the efficiency of spray operations.

[0149] Before the spraying robot starts working, if the drone posture error is within the range of T1 to T2, this solution optimizes the robot's posture in space through an intelligent adjustment mechanism. After the spraying robot starts working, this solution can detect the current deformation of the overhead rail suspension structure in real time and adjust the spraying robot's posture according to the current deformation to avoid collisions caused by singular points and ensure smooth operation.

[0150] This solution can automatically correct the spraying path and robot posture according to real-time data, so that similar drones can be reused multiple times after one path planning, saving time for re-path planning and process approval, and reducing the need for manual intervention. This solution not only ensures the continuity and stability of the spraying operation, but also improves work efficiency, making the spraying system more intelligent and convenient.

[0151] See also Figure 6 , Figure 6 A flowchart of a robot posture optimization method for multi-source error hierarchical compensation provided in an embodiment of the present application specifically includes the following steps:

[0152] S601: The operator adjusts the drone to the designated marking position.

[0153] Specifically, the operator can adjust the drone to the identification position specified by the spraying factory to complete the aircraft entering the factory. The above identification position can be the initial reference point.

[0154] S602: Determine the actual position and posture of the drone based on machine vision.

[0155] Specifically, this embodiment can use a machine vision system to capture the surface features of the drone, calculate the actual coordinate system, and use the posture of the actual coordinate system as the actual posture of the drone to achieve visual positioning and feature calculation. This embodiment can also obtain the theoretical coordinate system of the drone from theoretical reference data, and use the posture of the theoretical coordinate system as the posture that the drone should theoretically maintain, that is, the theoretical posture.

[0156] S603: Compare the actual posture of the UAV with the theoretical posture to obtain a posture error.

[0157] Among them, this step can obtain the deviation information between the actual state of the drone and the theoretical benchmark, providing data support for subsequent error comparison. As a feasible implementation method, this embodiment can compare the actual coordinate system of the drone with the theoretical coordinate system, calculate the deviation value between the two, and use the deviation value as the posture error.

[0158] S604: Perform threshold determination on the posture error.

[0159] This scheme can compare the pose error with the reference error interval, and the following situations may exist:

[0160] If the posture error is less than the lower limit value T1 of the reference error interval, it means that the error is within an acceptable range, and the robot can enter step S608 to directly enter the spraying operation state;

[0161] If the posture error is within the reference error range, the process proceeds to S605 to adjust the posture and output a new spraying path.

[0162] If the posture error is greater than the upper limit value T2 of the reference error interval, it means that the error is too large. At this time, manual intervention can be prompted and the operator needs to readjust the position of the drone, that is, re-execute step S601.

[0163] S605: Calculate the deviation matrix by using the characteristic maximum fitting method.

[0164] Among them, this step is based on the situation that the posture error is between the upper limit value T1 and the lower limit value T2. At this time, it means that the error is judged to be within the allowable range, and the error compensation operations of S605~S607 can be performed.

[0165] In this step, the deviation matrix is ​​calculated by the characteristic maximum fitting method to obtain the optimal rotation parameters and translation parameters, so as to achieve the maximum fit between the ideal posture and the actual posture and optimize the position and posture.

[0166] S606: Correct the position and posture of the painting robot using the deviation matrix.

[0167] Specifically, this step adjusts the position and posture of the spraying robot according to the deviation matrix so that the spraying robot is more accurately aligned with the surface of the drone to ensure the spraying accuracy.

[0168] S607: Generate a new spraying path.

[0169] In this step, the initial spraying path can be corrected according to the corrected position and posture of the spraying robot to obtain a new spraying path. The new spraying path conforms to the actual position of the drone, enabling the spraying robot to move accurately along the updated path to achieve uniform coverage.

[0170] This solution avoids robot collisions caused by track deformation or singular points through early warning and dynamic posture adjustment, significantly improving the safety of the system. The corrected spraying path generated in real time by this solution ensures uniform spraying and ensures more stable and high-quality spraying quality on the drone surface.

[0171] S608: Control the spraying robot to perform operations, and correct the position and posture of the spraying robot based on real-time track deformation monitoring data.

[0172] After all necessary adjustments, the spraying robot can enter the formal operation state and perform the spraying task on the surface of the drone. This solution can dynamically adjust the robot position and posture during the spraying process based on the track deformation monitoring data to compensate for the track error and ensure the accuracy of continuous spraying.

[0173] It can be seen that this solution obtains the UAV posture error in real time through the visual detection system, and uses the sky rail monitoring system to collect the track deformation error, realizing multi-level error data acquisition and hierarchical compensation. The above-mentioned robot posture optimization solution with multi-source error hierarchical compensation can solve the posture deviation problem caused by the inconsistency between the simulation environment and the actual environment, and avoid collisions caused by singular points or track deformation.

[0174] Specifically, this solution evaluates the posture error and sets multiple thresholds (such as T1 and T2). When the posture error is within the allowable range (between T1 and T2), this embodiment can automatically optimize the position and posture of the robot through the intelligent adjustment function; when the error exceeds the threshold T2, this embodiment can prompt manual intervention to ensure that the error is controlled within a safe range. The above strategy not only ensures the accuracy of the spraying operation, but also significantly reduces accidental collisions caused by excessive errors, and improves the safety and stability of the spraying robot operation. The multi-source data fusion and intelligent adjustment algorithm provided by this solution effectively solve the posture deviation problem in the spraying operation, and provides strong technical support for drone surface spraying.

[0175] The various embodiments in the specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same and similar parts between the various embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of this application, several improvements and modifications can be made to this application, and these improvements and modifications also fall within the scope of protection of this application.

[0176] It should also be noted that, in this specification, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprises", "comprising" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the statement "comprising a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element.

Claims

1. A medium to large UAV spraying system, characterized in that: include: Overhead rail suspension structure, overhead rail deformation monitoring device and robot posture optimization module; The overhead rail suspension structure includes an X-axis track, a Y-axis track, a Z-axis track and a spraying robot, wherein the X-axis track is fixedly installed on the top of the robot working area, the Y-axis track is installed on the X-axis track, the Z-axis track is suspended and installed on the Y-axis track, and the spraying robot is installed on the Z-axis track; the Y-axis track is used to slide along the X-axis track, the Z-axis track is used to slide along the Y-axis track, the spraying robot is used to slide along the Z-axis track, and the spraying robot is also used to spray the drone; The overhead rail deformation monitoring device includes a sensor and a data processing module; the sensor is arranged on any one of the X-axis track, the Y-axis track and the Z-axis track or a combination of any several of the tracks; the data processing module is used to process the data collected by the sensor to obtain the deformation of the overhead rail suspension structure; wherein the sensor includes any one of a laser rangefinder, a strain sensor and a vibration sensor or a combination of any several of the above; The robot posture optimization module is used to adjust the posture of the spraying robot according to the deformation of the overhead rail suspension structure and / or the posture error of the drone; wherein the posture error of the drone is determined according to the actual posture and standard posture of the drone; Wherein, the robot posture optimization module includes: The error interval determination unit is used to set a reference error interval according to the deformation of the overhead rail suspension structure before the operation starts; wherein the calculation formula of the lower limit value T1 of the reference error interval is: ; The calculation formula of the upper limit value T2 of the reference error interval is: ; represents the deformation of the overhead rail type suspension structure, the deformation of the overhead rail type suspension structure includes displacement and / or strain; , , and It is an empirical parameter determined by experiments or simulations, and T1 is less than T2; An error compensation unit is used to calculate a deviation matrix between the actual posture of the UAV and the standard posture when the overhead rail suspension structure is not in an operating state and the posture error of the UAV is within the reference error interval, and adjust the posture of the spraying robot using the rotation parameter and the translation parameter in the deviation matrix; The error judgment unit is used to determine that the spray robot meets the prerequisites for spraying operations if the UAV posture error is less than the lower limit of the reference error interval; the error judgment unit is also used to determine that the spray robot does not meet the prerequisites for spraying operations if the UAV posture error is greater than the upper limit of the reference error interval, and generate a warning message.

2. According to the medium and large UAV spraying system of claim 1, it is characterized in that: The Y-axis track comprises a Y-axis track body, an X-axis slide and an X-axis driving assembly, wherein the Y-axis track body is mounted on the X-axis track through the X-axis slide, and the X-axis driving assembly is used to drive the Y-axis track body to move along the X-axis through the X-axis slide; The Z-axis track includes a Z-axis track body, a Y-axis skateboard, a Y-axis drive assembly, a Z-axis skateboard and a Z-axis drive assembly. The Z-axis track body is suspended on the Y-axis track body through the Y-axis skateboard, and the Y-axis drive assembly is used to drive the Z-axis track body to move along the Y-axis through the Y-axis skateboard; the spraying robot is installed on the Z-axis track body through the Z-axis skateboard, and the Z-axis drive assembly is used to drive the spraying robot to move along the Z-axis through the Z-axis skateboard.

3. The medium and large UAV spraying system according to claim 1 is characterized in that: A paint supply device is arranged on the Z-axis track, and the paint supply device is connected to the spraying robot.

4. The medium and large UAV spraying system according to claim 1 is characterized in that: If the sensor includes the laser rangefinder, the laser rangefinder is arranged at the midpoint of the track of the overhead rail suspension structure, and / or at a point where the track is divided into three equal parts, and / or at a point where the track is divided into four equal parts; If the sensor comprises a strain sensor, the strain sensor is arranged at the load-bearing part and / or the connecting part of the overhead rail type suspension structure; If the sensor comprises a vibration sensor, the vibration sensor is arranged at a supporting portion and / or a connecting portion of the overhead rail type suspension structure.

5. The medium and large UAV spraying system according to claim 1 is characterized in that: The overhead rail deformation monitoring device also includes an early warning module, which is used to control the overhead rail suspension structure to shut down and output early warning information when the deformation of the overhead rail suspension structure is greater than a threshold value.

6. The medium and large UAV spraying system according to claim 1 is characterized in that: The data processing module comprises: A preprocessing unit, used for performing Kalman filtering on the data collected by the sensor to obtain filtered data; The deformation analysis unit is used to process the filtered data using a long short-term memory artificial neural network to obtain deformation information; wherein the deformation information includes the current deformation amount and deformation trend of the overhead rail suspension structure.

7. The medium and large UAV spraying system according to claim 6 is characterized in that: The data processing module comprises: The auxiliary adjustment unit is used to determine the deformation cause according to the deformation information and output a deformation adjustment strategy corresponding to the deformation cause.

8. The medium-to-large UAV spraying system according to claim 1, characterized in that: Also includes: A visual detection device is arranged in the working area of ​​the robot, and is used to detect the actual position and posture of the drone.

9. The medium-to-large UAV spraying system according to claim 1, characterized in that: The robot posture optimization module comprises: The online compensation unit is used to adjust the posture of the spraying robot according to the deformation of the overhead rail type suspension structure at a current moment when the overhead rail type suspension structure is in an operating state.

Images