Spraying track control method, system and equipment based on deep learning and medium

Abstract

The invention relates to the technical field of spraying, in particular to a spraying track control method, system and equipment based on deep learning and a medium, according to a spraying track of a to-be-sprayed product and based on the distance between front and back products of the sprayed product, a trial spraying moving track and a spraying stopping moving track are constructed, and in a first interval, the spraying stopping moving track and the spraying stopping moving track are separated from each other; the robot is controlled to control the spray gun to conduct pre-spraying test according to the distance between the spraying test point and the to-be-sprayed product, it is guaranteed that after the spray gun enters the spraying track, uniform spraying can be achieved, and after the spray gun completes running of the spraying track, spraying is conducted within the second interval; and the control robot controls the spray gun to carry out spraying stopping operation according to the distance between the spraying stopping point and the to-be-sprayed product, it is guaranteed that spraying operation of the to-be-sprayed product is completed, meanwhile, a large amount of paint is prevented from being wasted in the interval area, automatic spraying of the to-be-sprayed product is completed in next spraying based on construction of the deep learning model, and the spraying efficiency is improved. And flow line production and intelligent prediction control work are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of spraying, and more specifically, it relates to a spraying trajectory control method, system, device and medium based on deep learning. Background Art

[0002] A spraying robot, also known as a painting robot, is an industrial robot used to replace manual operations and can perform automatic painting or spraying of other coatings. It was invented by the Norwegian company Trallfa in 1969. A painting robot mainly consists of a robot body, a computer and a corresponding control system. A hydraulically driven painting robot also includes the trajectory movement of a hydraulic oil source. Its wrist generally has 2-3 degrees of freedom and can move flexibly. The wrist of a more advanced painting robot uses a flexible wrist, which can bend in all directions and rotate. Its movement is similar to that of a human wrist and can easily reach into the workpiece through a small hole to spray the inner surface. Painting robots generally use hydraulic drive, have the characteristics of fast action speed and good explosion-proof performance, and can be taught by hand guiding or point number indication. Painting robots are widely used in the production departments of processes such as automobiles, instruments, electrical appliances, and enamel.

[0003] The Chinese invention patent with the patent number 202011502029.1 and the patent name of a sheet metal part spraying system. The present invention discloses that it includes a spraying room, a conveying device, a spraying device, a cleaning device, a drying device and a transfer mechanism. A cleaning area, a preheating area, a front spraying area, a first drying area, a back spraying area and a second drying area are sequentially arranged in the spraying room. The conveying device includes a first conveyor belt, a second conveyor belt and a third conveyor belt arranged in sequence. The drying device includes a number of drying boxes and a number of negative pressure boxes. The transfer mechanism includes a horizontal slide rail, and three horizontal moving frames are sequentially installed on the horizontal slide rail. The spraying device includes a number of paint spray guns. The cleaning device includes high-pressure air nozzles. The sheet metal part spraying system of this invention is suitable for the front and back spraying operations of non-porous sheet metal parts that cannot be hung by hanging parts, and can effectively protect the operators of manual operations and avoid the spraying operation from affecting the workshop production environment.

[0004] For the above-mentioned sheet metal part spraying system, when the paint spray gun is turned on or off, the outermost layer of paint in the paint spray gun is still in a flowing state under the influence of gravity. At this time, when the robot controls the paint spray gun to spray paint for spraying, an air spraying situation will occur, which will lead to uneven spraying. Based on the flow production of modern factories, in order to avoid the situation of uneven spraying, currently, continuous spraying is selected between adjacent products to be sprayed, so as to avoid the uneven spraying caused by air spraying. However, since the products to be sprayed require time during the loading process, the adjacent products to be sprayed are placed at intervals. If the traditional continuous spraying method is still used, a large amount of paint will be wasted in the space between adjacent products to be sprayed. Summary of the Invention

[0005] In view of the deficiencies of the above-mentioned existing technologies, the purpose of the present invention is to provide a spray trajectory control method based on deep learning, which has the advantages of ensuring uniform spraying of the spray gun, avoiding a large amount of paint waste in the interval area, realizing pipeline production, and intelligent predictive control of operations.

[0006] The above technical object of the present invention is achieved through the following technical solutions: A spray trajectory control method based on deep learning, comprising: Obtaining that the distances between the front and rear products to be sprayed are the first spacing and the second spacing respectively, and the spray trajectory of the product to be sprayed; Both the first spacing and the second spacing are provided with a trial spraying operation trajectory and a stop spraying operation trajectory, and each product to be sprayed experiences at least one spraying trajectory, one trial spraying operation trajectory, and one stop spraying operation trajectory; Controlling the robot to move the spray gun to the first spacing or the second spacing, selecting a trial spraying point in the trial spraying operation trajectory according to the spacing distance of the first spacing or the second spacing, and selecting a stop spraying point in the stop spraying operation trajectory according to the spacing distance of the first spacing or the second spacing; Constructing a deep learning model in advance according to the data sets of the first spacing, the second spacing, the spraying trajectory, the trial spraying operation trajectory, and the stop spraying operation trajectory obtained previously; in the next spraying based on the deep learning model, pre-testing the trial spraying operation trajectory and the stop spraying operation trajectory in advance through the spacing of the first spacing and the second spacing, and completing the automatic spraying of the product to be sprayed through the connection with the spraying trajectory.

[0007] Preferably, in the spraying trajectory, the robot is controlled to spray evenly according to the product to be sprayed by controlling the spray gun; based on the trial spraying operation trajectory, the robot is controlled to control the spray gun to perform a pre-trial spray according to the distance between the trial spraying point and the product to be sprayed; based on the stop spraying operation trajectory, the robot is controlled to control the spray gun to perform a stop spraying operation according to the distance between the stop spraying point and the product to be sprayed.

[0008] Preferably, the edge points of the product to be sprayed are used as the starting point and the ending point of the spraying trajectory, the trial spraying operation trajectory is the path between the starting point and the trial spraying point, and the stop spraying operation trajectory is the path between the ending point and the stop spraying point.

[0009] Preferably, if the path between the starting point and the trial spraying point of the trial spraying operation trajectory is greater than the first spacing or the second spacing, several trial spraying turning points are arranged on the path between the starting point and the trial spraying point of the trial spraying operation trajectory, and the robot is controlled to control the spray gun to turn the moving direction of the spray gun according to the trial spraying turning points and passing through the starting point twice to complete the trial spraying operation trajectory.

[0010] Preferably, if the path of the stopped spraying trajectory between the end point and the stopping spraying point is greater than the first spacing or the second spacing, a plurality of stopping spraying turning points are provided between the starting point of the previous product to be sprayed and the stopping spraying point of the stopped spraying trajectory. The robot is controlled to turn the moving direction of the spray gun according to the stopping spraying turning points to complete the stopped spraying trajectory.

[0011] A spraying system, comprising: An acquisition module, which acquires the size of a single product to be sprayed and the distances between a plurality of products to be sprayed before and after; A spraying module, which controls a robot to control a spray gun to sequentially complete the spraying operations of a trial spraying operation trajectory, a spraying trajectory, and a stopped spraying trajectory for a single product to be sprayed; A control module, based on the distances between the plurality of products to be sprayed before and after, predicts the spraying trajectory, the trial spraying operation trajectory, and the stopped spraying trajectory through a deep learning model, and quickly matches the spraying trajectory, the trial spraying operation trajectory, and the stopped spraying trajectory corresponding to the same distance based on the continuous accumulation of the operation data of the products to be sprayed that have completed spraying.

[0012] An electronic device, the electronic device includes a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, the above-mentioned spraying trajectory control method based on deep learning is implemented.

[0013] A computer-readable storage medium, the storage medium stores a computer program, characterized in that when the computer program is executed by a processor, the above-mentioned spraying trajectory control method based on deep learning is implemented.

[0014] In summary, the beneficial effects of the present invention are as follows: 1. The spraying trajectory of the product to be sprayed is obtained according to the size of the product to be sprayed, and based on the recognition of the distances between the products before and after the sprayed product, that is, the first spacing and the second spacing, a trial spraying operation trajectory and a stopped spraying operation trajectory are constructed. Within the first spacing, the robot is controlled to control the spray gun to perform a trial spray in advance according to the distance between the trial spraying point and the product to be sprayed, so as to ensure that the spray gun can perform uniform spraying after entering the spraying trajectory. After the spray gun completes the operation of the spraying trajectory, within the second spacing, the robot is controlled to control the spray gun to perform a stopping spray operation according to the distance between the stopping spraying point and the product to be sprayed, while ensuring the completion of the spraying operation of the product to be sprayed, avoiding a large amount of paint waste in the interval area; 2. Construct a deep learning model in advance according to the datasets of the first spacing, the second spacing, the spraying trajectory, the trial spraying operation trajectory, and the stopping operation spraying trajectory obtained multiple times previously; in the next spraying based on the deep learning model, pre-test the trial spraying operation trajectory and the stopping operation spraying trajectory in advance through the spacings of the first spacing and the second spacing, and complete the automatic spraying of the product to be sprayed through the connection with the spraying trajectory, so as to achieve pipeline production and intelligent predictive control operations. Brief Description of the Drawings

[0015] Figure 1 is a schematic diagram of the process step framework of an embodiment of the present invention; Figure 2 is a schematic diagram of the trajectory prediction of the front and rear product intervals of the product to be sprayed in an embodiment of the present invention; Figure 3 is another schematic diagram of the trajectory prediction of the front and rear product intervals of the product to be sprayed in an embodiment of the present invention; Figure 4 is a schematic diagram of the connection of system modules in an embodiment of the present invention.

[0016] Reference Numerals: 1. Acquisition Module; 2. Spraying Module; 3. Control Module. Detailed Embodiments

[0017] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to 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 used to limit the present invention.

[0018] It should be noted that although the functional modules are divided in the device schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order from the module division in the device or the flowchart. Terms such as "first" and "second" in the specification, claims and the above-mentioned drawings are used to distinguish similar objects and do not necessarily need to describe a specific order or sequence.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs. The terms used herein are only for the purpose of describing the embodiments of the present invention and are not intended to limit the present invention.

[0020] The spraying trajectory control method, system, electronic device and storage medium based on deep learning provided by the embodiments of the present invention will be specifically described through the following embodiments. First, the paint spraying trajectory control method based on deep learning in the embodiments of the present invention will be described.

[0021] The paint spraying trajectory control method based on deep learning provided by the embodiments of the present invention relates to, but is not limited to, the technical field of sheet metal paint spraying. The paint spraying trajectory control method based on deep learning provided by the embodiments of the present invention can be applied to a terminal, can also be applied to a server side, or can also be software running on a terminal or a server side. In some embodiments, the terminal can be a smart phone, a tablet computer, a notebook computer, a desktop computer, a robot, etc.; the server side can be configured as an independent physical server, can also be configured as a server cluster or a distributed system composed of multiple physical servers, or can also be configured as a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application for implementing the method of paint spraying, etc., but is not limited to the above forms.

[0022] The present invention can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multi-processor systems, microprocessor-based systems, set-top boxes, programmable consumer electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and so on. The present invention can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present invention can also be practiced in a distributed computing environment where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0023] A paint spraying trajectory control method based on deep learning, see Figure 1 , including: Obtain that the distances between the product before and after the product to be sprayed are the first spacing and the second spacing respectively, and the spraying trajectory of the product to be sprayed; Both the first spacing and the second spacing are provided with a trial spraying operation trajectory and a stop spraying operation trajectory, and each product to be sprayed experiences at least one spraying trajectory, one trial spraying operation trajectory, and one stop spraying operation trajectory; Control the robot to move the spray gun to the first spacing or the second spacing, select a trial spraying point in the trial spraying operation trajectory according to the spacing distance of the first spacing or the second spacing, and select a stop spraying point in the stop spraying operation trajectory according to the spacing distance of the first spacing or the second spacing; Construct a deep learning model in advance according to the data sets of the first spacing, the second spacing, the spraying trajectory, the trial spraying operation trajectory, and the stopping operation spraying trajectory obtained multiple times previously; in the next spraying based on the deep learning model, predict the trial spraying operation trajectory and the stopping operation spraying trajectory in advance through the spacing between the first spacing and the second spacing, and complete the automatic spraying of the product to be sprayed through the connection with the spraying trajectory.

[0024] In this embodiment, the spraying trajectory of the product to be sprayed is obtained according to the size of the product to be sprayed, and based on the recognition of the distances between the products before and after the sprayed product, that is, the first spacing and the second spacing, the trial spraying operation trajectory and the stopping spraying operation trajectory are constructed. Within the first spacing, the robot is controlled to control the spray gun to perform a trial spray in advance according to the distance between the trial spray point and the product to be sprayed, so as to ensure uniform spraying after the spray gun enters the spraying trajectory. After the spray gun completes the operation of the spraying trajectory, within the second spacing, the robot is controlled to control the spray gun to perform a stop spraying operation according to the distance between the stop spraying point and the product to be sprayed. While ensuring the completion of the spraying operation of the product to be sprayed, a large amount of paint is avoided from being wasted in the interval area. Construct a deep learning model in advance according to the data sets of the first spacing, the second spacing, the spraying trajectory, the trial spraying operation trajectory, and the stopping operation spraying trajectory obtained multiple times previously; in the next spraying based on the deep learning model, predict the trial spraying operation trajectory and the stopping operation spraying trajectory in advance through the spacing between the first spacing and the second spacing, and complete the automatic spraying of the product to be sprayed through the connection with the spraying trajectory, so as to realize pipeline production and intelligent predictive control operation.

[0025] Among them, in the spraying trajectory, the robot is controlled to control the spray gun to perform uniform spraying according to the product to be sprayed; based on the trial spraying operation trajectory, the robot is controlled to control the spray gun to perform a trial spray in advance according to the distance between the trial spray point and the product to be sprayed; based on the stopping operation spraying trajectory, the robot is controlled to control the spray gun to perform a stop spraying operation according to the distance between the stop spraying point and the product to be sprayed.

[0026] In addition, the edge points of the product to be sprayed are used as the starting point and the ending point of the spraying trajectory. The trial spraying operation trajectory is the path between the starting point and the trial spray point, and the stopping operation spraying trajectory is the path between the ending point and the stop spraying point. In this way, a complete path for the spray gun to perform the spraying operation for a single product to be sprayed is constructed.

[0027] During the spraying process, the product to be sprayed is fed onto the spraying point through a conveyor belt. This feeding method is based on a selection method for screening the product to be sprayed. For example, during the production and manufacturing of sheet metal, there are defective products. Therefore, after screening, the product is fed onto the spraying point through the conveyor belt for paint spraying. However, during the screening process, the good and defective products are random, which results in an uncertain distance between adjacent products to be sprayed. Moreover, due to the different sizes of sheet metal, during the centralized spraying process, classifying the sizes of different sheet metals in advance will undoubtedly further increase the complexity of the production process.

[0028] Based on this, in order to perform operation spraying on sheet metal with different sizes and after screening, during the process of realizing streamlined operation production, for the trial spraying operation trajectory, if the path between the starting point and the trial spraying point of the trial spraying operation trajectory is greater than the first distance or the second distance, several trial spraying turning points are set between the starting point and the trial spraying point of the trial spraying operation trajectory. The robot is controlled to turn the moving direction of the spray gun according to the trial spraying turning points and passing through the starting point twice to complete the trial spraying operation trajectory.

[0029] For the stopped spraying operation trajectory, if the path between the end point and the stopped spraying point of the stopped spraying operation trajectory is greater than the first distance or the second distance, several stopped spraying turning points are set between the starting point of the previous product to be sprayed and the stopped spraying point of the stopped spraying operation trajectory. The robot is controlled to turn the moving direction of the spray gun according to the stopped spraying turning points to complete the stopped spraying operation trajectory.

[0030] See Figure 2 , for example, taking the edge points of the product to be sprayed as the starting point and the end point of the spraying trajectory, the starting point is A1, the end point is A2, the distance between the product to be sprayed and the previous product to be sprayed is the first distance, and the trial spraying point is selected within the first distance, and this selected trial spraying point is B1. The distance between the product to be sprayed and the next product to be sprayed is the second distance, and the stopped spraying point is selected within the second distance, and this selected stopped spraying point is B2.

[0031] See Figure 3 , if the distance from B1 to A1 is greater than the distance of the first distance, then the trial spraying turning point C1 is set, and the robot is controlled to turn the moving direction of the spray gun according to the trial spraying turning point C1 and passing through the starting point A1 twice to complete the trial spraying operation trajectory.

[0032] If the distance from B2 to A2 is greater than the distance of the second distance, then the stopped spraying point C2 is set, and the robot is controlled to turn the moving direction of the spray gun according to the stopped spraying point C2 to complete the stopped spraying operation trajectory.

[0033] In addition, a deep learning model is constructed in advance according to the data sets of the first spacing, the second spacing, the spraying trajectory, the trial spraying operation trajectory, and the stop operation spraying trajectory obtained multiple times previously; among them, the first spacing, the second spacing, and the spraying trajectory are all obtained in real time based on a 3D camera.

[0034] Regarding the acquisition of the trial spraying points, the purpose of the trial spraying operation trajectory is to contact the spraying trajectory and avoid air spraying caused by paint loss. Therefore, the actual distance of the trial spraying operation trajectory can be adjusted according to the moving speed of the spray gun and the setting of the spray gun pressure, etc.

[0035] Regarding the acquisition of the stop spraying points, in addition to contacting the spraying trajectory, the purpose of the stop spraying operation trajectory is more importantly to avoid ineffective spraying of paint within the spacing. Therefore, the actual distance of the trial spraying operation trajectory can be obtained by adjusting according to the moving speed of the spray gun and the time when the spray gun leaves the spraying range, etc.

[0036] In addition, regarding the construction of the deep learning model, after the acquisition of the data set in this embodiment, first through: 1. Data collection and preprocessing S1, Input data: The first spacing d1, the second spacing d2.

[0037] The spraying trajectory coordinate sequence T = {(xt, yt, zt)}, based on the source of the encoder for controlling the movement of the robot.

[0038] The start and stop point markers of the trial spraying trajectory and the stop trajectory, such as timestamps or spatial coordinates.

[0039] S2, Output data: The dynamic trajectory planning results of the next spraying cycle, including the trial spraying points, the stop spraying points, and the spraying trajectory.

[0040] S3, Data preprocessing: Temporal alignment: Align the spacing data and the trajectory data according to the timestamps.

[0041] Normalization: Standardize physical quantities such as spacing and coordinates, such as scaling to the [0, 1] interval.

[0042] Enhancement strategy: Expand the data set by adding noise, simulating spacing mutations, etc.

[0043] 2. Model architecture design Regarding the spatio-temporal characteristics of spraying control, the following hybrid model structure can be selected, including: Core architecture: Transformer + LSTM Transformer Encoder: Processes the spatial relationships of spacing and trajectories. For example, the self-attention mechanism captures the mutual influence between d1 and d2.

[0044] LSTM Decoder: Generates a temporally continuous spraying trajectory, that is, uses memory cells to maintain the coherence of the spraying start and stop states.

[0045] Input and Output Design: Input: Historical temporal data, d1, d2, T at the past N time steps + current spacing d1′, d2′.

[0046] Output: Spraying trajectory T′ and trial spraying point P for the next M time steps test以及 Stop Point P stop 。

[0047] 3. Training and Validation Training Strategy: Introduce transfer learning: If the small sample data is insufficient, the pre-trained model can be used for spraying tasks in the simulation environment and then fine-tuned to the real scenario.

[0048] Validation Method: Offline Testing: Calculate the trajectory prediction accuracy and start / stop point classification accuracy through historical data.

[0049] Based on this, this embodiment also provides a spraying system, see Figure 4 ,including: Acquisition Module 1, which acquires the size of a single product to be sprayed and the distances between multiple products to be sprayed before and after; Spraying Module 2, which controls the spray gun through a robot to sequentially complete the spraying operations of the trial spraying operation trajectory, spraying trajectory, and stop spraying operation trajectory for a single product to be sprayed; Control Module 3, based on the distances between multiple products to be sprayed before and after, predicts the spraying trajectory, trial spraying operation trajectory, and stop spraying operation trajectory through a deep learning model, and quickly matches the corresponding spraying trajectory, trial spraying operation trajectory, and stop spraying operation trajectory at the same distance based on the continuous accumulation of the running data of the products to be sprayed that have completed spraying.

[0050] In the embodiment of the present invention, through a series of modular designs, a system suitable for different structures is realized. This system can achieve closed-loop, reliable, and efficient execution through acquisition, analysis, and control.

[0051] This application embodiment also provides an electronic device. The electronic device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the above-mentioned spraying trajectory control method based on deep learning. This electronic device can be any intelligent terminal including a tablet computer, in-vehicle computer, etc.

[0052] The embodiment of the present application also provides a storage medium, which is a computer-readable storage medium. The storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned spraying trajectory control method based on deep learning is implemented.

[0053] As a non-transitory computer-readable storage medium, the memory can be used to store non-transitory software programs and non-transitory computer-executable programs. In addition, the memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories can be connected to the processor through a network. Examples of the above network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0054] The embodiments described in the embodiments of the present application are for more clearly illustrating the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. Those skilled in the art will know that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of the present application are equally applicable to similar technical problems.

[0055] Those skilled in the art can understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than those shown in the figures, or combine certain steps, or different steps.

[0056] Those of ordinary skill in the art can understand that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, and appropriate combinations thereof.

[0057] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above figures are used to distinguish similar objects, and do not have to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present application described here can be implemented in an order other than those illustrated or described here. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0058] It should be understood that in this application, "at least one (item)" means one or more, and "a plurality" means two or more. "And / or" is used to describe the relationship between associated objects and indicates that there can be three relationships. For example, "A and / or B" can mean: only A exists, only B exists, and both A and B exist simultaneously. Here, A and B can be singular or plural. The character " / " generally indicates an "or" relationship between the associated objects before and after. "At least one (item) of the following" or a similar expression refers to any combination of these items, including any combination of a single item or multiple items. For example, at least one (item) of a, b, or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0059] The technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute All or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes: various media that can store programs, such as USB flash drives, mobile hard disks, read-only memories (ROM for short), random access memories (RAM for short), magnetic disks, or optical discs.

[0060] The preferred embodiments of the embodiments of this application have been described above with reference to the accompanying drawings. However, this does not limit the scope of the rights of the embodiments of this application. Any modifications, equivalent replacements, and improvements made by those skilled in the art without departing from the scope and essence of the embodiments of this application shall be within the scope of the rights of the embodiments of this application.

[0061] The above embodiments are only explanations of the present invention and do not limit the present invention. After reading this specification, those skilled in the art can make modifications that do not contribute creatively to this embodiment as needed, but as long as they are within the scope of the claims of the present invention, they are protected by the patent law.

Claims

1. A spraying trajectory control method based on deep learning, characterized by: include: Obtaining the distances between the products to be sprayed and the products to be sprayed, which are respectively the first spacing and the second spacing, and the spraying trajectory of the product to be sprayed; The first spacing and the second spacing are both provided with a trial spraying running track and a stop spraying running track, and each product to be sprayed undergoes at least one spraying track, one trial spraying running track and one stop spraying running track; Control the robot to move the spray gun to the first spacing or the second spacing, select a test spraying point in the test spraying operation trajectory according to the spacing distance of the first spacing or the second spacing, and select a stop spraying point in the stop running spraying trajectory according to the spacing distance of the first spacing or the second spacing; A deep learning model is constructed in advance according to the data sets of the first spacing, the second spacing, the spraying trajectory, the trial spraying running trajectory and the stopped spraying running trajectory acquired multiple times previously; Based on the deep learning model, in the next spraying, the spraying running trajectory and the stopping running spraying trajectory are pre-tested through the intervals of the first interval and the second interval, and the automatic spraying of the product to be sprayed is completed by connecting with the spraying trajectory.

2. A spraying trajectory control method based on deep learning according to claim 1, characterized in that: In the spraying trajectory, the control robot controls the spray gun to perform uniform spraying according to the product to be sprayed; based on the trial spraying operation trajectory, the control robot controls the spray gun to perform early trial spraying according to the distance between the trial spraying point and the product to be sprayed; based on the stop running spraying trajectory, the control robot controls the spray gun to stop spraying according to the distance between the stop spraying point and the product to be sprayed.

3. A spraying trajectory control method based on deep learning according to claim 2, characterized in that: The edge point of the product to be sprayed is used as the starting point and the end point of the spraying trajectory. The trial spraying operation trajectory is the path from the starting point to the trial spraying point, and the stopped spraying trajectory is the path from the end point to the stopped spraying point.

4. A spraying trajectory control method based on deep learning according to claim 3, characterized in that: If the path of the trial spraying running trajectory from the starting point to the trial spraying point is greater than the first spacing or the second spacing, the trial spraying running trajectory is provided with a plurality of trial spraying turning points between the starting point and the trial spraying point, and the control robot controls the spray gun to turn the moving direction of the spray gun according to the trial spraying turning points and the second passing through the starting point to complete the trial spraying running trajectory.

5. The spraying trajectory control method based on deep learning according to claim 3 is characterized in that: If the path between the end point of the stopped spraying trajectory and the stopped spraying point is greater than the first spacing or the second spacing, the stopped spraying trajectory is provided with several stopped spraying turning points between the starting point of the previous product to be sprayed and the stopped spraying point, and the control robot controls the spray gun to turn the moving direction of the spray gun according to the stopped spraying turning point to complete the stopped spraying trajectory.

6. A spraying system, characterized in that: include: An acquisition module is used to acquire the size of a single product to be sprayed and the distance between multiple products to be sprayed; The spraying module controls the spray gun by controlling the robot to sequentially complete the spraying operation of the trial spraying operation track, the spraying track and the spraying operation of the stop spraying track for a single product to be sprayed; The control module predicts the spraying trajectory, the trial spraying operation trajectory and the stopped spraying trajectory through a deep learning model based on the distances before and after the multiple products to be sprayed, and based on the continuously increasing accumulation of operation data of the products to be sprayed that have completed spraying, quickly matches the spraying trajectory, the trial spraying operation trajectory and the stopped spraying trajectory at the same distance.

7. An electronic device, characterized in that: The electronic device comprises a memory and a processor, the memory stores a computer program, and the processor implements the spray trajectory control method according to any one of claims 1 to 5 when executing the computer program.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the spraying trajectory control method according to any one of claims 1 to 5 is implemented.

Images