CN117031986B - Control method and system for automatic cleaning of building curtain wall - Google Patents
Control method and system for automatic cleaning of building curtain wall Download PDFInfo
- Publication number
- CN117031986B CN117031986B CN202311303309.3A CN202311303309A CN117031986B CN 117031986 B CN117031986 B CN 117031986B CN 202311303309 A CN202311303309 A CN 202311303309A CN 117031986 B CN117031986 B CN 117031986B
- Authority
- CN
- China
- Prior art keywords
- curtain wall
- control
- cleaning
- data
- node
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000004140 cleaning Methods 0.000 title claims abstract description 200
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000001514 detection method Methods 0.000 claims abstract description 48
- 238000004088 simulation Methods 0.000 claims abstract description 27
- 230000004044 response Effects 0.000 claims abstract description 23
- 238000013461 design Methods 0.000 claims abstract description 13
- 230000003993 interaction Effects 0.000 claims abstract description 12
- 230000000452 restraining effect Effects 0.000 claims abstract description 6
- 230000003749 cleanliness Effects 0.000 claims description 24
- 238000005457 optimization Methods 0.000 claims description 14
- 230000003044 adaptive effect Effects 0.000 claims description 12
- 230000011218 segmentation Effects 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000003709 image segmentation Methods 0.000 claims description 5
- 230000000694 effects Effects 0.000 abstract description 14
- 230000006870 function Effects 0.000 description 14
- 230000008569 process Effects 0.000 description 12
- 230000033001 locomotion Effects 0.000 description 11
- 238000005507 spraying Methods 0.000 description 8
- 239000000428 dust Substances 0.000 description 6
- 230000007613 environmental effect Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 239000007788 liquid Substances 0.000 description 4
- 239000000463 material Substances 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 230000000670 limiting effect Effects 0.000 description 3
- 230000002829 reductive effect Effects 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 239000007921 spray Substances 0.000 description 2
- 238000004381 surface treatment Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000004888 barrier function Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000036961 partial effect Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000035699 permeability Effects 0.000 description 1
- 230000036544 posture Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000009877 rendering Methods 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 239000002344 surface layer Substances 0.000 description 1
- 238000010408 sweeping Methods 0.000 description 1
- 230000026676 system process Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000005406 washing Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B17/00—Systems involving the use of models or simulators of said systems
- G05B17/02—Systems involving the use of models or simulators of said systems electric
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The application discloses a control method and a system for automatic cleaning of a building curtain wall, belonging to the field of intelligent control, wherein the method comprises the following steps: collecting design information and establishing a curtain wall data set of a building curtain wall; establishing a simulation model by using a curtain wall data set, fitting a cleaning device to the simulation model, and completing path planning by restraining response accuracy of the cleaning device; configuring control parameters of the cleaning device; configuring an obstacle avoidance control node interval based on path planning, and executing obstacle feedback detection of the interaction sensor in the obstacle avoidance control node interval; generating real-time compensation data according to the obstacle feedback detection result, optimizing control data by using the real-time compensation data, and generating optimized control parameters; and (5) carrying out automatic cleaning control on the building curtain wall by optimizing control parameters. The application solves the technical problems of low cleaning operation efficiency and poor cleaning effect of the building curtain wall in the prior art, and achieves the technical effects of improving the automatic cleaning effect and the cleaning operation efficiency of the building curtain wall.
Description
Technical Field
The application relates to the field of intelligent control, in particular to a control method and a system for automatic cleaning of a building curtain wall.
Background
With the development of the building industry, curtain walls are widely used as important building facade forms. The outer vertical face of the building curtain wall is easy to be polluted, and regular cleaning maintenance is needed to ensure the permeability and the comfortableness of the outer vertical face and the internal environment of the building. At present, the building curtain wall cleaning industry is mainly manual cleaning and machine cleaning, however, the automatic cleaning efficiency of the existing building curtain wall is still limited, the operation period is longer, the adaptability to complex facade environments is poor, and the cleaning is difficult to accurately ensure to meet the expected cleanliness requirement. For high-rise buildings, the existing curtain wall automatic cleaning is difficult to realize intelligent obstacle avoidance and path planning, so that the cleaning effect is poor.
Disclosure of Invention
The application provides a control method and a system for automatically cleaning a building curtain wall, and aims to solve the technical problems of low cleaning operation efficiency and poor cleaning effect of the building curtain wall in the prior art.
In view of the above problems, the application provides a control method and a system for automatic cleaning of a building curtain wall.
In a first aspect of the disclosure, a control method for automatic cleaning of a building curtain wall is provided, the method comprising: building a curtain wall data set of the building curtain wall, wherein the curtain wall data set is a curtain wall data set constructed by collecting design information, and comprises size data and curtain wall characteristic data; establishing a simulation model by using a curtain wall data set, fitting a cleaning device to the simulation model, and completing path planning by restraining response accuracy of the cleaning device; configuring control parameters of the cleaning device, wherein the control parameters take path planning and curtain wall data sets as basic data, curtain wall cleanliness as auxiliary parameters, and processing the generated control data through a fitting control network; configuring an obstacle avoidance control node interval based on path planning, and executing obstacle feedback detection of the interaction sensor in the obstacle avoidance control node interval; generating real-time compensation data according to the obstacle feedback detection result, optimizing the control data by using the real-time compensation data, generating optimized control parameters, and carrying out automatic cleaning control on the building curtain wall through the optimized control parameters.
In another aspect of the present disclosure, a control system for automated cleaning of building curtain walls is provided, the system comprising: the curtain wall data set building module is used for building a curtain wall data set of a building curtain wall, wherein the curtain wall data set is a curtain wall data set built by collecting design information, and comprises size data and curtain wall characteristic data; the simulation model building module is used for building a simulation model by using the curtain wall data set, fitting the cleaning device to the simulation model, and completing path planning by restraining the response precision of the cleaning device; the control coefficient configuration module is used for configuring control parameters of the cleaning device, wherein the control parameters take path planning and curtain wall data sets as basic data, curtain wall cleanliness as auxiliary parameters, and the generated control data are processed through a fitting control network; the obstacle feedback detection module is used for configuring an obstacle avoidance control node interval based on path planning, and executing obstacle feedback detection of the interaction sensor in the obstacle avoidance control node interval; the control data optimizing module is used for generating real-time compensation data according to the obstacle feedback detection result, optimizing the control data by the real-time compensation data and generating optimized control parameters; and the automatic cleaning control module is used for carrying out automatic cleaning control on the building curtain wall through optimizing control parameters.
One or more technical schemes provided by the application have at least the following technical effects or advantages:
because the curtain wall data set is constructed by adopting the collected design information, the curtain wall data set comprises data such as size, characteristics and the like, so that the subsequent modeling and control are facilitated; establishing a simulation model by using a curtain wall data set, fitting a cleaning device, and completing path planning aiming at response precision of the device so as to achieve the purpose of intelligent planning; based on path planning and curtain wall data sets, the control parameters for configuring the cleaning device are generated through control network processing by combining the cleanliness requirement, so that automatic optimization configuration of the parameters is realized; setting an obstacle avoidance control area aiming at path planning, and adopting a sensor to perform obstacle feedback detection so as to realize automatic obstacle avoidance; generating real-time compensation data by using obstacle feedback, optimizing control data, and realizing real-time compensation and optimization of control parameters; through the optimized control parameters, the automatic cleaning process of the curtain wall is controlled, the technical scheme of accurate and efficient cleaning operation is realized, the technical problems of low cleaning operation efficiency and poor cleaning effect of the building curtain wall in the prior art are solved, and the technical effects of improving the automatic cleaning effect and the cleaning operation efficiency of the building curtain wall are achieved.
The foregoing description is only an overview of the present application, and is intended to be implemented in accordance with the teachings of the present application in order that the same may be more clearly understood and to make the same and other objects, features and advantages of the present application more readily apparent.
Drawings
FIG. 1 is a schematic flow chart of a control method for automatically cleaning a building curtain wall according to an embodiment of the application;
FIG. 2 is a schematic flow chart of adjusting optimized control parameters in a control method for automatically cleaning a building curtain wall according to an embodiment of the present application;
fig. 3 is a schematic structural view of a control system for automatic cleaning of building curtain walls according to an embodiment of the present application.
Reference numerals illustrate: the system comprises a curtain wall data set building module 11, a simulation model building module 12, a control coefficient configuration module 13, an obstacle feedback detection module 14, a control data optimization module 15 and an automatic cleaning control module 16.
Detailed Description
The technical scheme provided by the application has the following overall thought:
the embodiment of the application provides a control method and a system for automatic cleaning of a building curtain wall, which realize the accurate control of the automatic cleaning process of the building curtain wall and effectively improve the cleaning efficiency and effect by constructing a digital curtain wall data set, establishing a simulation environment for path planning, utilizing a control network to generate optimized control parameters, configuring real-time obstacle avoidance feedback control and other technical means.
Specifically, firstly, building design information is collected, and a digital curtain wall data set containing data such as size, characteristics and the like is constructed to serve as basic data for subsequent modeling and simulation; then, a simulation model is established according to the curtain wall data set, in the established simulation environment, the cleaning device is fitted, and path planning is conducted according to the response precision requirement of the device, so that intelligent planning of a cleaning path is completed; then, comprehensively considering the cleanliness requirement based on a path planning result and a curtain wall data set, generating control parameters through training treatment of a control network, and realizing automatic optimal configuration of the control parameters of the cleaning process; in order to realize obstacle avoidance, an obstacle avoidance control interval is further arranged on the basis of path planning, obstacle feedback detection is carried out by adopting real-time sensing, and control parameters are adjusted through feedback results, so that real-time compensation and optimization of a cleaning path are realized, and optimized control parameters are obtained; finally, the automatic cleaning control of the building curtain wall is realized by optimizing the control parameters, the accurate automatic cleaning control of the complex building curtain wall is realized, and the cleaning efficiency and the cleaning quality are improved.
Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.
Example 1
As shown in fig. 1, an embodiment of the present application provides a control method for automatic cleaning of a building curtain wall, the method comprising:
building a curtain wall data set of a building curtain wall, wherein the curtain wall data set is a curtain wall data set constructed by collecting design information, and comprises size data and curtain wall characteristic data;
in the embodiment of the application, a curtain wall data set refers to a set of technical parameters and characteristic data of a building curtain wall to be cleaned, and comprises dimension data and curtain wall characteristic data, wherein the dimension data refers to specific dimensions of the curtain wall, such as data of length, width, thickness and the like; curtain wall characteristic data refer to curtain wall characteristics affecting cleaning control, such as data of materials, surface treatment, spacing, whether obstacle avoidance protrusions exist or not.
First, referring to a building design drawing to obtain design dimension data of the curtain wall, such as dimension parameters of length, width, height, thickness and the like; meanwhile, the specification of curtain wall materials is consulted, and information such as specific materials and structures of the curtain wall is known. Secondly, carrying out field measurement on the actually built curtain wall to obtain accurate size data; and (5) observing the actual conditions of the curtain wall, such as the characteristics of a surface treatment mode, a component connection mode, obstacle avoidance protrusions and the like. Then, the collected size data are tidied and recorded to form digital size data; and classifying and recording the collected curtain wall characteristics to form structural curtain wall characteristic data. The dimensional data and curtain wall characterization data are then integrated into a curtain wall dataset.
By systematically collecting and arranging various key data forming a curtain wall data set, a foundation is laid for subsequent modeling simulation and cleaning control parameter design.
Establishing a simulation model by using the curtain wall data set, fitting a cleaning device to the simulation model, and completing path planning by restraining response accuracy of the cleaning device;
in the embodiment of the application, after a curtain wall data set of a building curtain wall is obtained, firstly, the collected curtain wall data set is imported into three-dimensional modeling software, a modeling module is called in the three-dimensional modeling software, and a high-precision integral model of the three-dimensional digital twin curtain wall is constructed according to the dimension parameters contained in the imported curtain wall data set. Then, a rendering module is called, and material textures are added on the established digital twin curtain wall model according to curtain wall characteristic data in the curtain wall data set, so that the appearance visual effect is highly vivid; and adding a model of the protruding structure at a proper position according to parameters such as the curtain wall structure and the like. And further, the physical characteristics of the digital twin curtain wall model, such as parameters of mass, friction coefficient and the like, are set so that the physical behaviors of movement, collision and the like are consistent with the physical behaviors of objects. Then, a three-dimensional model representing the actual cleaning device is imported, the kinematic parameters of the cleaning device are adjusted, and a reasonable control precision range is set, so that the cleaning device can realize the motion response characteristic matched with a real object in a digital environment. And then, based on the established digital twin environment, taking factors such as coverage area, obstacle avoidance performance and the like into consideration, calling a path planning algorithm module to plan an initial cleaning route and finish path planning.
By applying the digital twin technology, the time cost of actual test errors is saved, and the fact that an automatic cleaning system is subjected to detailed virtual verification before actual use is ensured, so that the reliability and the intelligence of control are improved.
Configuring control parameters of the cleaning device, wherein the control parameters take the path planning and the curtain wall data set as basic data, curtain wall cleanliness as auxiliary parameters, and the generated control data are processed through a fitting control network;
further, generating curtain wall cleanliness specifically includes:
cleaning information is called for the building curtain wall, and a call data set is generated;
collecting an environment data set, wherein the environment data set is an environment characteristic set after the last cleaning;
and taking the current date as a reference standard, and carrying out pollution prediction according to the calling data set and the environment data set to generate the curtain wall cleanliness.
In one possible embodiment, control parameters for the cleaning device are generated for fitting to the control network process, requiring curtain wall cleanliness as an auxiliary parameter. Firstly, monitoring the cleaning condition and frequency of a curtain wall in real time through sensors such as an image sensor, a pressure sensor and the like deployed on the curtain wall of a building, sending the monitoring data to a central server, and forming a complete curtain wall cleaning log at the server end, wherein the log contains information such as time, cleaning range, cleaning strength and the like of each cleaning; and calling the historical cleaning logs to extract and process data to form a calling data set. Secondly, various sensors arranged on the curtain wall are used, after each cleaning, the environment is collected and monitored, environmental parameters such as air quality, temperature and humidity, dust accumulation rate and the like are obtained, and the environmental data after the last cleaning is called to form an environmental characteristic set which is an environmental data set. Then, a prediction model based on deep learning is constructed, and the prediction model is input into a current date, a calling data set and an environment data set and output into the expected pollution degree of the curtain wall. And then, based on the constructed call data set and the environment data set, carrying out pollution prediction on the curtain wall by combining the current date, and then making a cleaning requirement on the building curtain wall according to the predicted pollution prediction degree to generate the expected curtain wall cleanliness.
After the cleanliness of the curtain wall is obtained, obtaining a path plan of the cleaning device from the digital twin environment, and taking the path plan as one of basic data generated by control parameters; meanwhile, the established curtain wall data set is read, and data such as size parameters, obstacle avoidance structure parameters, layout parameters and the like of the curtain wall are analyzed and extracted to serve as one of basic data for generating control parameters. And then, on the basis of path planning parameters, comprehensively considering a curtain wall data set, adjusting speed parameters and acceleration parameters, and realizing the adaptation of the actual curtain wall size and layout to form adaptive control parameters. And then, constructing a fitting control network in the three-dimensional digital twin curtain wall by taking the path planning and the curtain wall data set as basic data, performing cleaning simulation on the fitting control network to obtain the simulated curtain wall cleanliness, and comparing the simulated curtain wall cleanliness with the curtain wall cleanliness. If the cleanliness of the curtain wall is met, the current control data is stored and used as the control parameter of the cleaning device; if the cleanliness of the curtain wall is not met, the current control data are adjusted until the cleanliness of the simulated curtain wall meets the cleanliness of the curtain wall, and the control parameters of the cleaning device are obtained.
Configuring an obstacle avoidance control node interval based on the path planning, and executing obstacle feedback detection of the interaction sensor in the obstacle avoidance control node interval;
further, the method specifically comprises the following steps:
setting an adaptive node detection trigger function, wherein the adaptive node detection trigger function is as follows:
；
wherein,detecting a trigger function for an adaptive node of an ith node,>is the node coordinates of the i-th node,for the node coordinates of the last trigger node, +.>Stability factor for the i-1 th trigger node, < ->Stability factor of pre-trigger node cluster being i-1 th trigger node,/for the pre-trigger node cluster>Weight coefficient for i-1 th trigger node,/->T is a preset period length for triggering the weight coefficient of the node cluster;
and judging whether to execute obstacle feedback detection of the interaction sensor in the obstacle avoidance control node interval or not through the self-adaptive node detection triggering function.
In a preferred embodiment, path planning is analyzed, key points such as inflection points of paths and path junctions are judged to serve as obstacle avoidance control nodes, all the obstacle avoidance control nodes are marked on the planned clean paths, an obstacle avoidance control node interval is obtained, and a foundation is laid for obstacle avoidance detection.
Then, in the obstacle avoidance control node section, whether an obstacle exists in the cleaning work area is monitored in real time through an interactive sensor (such as TOF laser radar, RGB-D cameras and the like) mounted on the cleaning device. And whether the obstacle feedback detection of the interaction sensor is executed in the obstacle avoidance control node interval is judged by the self-adaptive node detection trigger function. Preferably, the adaptive node detection trigger function isWherein->Detecting a trigger function for an adaptive node of an ith node,>node coordinates for the ith node, +.>For the node coordinates of the last trigger node, +.>Stability factor for the i-1 th trigger node, < ->Stability factor of pre-trigger node cluster being i-1 th trigger node,/for the pre-trigger node cluster>Weight coefficient for i-1 th trigger node,/->T is a preset period length for triggering the weight coefficient of the node cluster;through the function, whether the obstacle detection needs to be executed at the current obstacle avoidance control node i or not can be calculated and judged in real time.
According to the self-adaptive node detection trigger function, when the cleaning device moves drastically or the position change is abnormal between the nodes,the value will be larger, indicating that obstacle avoidance detection of the current node i needs to be enhanced. Conversely, if the position change is smooth, the position is +>And if the detection frequency of the current node is reduced or the current node is not detected, so that the node obstacle avoidance detection strategy is adaptively adjusted according to the real-time motion state of the cleaning device.
Generating real-time compensation data according to the obstacle feedback detection result, optimizing the control data by using the real-time compensation data, and generating optimized control parameters;
in the embodiment of the application, in the process of cleaning the building curtain wall by the cleaning device, the interactive sensor on the cleaning device can scan the operation space and feed back the obstacle information to obtain the obstacle feedback detection result, and various obstacles in the space are reflected. After the obstacle feedback detection result is obtained, the control system calculates the compensation control quantity for the cleaning device to avoid the obstacle and adjust the path according to the specific position, the size and other data of the obstacle, and real-time compensation data is obtained. For example, if the obstacle feedback detection result shows that an outstanding obstacle is detected in front of the obstacle feedback detection result, calculating the distance and the speed compensation of the lateral avoidance; for another example, if sundries are detected on the curtain wall, the corresponding attack angle and speed compensation are calculated. And then feeding the obtained real-time compensation data back to a control system of the cleaning device, correcting and optimizing the original control parameters, and generating new optimized control parameters so as to realize autonomous avoidance of the obstacle, thereby adaptively adjusting the control parameters and realizing closed-loop control of flexible obstacle avoidance.
And carrying out automatic cleaning control on the building curtain wall through the optimized control parameters.
In the embodiment of the application, the optimized control parameters aiming at the current cleaning state are obtained, and compared with the original control plan, the control parameters are more accurate and reliable, so that the cleaning device can flexibly avoid and control the movement according to the obstacles in the actual environment. The driving system of the cleaning device directly calls the optimized control parameters as input instructions, and drives the cleaning device to move according to the control quantities such as paths, speeds, waypoints, postures and the like contained in the parameters so as to accurately and rapidly avoid space barriers and complete automatic cleaning operation of the building curtain wall.
Autonomous control and optimization adjustment of the cleaning device are realized through a closed loop feedback mode, the self-adaptive capacity and the motion control precision of the clean environment of the curtain wall are improved, and the automatic and intelligent cleaning of the building curtain wall is realized.
Further, as shown in fig. 2, the embodiment of the present application further includes:
when the cleaning device executes building curtain wall cleaning of any area, triggering a data acquisition instruction;
controlling a CCD sensor to acquire the advancing path and the building curtain wall data of the next area through the data acquisition instruction, and constructing a curtain wall image;
taking the obstacle avoidance protrusion as a segmentation feature, performing image segmentation on the curtain wall image to generate a segmentation image, wherein the segmentation image comprises a path image and a neighborhood image;
and adjusting the optimal control parameters according to the path image and the neighborhood image.
In a preferred embodiment, the system triggers the data acquisition command to activate the CCD sensor on the cleaning device into an operational mode to detect and acquire data from the current and surrounding environment while the cleaning device is performing a cleaning operation on an area of the building curtain. Then, the data acquisition instruction controls the CCD sensor to adjust the angle, simultaneously starts an image acquisition mode, aims at the advancing direction of the device, continuously shoots the curtain wall in the front path area, and acquires the image information in the visual field range of the route. Meanwhile, the CCD sensor properly swings and adjusts the lens left and right according to the movement direction of the device, the visual field range is enlarged, and the image of the curtain wall in the next operation area is acquired. The two partial images are then brought together into an integral curtain wall image covering the current and subsequent working areas. The image reflects the information of the spatial structure, dirt distribution and the like of the path and the surrounding area, and provides image data support for subsequent closed-loop control.
Then, the control system uses image processing algorithm, such as Canny edge detection, to identify and extract the edge contours of the protruding structures by taking the structural protrusions of the building curtain wall as the image segmentation features. Based on the convex contours of the cleaning devices, which need to avoid the obstacle, as the segmentation basis, an image segmentation method is adopted to segment the whole curtain wall image into a path image and a neighborhood image. The path image comprises an image area in front of the movement direction of the cleaning device, and reflects the field of view of the current cleaning path; the neighborhood image includes a peripheral region image, corresponding to a space to be entered in the next stage of the device. After the path image and the neighborhood image are obtained, the system generates a series of updated values of control parameters such as speed, path adjustment, spraying density and the like according to the strategies of obstacle avoidance and cleaning optimization, and feeds back and adjusts the updated control parameters to the control system of the cleaning device to complete the optimization of motion control so as to adapt to the actual environment.
Further, the embodiment of the application further comprises:
establishing a dirty characteristic set, wherein the dirty characteristic set is constructed through big data;
performing dirt recognition on the neighborhood image through the dirt feature set, and generating cleaning features according to recognition results, wherein the recognition results comprise dirt types and dirt areas;
the optimization control parameter is adjusted based on the cleaning characteristics.
In one possible implementation, first, a large number of building fouling images are collected and marked by using a web crawler, a fouling feature set is obtained, and a fouling feature extraction model is trained by machine learning through the feature set, wherein the feature set comprises various image examples of common fouling and feature vector representations, and the image examples comprise different types of fouling features such as stains, dust and the like. Then, preprocessing the neighborhood image, including denoising, enhancing and the like, inputting the processed neighborhood image into a dirty feature extraction model, matching the model with samples in a dirty feature set to obtain a recognition result, wherein the recognition result comprises types of dirty in the image, and simultaneously giving out a specific coverage area of each type of dirty. After the dirt type and area are obtained, a cleaning characteristic indicating the cleaning requirement is generated according to the identification result, for example, the type of cleaning liquid is adjusted, the spraying area is increased, the running speed of the cleaning device is adjusted, and the like. If more dust stains are detected, cleaning liquid for specifically removing dust is used; if a larger area of stains exists, the spraying area is increased to completely remove the stains; if no soil is detected, the running speed of the cleaning device can be increased, etc.
The system then uses the generated cleaning characteristics, which are indicative of the soiling of the area to be cleaned, to adjust and optimize the control parameters of the cleaning device accordingly. For example, when a large area of dust and dirt is identified, the air pressure of the nozzle is increased to enhance the spraying force, and the sweeping range of the nozzle is widened; if the condition of uneven distribution of stains is detected, non-uniform spraying is carried out, and the spraying times are increased for the heavy-dirt area; for more stubborn stains, the moving speed is reduced to increase the spraying action time; the matched cleaning liquid is selected for different types of dirt, or multiple cleaning liquids are prepared for simultaneous use. The cleaning device can intelligently adapt to complex environments by pertinently adjusting control parameters according to dirt conditions, so that the cleaning process is optimized, and the effective automatic cleaning operation is completed.
Further, the embodiment of the application further comprises:
establishing a matching coefficient of the running speed and the rotating speed;
generating a speed adjustment value for the operating speed from the cleaning feature;
performing associated calculation of the rotation speed according to the speed adjustment value and the matching coefficient to generate an adjustment rotation speed;
and adjusting the optimized control parameter according to the speed adjustment value and the adjusted rotation speed.
In a preferred embodiment, the running speed and the rotation speed of the cleaning device have a cooperative relationship, and in order to achieve optimization of the cleaning effect, the cooperative coefficients of the running speed and the rotation speed are obtained through theoretical calculation and experimental test, so that the best matched rotation speed is determined at each running speed, and the spraying coverage is uniform. For example, at an operating speed of 1 meter per second, the optimum rotational speed of the spray head is 120 revolutions per second, and so on, and the matching coefficients of different operating speeds and different rotational speeds are established. The control system then determines the amount of speed change, i.e., the speed adjustment value, that the operating speed of the cleaning device needs to be adjusted based on the generated cleaning characteristics. For example, when a small amount of dust is recognized on the surface, the current normal operating speed is maintained, and the speed adjustment value is 0; if more refractory stains are detected, the speed is required to be reduced to increase the spray washing time, and the speed adjustment value is-0.1 m/s; for large-area stains with shallow surface layers, the working speed is properly increased, and the speed adjustment value is +0.05m/s. And the speed adjustment is intelligently planned according to the cleaning characteristics, so that the cleaning process is optimally controlled, and the running action of the device is adapted to the environmental requirement.
And then, carrying out the correlation calculation of the rotation speed based on the obtained speed adjustment value and the established matching coefficient of the running speed and the rotation speed, and generating the adjustment rotation speed of the rotation speed. For example, the control system queries the matching coefficient of the running speed and the rotating speed, finds that the rotating speed value corresponding to the current running speed is 120 revolutions/min, calculates the rotating speed corresponding to the running speed adjusted according to the speed adjusting value according to the matching coefficient, and obtains the adjusted rotating speed, for example, -10 revolutions/min. The coordination of the running speed and the rotating speed is ensured, and the kinematic conflict is avoided. Then, the control system feeds the obtained speed adjustment value and the corresponding adjustment rotation speed back to the cleaning device, and the adjustment of the optimized control parameters of the cleaning device is completed together. For example, the actual running speed control parameter is adjusted to the original amount by 0.1m/s, and the rotation speed control parameter is adjusted to the original amount by 10 revolutions/min. The running speed and the rotating speed of the cleaning device are adjusted simultaneously, so that kinematic coordination of the cleaning device is ensured, and the cleaning process is optimized.
Further, the embodiment of the application further comprises:
carrying out intra-image feature recognition on the path image to generate an obstacle recognition result;
establishing an avoidance track according to the obstacle recognition result;
establishing a response safety space of the cleaning device, expanding the avoidance track by using the response safety space, and generating an adjustment track;
and carrying out avoidance control of automatic cleaning based on the adjustment track.
In a preferred embodiment, the control system processes the path image, extracts image content characteristics, matches the image content characteristics with a pre-stored obstacle characteristic library, and judges whether obstacle targets such as protrusions and sundries exist in the image. For example, a convolutional neural network is used for processing the path image, identifying whether a convex or hard sundries appear on a travel path, and then giving information such as specific position coordinates, occupied space size and the like of the obstacle to form an obstacle identification result. If no obstacle is detected, the obstacle recognition result is null; if the obstacle is detected, specific information of the obstacle is given, and a basis is provided for subsequent obstacle avoidance track planning. Then, if the obstacle recognition result contains specific position and occupied space information of the obstacle, the control system calculates an avoidance track which bypasses the obstacle according to the obstacle avoidance principle, and the avoidance track can meet the condition that the cleaning device main body and the obstacle keep a safe distance as far as possible. For example, if a bulge of 0.5 m wide is identified 2 m in front of the bulge, the control system plans to detour to the left of the bulge, the distribution of the trace points keeping the cleaning device at a distance of 0.2 m from the pipe.
Subsequently, in step S753, the system needs to build a motion response safety space model of the cleaning device, and expand the pre-planned avoidance trajectory with the safety space to obtain an adjusted final trajectory.
Because the cleaning device has certain mechanical response inertia, the millisecond-level rapid obstacle avoidance cannot be realized, and therefore, a response buffer zone of the cleaning device, namely a response safety space, is established according to dynamic parameters such as the braking distance, the steering amplitude and the like of the cleaning device. And then, expanding a response safety space outside the planned avoidance track to form an increased adjustment track, wherein the adjustment track is smoother and buffered compared with the avoidance track, and can adapt to the response requirement of the device so as to realize stable and safe automatic obstacle avoidance of the cleaning device. After a smooth and safe adjustment track is obtained, the cleaning device inputs data of the track as a reference for motion control, and the cleaning device obtains the position of the cleaning device in real time through a course sensor in the motion process of the cleaning device and performs closed-loop feedback comparison with the adjustment track. If the deviation track is detected, the compensation control quantity is automatically calculated, the wheel set and the steering system are driven to enable the device to be readjusted, the device is locked on the adjustment track, the closed-loop obstacle avoidance control of the cleaning device is carried out, and the capability of automatically avoiding obstacles on a path is achieved.
In summary, the control method for automatically cleaning the building curtain wall provided by the embodiment of the application has the following technical effects:
and (3) building a curtain wall data set of the building curtain wall, wherein the curtain wall data set is a curtain wall data set constructed by collecting design information, and comprises size data and curtain wall characteristic data, so as to acquire digital data of the curtain wall and provide basic data for subsequent modeling and control. And establishing a simulation model by using the curtain wall data set, fitting the cleaning device to the simulation model, and completing path planning by restricting the response precision of the cleaning device so as to provide a reference scheme for cleaning control. And configuring control parameters of the cleaning device, wherein the control parameters take path planning and curtain wall data sets as basic data, curtain wall cleanliness as auxiliary parameters, and the intelligent and accurate configuration of parameters is realized by fitting control data generated by control network processing. And configuring an obstacle avoidance control node interval based on path planning, and executing obstacle feedback detection of the interaction sensor in the obstacle avoidance control node interval to realize an obstacle avoidance function in the cleaning process. Generating real-time compensation data according to the obstacle feedback detection result, optimizing the control data by using the real-time compensation data, generating optimized control parameters, and compensating and optimizing the cleaning path. The automatic cleaning control of the building curtain wall is carried out by optimizing the control parameters, so that the technical effects of improving the automatic cleaning effect and the cleaning operation efficiency of the building curtain wall are achieved.
Example two
Based on the same inventive concept as the control method for automatic cleaning of building curtain walls in the foregoing embodiments, as shown in fig. 3, an embodiment of the present application provides a control system for automatic cleaning of building curtain walls, including:
the curtain wall data set establishing module 11 is used for establishing a curtain wall data set of a building curtain wall, wherein the curtain wall data set is a curtain wall data set constructed by collecting design information, and comprises size data and curtain wall characteristic data;
a simulation model building module 12, configured to build a simulation model from the curtain wall data set, fit a cleaning device to the simulation model, and complete path planning by restricting response accuracy of the cleaning device;
the control coefficient configuration module 13 is configured to configure control parameters of the cleaning device, wherein the control parameters take the path planning and the curtain wall data set as basic data, take the curtain wall cleanliness as auxiliary parameters, and process the generated control data through fitting a control network;
the obstacle feedback detection module 14 is configured to configure an obstacle avoidance control node interval based on the path planning, and perform obstacle feedback detection of the interaction sensor in the obstacle avoidance control node interval;
the control data optimizing module 15 is used for generating real-time compensation data according to the obstacle feedback detection result, optimizing the control data by using the real-time compensation data, and generating optimized control parameters;
and the automatic cleaning control module 16 is used for carrying out automatic cleaning control on the building curtain wall through the optimized control parameters.
Further, the obstacle feedback detection module 14 includes the following execution steps:
setting an adaptive node detection trigger function, wherein the adaptive node detection trigger function is as follows:
；
wherein,detecting a trigger function for an adaptive node of an ith node,>is the node coordinates of the i-th node,for the node coordinates of the last trigger node, +.>Stability factor for the i-1 th trigger node, < ->Stability factor of pre-trigger node cluster being i-1 th trigger node,/for the pre-trigger node cluster>Weight coefficient for i-1 th trigger node,/->T is a preset period length for triggering the weight coefficient of the node cluster;
and judging whether to execute obstacle feedback detection of the interaction sensor in the obstacle avoidance control node interval or not through the self-adaptive node detection triggering function.
Further, the control coefficient configuration module 13 includes the following execution steps:
cleaning information is called for the building curtain wall, and a call data set is generated;
collecting an environment data set, wherein the environment data set is an environment characteristic set after the last cleaning;
and taking the current date as a reference standard, and carrying out pollution prediction according to the calling data set and the environment data set to generate the curtain wall cleanliness.
Further, the embodiment of the application also comprises an optimization control parameter adjustment module, which comprises the following execution steps:
when the cleaning device executes building curtain wall cleaning of any area, triggering a data acquisition instruction;
controlling a CCD sensor to acquire the advancing path and the building curtain wall data of the next area through the data acquisition instruction, and constructing a curtain wall image;
taking the obstacle avoidance protrusion as a segmentation feature, performing image segmentation on the curtain wall image to generate a segmentation image, wherein the segmentation image comprises a path image and a neighborhood image;
and adjusting the optimal control parameters according to the path image and the neighborhood image.
Further, the optimization control parameter adjustment module further includes the following execution steps:
establishing a dirty characteristic set, wherein the dirty characteristic set is constructed through big data;
performing dirt recognition on the neighborhood image through the dirt feature set, and generating cleaning features according to recognition results, wherein the recognition results comprise dirt types and dirt areas;
the optimization control parameter is adjusted based on the cleaning characteristics.
Further, the optimization control parameter adjustment module further includes the following execution steps:
establishing a matching coefficient of the running speed and the rotating speed;
generating a speed adjustment value for the operating speed from the cleaning feature;
performing associated calculation of the rotation speed according to the speed adjustment value and the matching coefficient to generate an adjustment rotation speed;
and adjusting the optimized control parameter according to the speed adjustment value and the adjusted rotation speed.
Further, the embodiment of the application also comprises an avoidance control module, which comprises the following execution steps:
carrying out intra-image feature recognition on the path image to generate an obstacle recognition result;
establishing an avoidance track according to the obstacle recognition result;
establishing a response safety space of the cleaning device, expanding the avoidance track by using the response safety space, and generating an adjustment track;
and carrying out avoidance control of automatic cleaning based on the adjustment track.
Any of the steps of the methods described above may be stored as computer instructions or programs in a non-limiting computer memory and may be called by a non-limiting computer processor to identify any method for implementing an embodiment of the present application, without unnecessary limitations.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the scope of the application. Thus, the present application is intended to include such modifications and alterations insofar as they come within the scope of the application or the equivalents thereof.
Claims (2)
1. A control method for automatic cleaning of a building curtain wall, the method comprising:
building a curtain wall data set of a building curtain wall, wherein the curtain wall data set is a curtain wall data set constructed by collecting design information, and comprises size data and curtain wall characteristic data;
establishing a simulation model by using the curtain wall data set, fitting a cleaning device to the simulation model, and completing path planning by restraining response accuracy of the cleaning device;
configuring control parameters of the cleaning device, wherein the control parameters take the path planning and the curtain wall data set as basic data, curtain wall cleanliness as auxiliary parameters, and the generated control data are processed through a fitting control network;
configuring an obstacle avoidance control node interval based on the path planning, and executing obstacle feedback detection of the interaction sensor in the obstacle avoidance control node interval;
generating real-time compensation data according to the obstacle feedback detection result, optimizing the control data by using the real-time compensation data, and generating optimized control parameters;
carrying out automatic cleaning control on the building curtain wall through the optimized control parameters;
the method further comprises the steps of:
setting an adaptive node detection trigger function, wherein the adaptive node detection trigger function is as follows:
；
wherein,detecting a trigger function for an adaptive node of an ith node,>node coordinates for the ith node, +.>For the node coordinates of the last trigger node, +.>Stability factor for the i-1 th trigger node, < ->Stability factor of pre-trigger node cluster being i-1 th trigger node,/for the pre-trigger node cluster>Weight coefficient for i-1 th trigger node,/->T is a preset period length for triggering the weight coefficient of the node cluster;
judging whether to execute obstacle feedback detection of the interaction sensor in the obstacle avoidance control node interval or not through the self-adaptive node detection triggering function;
cleaning information is called for the building curtain wall, and a call data set is generated;
collecting an environment data set, wherein the environment data set is an environment characteristic set after the last cleaning;
taking the current date as a reference standard, and carrying out pollution prediction according to the calling data set and the environment data set to generate the curtain wall cleanliness;
when the cleaning device executes building curtain wall cleaning of any area, triggering a data acquisition instruction;
controlling a CCD sensor to acquire the advancing path and the building curtain wall data of the next area through the data acquisition instruction, and constructing a curtain wall image;
taking the obstacle avoidance protrusion as a segmentation feature, performing image segmentation on the curtain wall image to generate a segmentation image, wherein the segmentation image comprises a path image and a neighborhood image;
adjusting the optimized control parameters according to the path image and the neighborhood image;
establishing a dirty characteristic set, wherein the dirty characteristic set is constructed through big data;
performing dirt recognition on the neighborhood image through the dirt feature set, and generating cleaning features according to recognition results, wherein the recognition results comprise dirt types and dirt areas;
adjusting the optimal control parameters based on the cleaning characteristics;
establishing a matching coefficient of the running speed and the rotating speed;
generating a speed adjustment value for the operating speed from the cleaning feature;
performing associated calculation of the rotation speed according to the speed adjustment value and the matching coefficient to generate an adjustment rotation speed;
adjusting the optimized control parameter according to the speed adjustment value and the adjusted rotation speed;
carrying out intra-image feature recognition on the path image to generate an obstacle recognition result;
establishing an avoidance track according to the obstacle recognition result;
establishing a response safety space of the cleaning device, expanding the avoidance track by using the response safety space, and generating an adjustment track;
and carrying out avoidance control of automatic cleaning based on the adjustment track.
2. A control system for automated cleaning of building curtain walls, for implementing a control method for automated cleaning of building curtain walls as claimed in claim 1, said system comprising:
the curtain wall data set building module is used for building a curtain wall data set of a building curtain wall, wherein the curtain wall data set is a curtain wall data set constructed by collecting design information, and comprises size data and curtain wall characteristic data;
the simulation model building module is used for building a simulation model by using the curtain wall data set, fitting a cleaning device to the simulation model, and completing path planning by restraining response accuracy of the cleaning device;
the control coefficient configuration module is used for configuring control parameters of the cleaning device, wherein the control parameters take the path planning and the curtain wall data set as basic data, curtain wall cleanliness as auxiliary parameters, and the generated control data are processed through a fitting control network;
the obstacle feedback detection module is used for configuring an obstacle avoidance control node interval based on the path planning, and executing obstacle feedback detection of the interaction sensor in the obstacle avoidance control node interval;
the control data optimization module is used for generating real-time compensation data according to the obstacle feedback detection result, optimizing the control data by the real-time compensation data and generating optimized control parameters;
and the automatic cleaning control module is used for carrying out automatic cleaning control on the building curtain wall through the optimized control parameters.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311303309.3A CN117031986B (en) | 2023-10-10 | 2023-10-10 | Control method and system for automatic cleaning of building curtain wall |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311303309.3A CN117031986B (en) | 2023-10-10 | 2023-10-10 | Control method and system for automatic cleaning of building curtain wall |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117031986A CN117031986A (en) | 2023-11-10 |
| CN117031986B true CN117031986B (en) | 2023-12-15 |
Family
ID=88626717
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311303309.3A Active CN117031986B (en) | 2023-10-10 | 2023-10-10 | Control method and system for automatic cleaning of building curtain wall |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117031986B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117284721B (en) * | 2023-11-23 | 2024-02-06 | 张家港市华申工业橡塑制品有限公司 | Intelligent decontamination method and system for rubber-plastic conveyor belt |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107440635A (en) * | 2017-09-29 | 2017-12-08 | 成都意町工业产品设计有限公司 | A kind of building glass-wall cleaning robot cleaning path planning system |
| CN109144067A (en) * | 2018-09-17 | 2019-01-04 | 长安大学 | A kind of Intelligent cleaning robot and its paths planning method |
| CN112057004A (en) * | 2020-09-16 | 2020-12-11 | 中原动力智能机器人有限公司 | Curtain wall cleaning robot obstacle identification method and control system |
| CN114895304A (en) * | 2022-07-13 | 2022-08-12 | 之江实验室 | Method, equipment and device for cleaning, detecting and planning building outer wall |
| CN115344034A (en) * | 2022-03-17 | 2022-11-15 | 西安达升科技股份有限公司 | Intelligent cleaning robot path planning method and intelligent cleaning device |
| CN115981323A (en) * | 2022-12-27 | 2023-04-18 | 广东智科电子股份有限公司 | Automatic obstacle avoidance method for multi-sensor fusion intelligent cleaning vehicle and intelligent cleaning vehicle |
-
2023
- 2023-10-10 CN CN202311303309.3A patent/CN117031986B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107440635A (en) * | 2017-09-29 | 2017-12-08 | 成都意町工业产品设计有限公司 | A kind of building glass-wall cleaning robot cleaning path planning system |
| CN109144067A (en) * | 2018-09-17 | 2019-01-04 | 长安大学 | A kind of Intelligent cleaning robot and its paths planning method |
| CN112057004A (en) * | 2020-09-16 | 2020-12-11 | 中原动力智能机器人有限公司 | Curtain wall cleaning robot obstacle identification method and control system |
| CN115344034A (en) * | 2022-03-17 | 2022-11-15 | 西安达升科技股份有限公司 | Intelligent cleaning robot path planning method and intelligent cleaning device |
| CN114895304A (en) * | 2022-07-13 | 2022-08-12 | 之江实验室 | Method, equipment and device for cleaning, detecting and planning building outer wall |
| CN115981323A (en) * | 2022-12-27 | 2023-04-18 | 广东智科电子股份有限公司 | Automatic obstacle avoidance method for multi-sensor fusion intelligent cleaning vehicle and intelligent cleaning vehicle |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117031986A (en) | 2023-11-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117031986B (en) | Control method and system for automatic cleaning of building curtain wall | |
| CN112099493B (en) | Autonomous mobile robot trajectory planning method, system and equipment | |
| CN112656311B (en) | Mobile robot zone cleaning | |
| CN110622085A (en) | Mobile robot and control method and control system thereof | |
| CN107981790A (en) | Room area division methods and sweeping robot | |
| CN103099583A (en) | Robot cleaner and control method thereof | |
| CN112363513B (en) | Obstacle classification obstacle avoidance control method based on depth information | |
| JP6240595B2 (en) | Self-position estimation apparatus and mobile body equipped with self-position estimation apparatus | |
| US20210213619A1 (en) | Robot and control method therefor | |
| Hess et al. | A probabilistic approach to high-confidence cleaning guarantees for low-cost cleaning robots | |
| CN111168669B (en) | Robot control method, robot, and readable storage medium | |
| CN112987725A (en) | Obstacle-based avoidance method, chip and cleaning robot | |
| CN114063615B (en) | Reversing navigation control method and system of intelligent pesticide spraying vehicle among ridges in greenhouse | |
| Pan et al. | GPU accelerated real-time traversability mapping | |
| CN112034837A (en) | Method for determining working environment of mobile robot, control system and storage medium | |
| JP7221549B2 (en) | Information processing device and mobile robot | |
| CN109753071B (en) | Robot welting walking method and system | |
| KR20190101326A (en) | Method for dividing moving space and moving robot for moving divided moving space | |
| US20200293051A1 (en) | Route determination method | |
| JP6438354B2 (en) | Self-position estimation apparatus and mobile body equipped with self-position estimation apparatus | |
| CN112308033B (en) | Obstacle collision warning method based on depth data and visual chip | |
| US20220100197A1 (en) | Self-propelled device and method for controlling the same | |
| CN115755888A (en) | AGV obstacle detection system with multi-sensor data fusion and obstacle avoidance method | |
| CN112947426A (en) | Cleaning robot motion control system and method based on multi-sensing fusion | |
| US11669096B2 (en) | Route determination method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |