Method and device for automatically identifying sweeping brush loss and automatically replacing sweeping brush of sweeping vehicle

Abstract

The invention relates to a method and a device for automatically identifying sweeper loss and automatically replacing a sweeper by a sweeper. The method generates an intermediate layer result by receiving sensor data representative of sweep loss and inputting it to a neural network. And triggering the classifier to generate a classifier score according to the intermediate layer result, and judging whether the sweeper needs to be replaced or not based on the score. When the score exceeds a preset threshold value, the controller sends out a signal to start the turnover motor, and the loosening and locking operation of the sweeper is automatically completed. The automatic sweeping device is suitable for unmanned sweeping vehicles and similar automatic sweeping equipment, has the advantages of improving the working efficiency, reducing the maintenance cost, improving the safety, prolonging the service life of the equipment and the like, and can realize the automation of sweeping brush replacement, reduce manual intervention, improve the sweeping efficiency and safety and reduce the operation cost.

Description

Technical Field

[0001] The present invention relates to the field of automated cleaning equipment for unmanned sweepers, and in particular to a method and device for automatically identifying brush wear and tear and automatically replacing the brushes of a sweeper. Background Art

[0002] In recent years, with the rapid development of automation technology and robotics technology, some unmanned sweepers have emerged in the field of cleaning to replace some manual cleaning operations. They use a variety of sensors (such as lidar, cameras, millimeter-wave radar, etc.) to perceive the surrounding environment, combined with high-precision maps and advanced algorithms, to achieve autonomous navigation, obstacle avoidance and path planning of vehicles. Applying unmanned driving technology to the field of sweepers overcomes the limitations of traditional sweepers and enables sweepers to automatically, efficiently and safely perform cleaning tasks in complex urban environments.

[0003] During the maintenance of the unmanned sweeper, the space around the sweeping disc is often cramped due to the body structure design of the unmanned sweeper. When replacing the sweeping disc, the maintenance personnel have a serious lack of operating space, their arms are difficult to stretch freely, and the use of tools is also greatly restricted, which makes it difficult to remove and install the sweeping disc smoothly, increasing the difficulty and time cost of operation; the connection parts between the sweeping disc and the unmanned sweeper are numerous and complex. It may involve multiple bolts and nuts. These connectors are often prone to rust and corrosion due to long-term exposure to complex working environments (such as moisture, dust and garbage), making the disassembly process extremely difficult. In addition, some sweeping discs are also connected to the vehicle's transmission or hydraulic system through special connectors. These connection methods require professional knowledge and specific tools for safe disassembly, which increases the complexity of replacement; the sweeping disc itself usually has a certain weight. Without suitable auxiliary equipment, it is difficult for maintenance personnel to carry and operate the sweeping disc for replacement alone. If multiple people collaborate, it is necessary to ensure the synchronization and coordination of the operation, otherwise the sweeping disc may accidentally fall or collide during the replacement process, causing damage to personnel safety and the sweeping disc itself.

[0004] Therefore the prior art still needs to be improved and enhanced. Summary of the invention

[0005] In view of the above-mentioned shortcomings of the prior art, the purpose of the present invention is to provide a method and device for a sweeper to automatically identify sweeping brush wear and replace the sweeping brush by itself. By adopting intelligent sensors and neural network technology, the wear state of the sweeping brush can be monitored in real time, and the sweeping brush replacement operation can be automatically performed according to preset trigger conditions, thereby improving the working efficiency and safety of unmanned sweepers.

[0006] In order to solve the above technical problems, the present invention adopts the following technical solutions: In a first aspect, the present invention provides a method for a sweeper to automatically identify brush wear and tear and automatically replace the brush, the method comprising the following steps: receiving sensor data indicative of brush wear; Inputting the sensor data into a neural network and generating intermediate layer results; Based on the intermediate layer results and the preset trigger conditions, applying a trigger classifier to classify the intermediate layer results to generate a classifier score; Based on the classifier score, it is determined whether to perform a brush replacement operation.

[0007] Furthermore, the trigger classifier is trained by a convolutional neural network, and the intermediate layer results come from the penultimate layer of the convolutional neural network.

[0008] Furthermore, the classifier score is compared with one or more preset thresholds to determine whether to perform a brush replacement operation.

[0009] Furthermore, the trigger classifier further includes a threshold filter for determining whether to send a brush replacement signal according to the size of the classifier score.

[0010] Furthermore, the trigger classifier performs classification based on different types of brush wear characteristics to automatically adjust replacement timing.

[0011] In a second aspect, the present invention further provides a device for a sweeper to automatically identify brush wear and tear and automatically replace the brush, the device comprising: Inclination sensor to monitor brush wear; a controller, receiving feedback data from the tilt sensor, and applying a trigger classifier based on a trigger condition to generate a classifier score; The motor is flipped to automatically perform the release and lock operations of the sweeping brush based on the classifier score.

[0012] Preferably, the controller further comprises an energy storage device for providing electrical energy to the tilt sensor and the flip motor.

[0013] Preferably, the control signal of the flip motor is generated by the controller according to the classifier score to automatically complete the loosening and fixing of the sweeping brush.

[0014] Preferably, the device also includes a radar module and an image detection module for assisting in locating worn brushes and determining the placement of new brushes.

[0015] Preferably, the device further comprises a communication module for transmitting the brush replacement data and its related metadata to the remote control system via a network.

[0016] The technical solution of the present invention realizes automatic identification of brush wear through the combination of intelligent sensors and neural networks, and ensures automatic replacement of the brush at the right time by triggering the application of classifiers. Compared with the prior art, the present invention has the following beneficial effects: Improve work efficiency. Traditional unmanned sweepers require manual replacement of sweeping brushes regularly, which includes multiple steps such as parking, removing old sweeping brushes, and installing new sweeping brushes. Manual operation of these steps is usually time-consuming, and in large-scale cleaning tasks, frequent replacement of sweeping brushes will cause the sweeper to be shut down for a long time. The automatic replacement of sweeping brushes can automatically complete the replacement when the unmanned sweeper's power allows or between tasks, with almost no manual waiting time, greatly reducing the interruption time of cleaning work, thereby improving the overall cleaning efficiency.

[0017] Improve safety. When manually replacing the sweeping brush, the operator needs to operate around the vehicle, which may face some potential safety risks. For example, when replacing the sweeping brush on the roadside, there may be a collision threat from passing vehicles; when operating in the equipment cabin, there may also be injuries due to unexpected movement of machine parts. The automatic replacement of sweeping brushes by unmanned sweepers can keep personnel away from these dangerous operating environments, and complete the replacement of sweeping brushes through pre-designed mechanical devices and automated processes, thereby effectively avoiding the safety hazards caused by manual operation.

[0018] Reduce operating costs. From a long-term perspective, reducing the number of people involved in replacing sweeping brushes can save a lot of labor costs. Enterprises do not need to arrange special personnel to carry out sweeping brush replacement work, or they can allow existing personnel to devote more energy to other more valuable maintenance and management work, such as remote monitoring of equipment, data analysis, etc.

[0019] To extend the service life of unmanned sweeping vehicles, the automatic brush replacement system usually has a more precise installation and disassembly mechanism, which can better avoid damage to the brush installation parts due to improper operation compared to manual operation. Moreover, since the worn brushes can be replaced in time, there will be no additional burden on the motor and other parts of the sweeper due to excessive wear of the brushes, thereby extending the service life of the overall sweeper equipment, reducing the frequency of equipment updates and repairs, and reducing the long-term operating costs of the equipment.

[0020] The present invention is applicable to a variety of scenarios, has good practicability and economy, and provides an efficient solution for garbage disposal by unmanned sweepers. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 A flow chart of the method and steps of a method and device for automatically identifying brush wear and automatically replacing brushes for a sweeper provided by a preferred embodiment of the present invention; Figure 2A device structure diagram of a method and device for automatically identifying brush wear and automatically replacing brushes for a sweeper provided by a preferred embodiment of the present invention; Figure 3 A structural diagram of a device without a sweeping brush for a method and device for automatically identifying sweeping brush wear and automatically replacing sweeping brushes provided by a preferred embodiment of the present invention; Figure 4 A structural diagram of a sweeping brush bayonet for a method and device for automatically identifying brush wear and automatically replacing a sweeping brush on a sweeping vehicle provided in a preferred embodiment of the present invention; Figure 5 for Figure 3 Enlarged view of point A in the middle. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0023] It should be noted that when a component is referred to as being "mounted on", "fixed on" or "disposed on" another component, it may be directly on the other component or there may be a central component at the same time. When a component is referred to as being "connected to" another component, it may be directly connected to the other component or there may be a central component at the same time.

[0024] It should also be noted that the directional terms such as left, right, up, and down in the embodiments of the present invention are merely relative concepts or are based on the normal use status of the product and should not be considered as restrictive.

[0025] The present invention provides a method for a sweeper to automatically identify brush wear and tear and automatically replace the brush, the method comprising the following steps: S100, receiving sensor data indicating brush loss; S200, inputting the sensor data into a neural network and generating an intermediate layer result; S300: Based on the intermediate layer results and a preset trigger condition, applying a trigger classifier to classify the intermediate layer results to generate a classifier score; S400: Based on the classifier score, determine whether to perform a brush replacement operation.

[0026] Specifically, the cleaning vehicle of the present invention, preferably an unmanned cleaning vehicle, is equipped with multiple sensors, including a tilt sensor 1, a laser radar, a camera, etc. The tilt sensor 1 is installed on the sweeping brush 2 to detect the degree of wear of the sweeping brush 2 in real time. As the cleaning operation proceeds, the tilt sensor 1 continuously collects angle data of the sweeping brush 2, and the data is transmitted to the controller 3 on the cleaning vehicle. The controller 3 performs preliminary processing on the data of the tilt sensor 1 and sends the results to the neural network for further analysis.

[0027] Furthermore, the controller 3 inputs the data into the vehicle-mounted neural network via the wireless network. The neural network used in the present invention is a convolutional neural network (CNN), which is trained to efficiently process the data of the inclination sensor 1 and identify the pattern of brush wear.

[0028] Furthermore, the output of the neural network is a multi-layer intermediate result, and the controller 3 determines whether the preset trigger condition is met based on these results. If it is detected that the wear degree of the sweeping brush 2 exceeds a certain preset value, preferably 15° in the embodiment of the present invention, the system will automatically determine that it is in a "need to be replaced" state.

[0029] In a further embodiment, the present invention adopts an advanced classifier training method to make it more adaptable to different types of brush 2 wear conditions. The trigger classifier is a support vector machine (SVM) for determining whether the brush needs to be replaced based on the results of the neural network middle layer. The trigger classifier is pre-trained to effectively distinguish between brushes that need to be replaced and brushes that do not need to be replaced.

[0030] In this embodiment, the trigger classifier receives the intermediate result from the penultimate layer of the convolutional neural network. This layer outputs a large amount of data and contains more characteristic information of brush wear, which can improve the accuracy of the classifier.

[0031] Furthermore, the trigger classifier calculates a classifier score based on the received intermediate result. The higher the score, the more serious the brush wear is and the closer it is to the threshold for replacement.

[0032] If the classifier score exceeds a preset threshold (eg, 0.8), the controller will send a signal to execute brush replacement, automatically starting the brush replacement process.

[0033] If the score is below the threshold, the wear of the brush continues to be monitored until the wear reaches the threshold.

[0034] Furthermore, when the trigger classifier score exceeds a preset threshold, the controller 3 determines whether to perform a sweep brush replacement operation according to the trigger classifier score. If the score is higher than the threshold, the controller sends a signal to the flip motor 4 to start the loosening and locking operation of the sweep brush 1.

[0035] Another embodiment of the present invention further provides a device for a sweeper to automatically identify brush wear and tear and automatically replace the brush, the device comprising: Inclination sensor 1, used to monitor the wear of the sweeping brush; A controller 3 receives feedback data from the tilt sensor 1 and applies a trigger classifier based on a trigger condition to generate a classifier score; Specifically, the controller 3 has the following functions: 1. receiving the angle parameter fed back by the inclination sensor 1 and transmitting it to the host of the unmanned sweeper; 2. receiving the signal of the unmanned sweeper and providing the forward or reverse current signal to the flip motor 4; The flip motor 4 automatically performs the release and lock operations of the sweeping brush 2 based on the classifier score.

[0036] In actual operation, the control signal of the flip motor 4 is generated by the controller 3 according to the classifier score to automatically complete the loosening and fixing of the sweeping brush. The control signal of the flip motor 4 is sent by the controller 3, and the flip motor 4 performs the removal and installation of the sweeping brush 2. This process is fully automated and does not require human intervention.

[0037] More specifically, the flip motor 4 is connected to the sweeping disc bayonet 5 and can control the loosening and installation of the sweeping brush 2 through an electrical signal. When the sweeping brush 2 needs to be replaced, the flip motor 4 will first loosen the existing sweeping brush and remove it from the bayonet 5.

[0038] After receiving the current signal 3 from the controller, the flip motor 4 flips the sweeping disc clamping plate 7 through the gear 6 to lock and release the sweeping disc; There are two bayonet holes 5 on the sweeping disc. When the sweeping disc clamping plate is turned down, the sweeping disc can be fastened so that the sweeping disc meets the working state and completes the cleaning task.

[0039] Then, the controller 3 controls the unmanned sweeper to move to a new sweeping brush position through the system navigation and positioning functions, automatically puts the new sweeping brush into the bayonet 5 and locks it through the flip motor 4. This process is monitored in real time by sensors to ensure the accuracy and safety of the operation.

[0040] The device also includes a radar module and an image detection module to assist in locating worn brushes and determining where to place new brushes. To ensure that new brushes are properly installed, the system uses onboard cameras, lidar, and image detection modules for auxiliary positioning. Through these sensors, the driverless sweeper can accurately reach the new brush placement point and automatically install the new brush after arriving.

[0041] The device also includes a communication module for transmitting the brush replacement data and its related metadata to the remote control system through the network. When the brush replacement process is completed, all relevant data (including classifier scores, brush wear data, brush replacement time, etc.) will be uploaded to the remote server through the communication module; Through the cloud platform or dedicated server, operators can monitor the working status of the driverless sweeper and the wear of the sweeping brush in real time. The system can also record the specific time and location of each sweeping brush replacement as a reference for subsequent maintenance.

[0042] Based on the uploaded historical data, the remote control system can perform data analysis and optimize the brush replacement strategy and control parameters. For example, if a certain type of brush often wears out too quickly under certain conditions, the system can automatically adjust the replacement frequency or recommend a different type of brush.

[0043] The remote control system can transmit new algorithms and control strategies to the driverless sweeper through wireless updates (OTA). In this way, the control system of the driverless sweeper can automatically adapt to new working environments and sweeping brush replacement requirements without human intervention.

[0044] Furthermore, the controller 3 further includes an energy storage device 9 for storing the electric energy converted by the solar panel 8 to provide electric energy for the tilt sensor and the flip motor.

[0045] Furthermore, the present invention is also provided with a rotary motor 10 and a solar panel 8. The solar panel 8 absorbs solar energy to provide energy for the device; the rotary motor 10 provides power for the sweeping disc to complete the cleaning action.

[0046] The control system of the present invention intelligently schedules the timing of brush replacement by combining map data and the working progress of the sweeper. For example, when the sweeper completes the cleaning task of a section of the area, the system automatically calculates the brush wear of the remaining tasks and automatically replaces the brushes between tasks to avoid long downtime.

[0047] The system automatically adjusts the brush wear monitoring threshold according to different working environments (such as road type, climate conditions, etc.). For example, in a slippery or muddy environment, the brush may wear faster, and the system will automatically lower the threshold to trigger the classifier to ensure that the brush is replaced in time.

[0048] To facilitate understanding of the present invention, the present invention summarizes the specific implementation steps and principles as follows: When the sweeping brush is worn too much, the inclination sensor detects that the sweeping disc has too large an inclination angle, and sends a signal to the vehicle-mounted host computer through the controller installed on the sweeping disc to request the sweeping disc to be replaced.

[0049] After receiving the signal that the sweeping disc needs to be replaced, the on-board host uses the radar, image detection module and map module installed on the unmanned sweeper to achieve precise positioning.

[0050] When the driverless sweeper reaches the position where the worn sweeping disc is placed, the on-board host sends a signal to the controller, and the controller starts the flip motor to release the sweeping disc. After the sweeping disc is released, the driverless sweeper reaches the new sweeping disc position again through the radar, image detection module and map module. After detecting that it has reached the new sweeping disc placement point, the on-board host sends a signal to the controller to start the flip motor and tighten the sweeping disc.

[0051] The automatic replacement of the scanning disk is completed.

[0052] principle: The wear of the sweeping brush is monitored through the inclination sensor. In conjunction with the host, radar, image detection module and map module carried by the unmanned sweeping vehicle, the unmanned sweeping vehicle is controlled to place the worn sweeping brush at the old sweeping brush placement point and go to replace the new sweeping brush. The main principle is to control the flip motor through electrical signals to achieve the loosening and locking state of the sweeping brush.

[0053] In summary, the present invention can fully demonstrate its characteristics of automation, high efficiency and intelligence. From the detection of brush wear, data analysis, generation of control signals, to the final brush replacement and remote monitoring, the entire process can be executed efficiently and accurately.

[0054] It is understandable that those skilled in the art can make equivalent substitutions or changes based on the technical solution and inventive concept of the present invention, and all these changes or substitutions should fall within the protection scope of the claims attached to the present invention.

Claims

1. A method for a sweeper to automatically identify brush wear and tear and automatically replace the brush, characterized in that: The method comprises the following steps: receiving sensor data indicative of brush wear; Inputting the sensor data into a neural network and generating intermediate layer results; Based on the intermediate layer results and the preset trigger conditions, applying a trigger classifier to classify the intermediate layer results to generate a classifier score; Based on the classifier score, it is determined whether to perform a brush replacement operation.

2. The method according to claim 1, characterized in that The trigger classifier is trained by a convolutional neural network, and the intermediate layer results come from the penultimate layer of the convolutional neural network.

3. The method according to claim 1, characterized in that The classifier score is compared with one or more preset thresholds to determine whether to perform a brush replacement operation.

4. The method according to claim 1, characterized in that The trigger classifier further includes a threshold filter for determining whether to send a brush replacement signal according to the size of the classifier score.

5. The method according to claim 1, characterized in that The trigger classifier performs classification based on different types of brush wear characteristics to automatically adjust replacement timing.

6. The sweeper automatically identifies the brush wear and tear and automatically replaces the brush, characterized in that: The device comprises: Inclination sensor to monitor brush wear; a controller, receiving feedback data from the tilt sensor, and applying a trigger classifier based on a trigger condition to generate a classifier score; The motor is flipped to automatically perform the release and lock operations of the sweeping brush based on the classifier score.

7. The device according to claim 6, characterized in that The controller further comprises an energy storage device for providing electric energy to the tilt sensor and the flip motor.

8. The device according to claim 6, characterized in that The control signal of the flip motor is generated by the controller according to the classifier score to automatically complete the loosening and fixing of the sweeping brush.

9. The device according to claim 6, characterized in that The device also includes a radar module and an image detection module for assisting in locating worn brushes and determining where to place new brushes.

10. The device according to claim 6, characterized in that The device also includes a communication module for transmitting the brush replacement data and its related metadata to a remote control system via a network.

Images