AI scene monitoring and early warning system for operation site

Abstract

The invention relates to an AI scene monitoring and early warning system for a working site, which is applied to the field of monitoring and early warning, and is characterized in that the scene of the working site and a working vehicle are photographed and monitored through a monitoring module, an abnormity analysis module judges whether early warning is needed or not according to an abnormal condition, and the analysis content mainly comprises abnormal obstacle distinguishing and early warning. According to the method, whether a scene obstacle is an abnormal person or not is distinguished, unsafe factors on the scene and unsafe behaviors of a vehicle are automatically recognized, early warning reminding of different levels and different types is achieved, an AI learning module summarizes the analysis processing result of an abnormity analysis module, the judgment capacity of an early warning system is continuously optimized, the abnormal state recognition accuracy is effectively improved, and the early warning efficiency is improved. And according to the processing operation modes made by most of the drivers, the corresponding AI assistance processing suggestions are obtained, and effective help is provided for the subsequent drivers encountering the abnormal problems, so that the processing efficiency of the drivers on the abnormal conditions is effectively improved.

Description

Technical Field

[0001] An AI scene monitoring and warning system involved in the present invention, in particular, an AI scene monitoring and warning system for the operation site applied to the field of monitoring and warning. Background Art

[0002] Video surveillance technology is a technology that uses video cameras and related devices to monitor a specific area. It is widely used in many fields such as security monitoring, traffic management, retail, healthcare, education, and industry. In some special working environments, it is necessary to accurately monitor the working environment and working equipment, and identify various behaviors in the working scene. According to the urgency and difficulty of the requirements, they are divided into three priorities: high, medium, and low. Timely handle potential safety anomalies that have not occurred.

[0003] The specification of Chinese invention patent CN202311498641.X discloses "A multi-scene network video monitoring and warning system". Through scene monitoring, abnormal frame pictures are confirmed, and then it is analyzed whether they exceed the standard. If there is an over-standard situation, an alarm is directly issued. If not, it is analyzed whether the corresponding abnormal area covers the border. If it does, it means that some areas may be located in other scenes. Subsequently, the monitoring pictures of adjacent monitoring terminals are confirmed, and then the corresponding pictures to be spliced are confirmed. The pictures to be spliced are merged with the corresponding scenes. Subsequently, the central analysis plane after splicing is confirmed, and the central analysis plane is comprehensively analyzed and processed with the corresponding abnormal area to determine whether it exceeds the standard, so as to confirm whether the warning condition can be triggered. Regarding several pictures as a whole for analysis, combining two groups of pictures to confirm the central analysis plane improves the accuracy of the determination.

[0004] Most traditional video surveillance technologies only process the content of a single video, and cannot summarize the processing process of this abnormal situation, which affects the recognition accuracy of the system. The AI learning model refers to algorithms and computational models that use artificial intelligence technology to simulate the human learning process. AI models can be used for various tasks, including image recognition, speech recognition, natural language processing, recommendation systems, etc. Existing video monitoring and warning systems lack this learning function, which affects the recognition accuracy of the system. Moreover, traditional monitoring systems can only alarm through voice and lights, but lack preventive protection for the operating vehicles in the working scene. When unqualified personnel sit on the operating vehicle, a single warning cannot effectively handle the situation, and there is a lack of a substantial way to drive them away. Summary of the Invention

[0005] Aiming at the above-mentioned existing technologies, the technical problem to be solved by the present invention is that most traditional video surveillance technologies only process the content of a single video, and cannot summarize the processing process of this abnormal situation, which affects the recognition accuracy of the system. A single warning cannot effectively handle the situation, and there is a lack of a substantial way to drive them away.

[0006] To solve the above problems, the present invention provides an AI scenario monitoring and early warning system for the operation site, including a central control server. The connection end of the central control server is connected to a monitoring module, an early warning module, and a mobile terminal module. The mobile terminal module is used for the management personnel to receive and process alarm information. The early warning module includes a scenario model module and an alarm module;

[0007] The monitoring module includes a scenario camera module and a vehicle-mounted camera module. Both the scenario camera module and the vehicle-mounted camera module use cameras and are respectively fixedly installed on the operation site and the operation vehicle. The scenario camera module monitors the operation scenario environment in real time, especially the scenario with operation vehicles. The vehicle-mounted camera module detects the driving environment of the operation vehicle and the operation status of the driver in real time;

[0008] The scenario model module includes an anomaly analysis module and a model supplement module. The connection end of the scenario model module is connected to an AI learning module. The connection end of the anomaly analysis module is connected to a behavior record module, a driving record module, and a scenario record module. The scenario model module is used to generate a model for the data collected by the monitoring module. The anomaly analysis module is used to analyze the abnormal models that appear in the newly formed scenario model and judge whether early warning is needed for the abnormal situation. The analysis content mainly includes distinguishing abnormal obstacles and determining whether the scenario obstacles are abnormal personnel. Then, the early warning module gives an early warning to the driver. The model supplement module is used to generate a new model for the newly appeared monitoring situation and add it to the scenario model module;

[0009] The AI learning module includes an AI interaction module. The AI learning module is used to summarize the analysis and processing results of the anomaly analysis module and the operation methods of the drivers of the operation vehicles to handle such problems, and obtain handling suggestions for dealing with such abnormal problems, so as to provide early warning suggestions for drivers who encounter such problems in the future. The AI interaction module is used to interact with the drivers of the operation vehicles, record the drivers' instructions, and give the recommended handling methods based on such early warnings to the drivers. The AI learning module summarizes the analysis and processing results of the anomaly analysis module, effectively improves the accuracy of abnormal state recognition by continuously optimizing the judgment ability of the early warning system, is more accurate in identifying key abnormal contents, and records the operation methods of the drivers to handle such problems, providing AI assistance and handling suggestions for drivers who encounter such problems in the future.

[0010] In the above-mentioned AI scenario monitoring and early warning system for the operation site, the scenario of the operation site and the operation vehicle are monitored by the monitoring module through camera, automatically identify the unsafe factors on the site and the unsafe behaviors of the operation vehicle, and realize early warning reminders of different levels and types. Especially for the alarm and expulsion of personnel without driving qualifications breaking into the operation vehicle, the operation safety effect is effectively improved.

[0011] As a further improvement of the present application, the AI learning module further includes a record update module, which is used to sort the abnormal situation content collected by the AI and the AI assistance processing suggestions according to time, and the newly added abnormal situation content and the newly generated AI assistance processing suggestions have a higher priority. The abnormal situation content that does not appear again is updated and replaced, and the new processing operation methods made by the driver for the old abnormal situation content generate new AI assistance processing suggestions to replace the old AI assistance processing suggestions.

[0012] As a further improvement of the present application, the alarm module includes a voice alarm unit and a light alarm unit, and the alarm module is also connected to a driver's seat evacuation module, which is used to automatically alarm and evacuate abnormal personnel entering the cab of the work vehicle and prevent abnormal personnel from staying in the cab for a long time.

[0013] As a further improvement of the present application, the driver's seat evacuation module includes a cab and a seat in the cab. The seat is a foldable seat, and a storage box is fixedly connected to the back of the seat. A preventive cover plate is slidably connected to the outside of the storage box. The front of the preventive cover plate correspondingly covers the steering wheel of the cab, and a power cylinder is hinged between the bottom of the preventive cover plate and the bottom of the cab.

[0014] As a further improvement of the present application, a cover groove is opened in the middle of the front of the preventive cover plate, and a locking tooth is fixedly connected to the inside of the cover groove. The locking tooth corresponds to the steering wheel of the cab. The steering wheel of the cab is covered by the cover groove and locked by the locking tooth, effectively preventing unqualified personnel from operating the steering wheel.

[0015] As a further improvement of the present application, drive-away support arms are hinged at both ends of the back of the preventive cover plate, and retractable piles are fixedly connected to both ends of the top of the storage box. The end of the drive-away support arm away from the hinged part contacts the retractable pile, and a torsion spring is sleeved on the hinged part of the drive-away support arm and the preventive cover plate. When the preventive cover plate extends out of the storage box, the drive-away support arms extend outwards, thus occupying the left and right spaces of the steering wheel in the cab, further preventing unqualified personnel from operating the steering wheel and driving unqualified personnel out of the cab. Extension pages are fixedly connected to one ends of the two drive-away support arms facing each other, and the two groups of extension pages are parallel to each other up and down. The extension pages are used to increase the space occupation area of the drive-away support arms and effectively improve the drive-away effect on unqualified personnel.

[0016] In summary, the present invention monitors the scene of the operation site and the operation vehicle through a monitoring module, generates a data model using the scene model module for the monitoring results, and uses an anomaly analysis module to judge the anomalies in the model, automatically identifying unsafe factors at the scene and unsafe behaviors of the operation vehicle, and realizing different-level and different-type warning reminders. The AI learning module summarizes the analysis and processing results of the anomaly analysis module to provide a judgment basis for subsequent warnings. By continuously optimizing the judgment ability of the warning system, the accuracy of anomaly state recognition is effectively improved, the recognition of key anomaly contents is more accurate, and assistance and treatment suggestions are given, effectively improving the driver's processing efficiency for abnormal conditions. When a person without a driving qualification breaks into the operation vehicle, while the alarm module issues an alarm, the driver's seat drive-away module synchronously conducts the drive-away work. The power cylinder pushes the bottom of the prevention cover plate, the seat in the cab flips and folds, and at the same time, the prevention cover plate is pushed out of the storage box, and finally the prevention cover plate covers the steering wheel of the cab, horizontally occupying the space of the cab, driving the person without a driving qualification out of the cab, and preventing the person with a driving qualification from accidentally operating the steering wheel. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 System diagram of the first embodiment of the present application;

[0018] Figure 2 System diagram of the monitoring module of the first embodiment of the present application;

[0019] Figure 3 System diagram of the anomaly analysis module of the first embodiment of the present application;

[0020] Figure 4 System diagram of the alarm module of the first embodiment of the present application;

[0021] Figure 5 Demonstration diagram of the driver's operation of avoiding obstacles in the first embodiment of the present application;

[0022] Figure 6 Demonstration diagram of the driver driving away abnormal personnel in the scene in the first embodiment of the present application;

[0023] Figure 7 Working demonstration diagram of the driver's seat drive-away module in the second embodiment of the present application;

[0024] Figure 8 Three-dimensional structure diagram of the cover groove in the second embodiment of the present application;

[0025] Figure 9 Rear-view three-dimensional diagram of the seat in the second embodiment of the present application;

[0026] Figure 10This is a demonstration diagram of the seat in the folded state of the second embodiment of the present application;

[0027] Figure 11 This is a three-dimensional structure diagram of the driving-away support arm of the second embodiment of the present application;

[0028] Figure 12 This is a demonstration diagram of the rotational change of the driving-away support arm of the second embodiment of the present application;

[0029] Figure 13 This is a three-dimensional structure diagram of the extension page of the second embodiment of the present application.

[0030] Explanation of the reference numerals in the figure:

[0031] 101, cab; 102, seat; 103, storage box; 104, preventive cover; 105, power cylinder; 106, cover groove; 107, engaging tooth; 108, telescopic column; 2, driving-away support arm; 201, retracting pile; 202, torsion spring; 203, extension page; 204, tribulus terrestris nail; 205, rubber head. Specific embodiments

[0032] The following will describe the two embodiments of the present application in detail with reference to the accompanying drawings.

[0033] The first embodiment:

[0034] Figures 1 to 4 As shown, an AI scene monitoring and early warning system for the operation site includes a central control server. The connection end of the central control server is connected with a monitoring module, an early warning module, and a mobile terminal module. The mobile terminal module is used for the management personnel to receive and process alarm information. The early warning module includes a scene model module and an alarm module. The monitoring module includes a scene camera module and a vehicle-mounted camera module. Both the scene camera module and the vehicle-mounted camera module use cameras and are respectively fixedly installed on the operation site and the operation vehicle. The scene model module includes an abnormal analysis module, a model supplement module, and an AI learning module. The connection end of the abnormal analysis module is connected with a behavior record module, a driving record module, and a scene record module. The AI learning module includes an AI interaction module. The scene model module is used for generating a model based on the data collected by the monitoring module. The abnormal analysis module is used for analyzing the abnormal models in the newly formed scene model and judging whether early warning is needed for the abnormal situation. The model supplement module is used for generating a new model based on the newly appeared monitoring situation and adding it to the scene model module. The AI learning module is used for summarizing the analysis and processing results of the abnormal analysis module. The AI interaction module is used for interacting with the driver of the operation vehicle, recording the driver's instructions. The alarm module includes a voice alarm unit and a light alarm unit;

[0035] When using the early warning system, both the scene camera module and the vehicle-mounted camera module use cameras, which are respectively fixedly installed at the operation site and on the operation vehicle. The operation vehicles monitored specifically include forklifts, tractors, bending plate vehicles, and other machinery applied to the work scene. The scene camera module monitors the operation scene environment in real time, and the vehicle-mounted camera module detects the driving environment of the operation vehicle and the operation status of the driver in real time. The monitoring results are used by the scene model module to generate a data model, and the anomaly analysis module is used to judge the abnormal situations in the model, automatically identify the unsafe factors on the site and the unsafe behaviors of the operation vehicle, and then use the alarm module to give voice alarms and light alarms;

[0036] Figure 5 and Figure 6 As shown, when the scene camera module detects an abnormality in the working environment, specifically when there is an obstacle on the driving route of the operation vehicle, after detecting the obstacle, first judge the obstacle to determine what the obstacle is, specifically whether it is a person, an operation vehicle, placed goods, or other sundries. First, alarm the driver on this route to prevent the driver from failing to observe the obstacle in time. When it is judged that the obstacle on the driving route of the operation vehicle is not a living being, that is, not a living being, after the driver avoids it according to the obstacle, the vehicle-mounted camera module monitors the driver's operation to judge whether the avoidance operation is difficult. An easy avoidance operation can reduce the alarm level of this abnormality, and a difficult avoidance operation can increase the alarm level of this abnormality, and send an alarm message to the mobile terminal module to facilitate the management personnel to quickly solve the problem. When it is judged that the obstacle on the driving route of the operation vehicle is a living being, that is, an abnormal person staying on the route, while the alarm module gives a warning alarm to the driver, it also alarms and drives away the abnormal person staying. The AI learning module records and learns the handling operations made by the driver of the operation vehicle this time. After multiple drivers encounter this type of problem and make the same operations, the AI interaction module generates corresponding AI assistance suggestions to give AI assistance suggestions for this type of problem when subsequent drivers encounter the same problem, facilitating the driver to make handling operations in time, Figure 5 In Figure 5 , what blocks the forklift route is a cargo box. For this type of problem, the driver usually adopts a detour method to handle it. At this time, the suggestion given by the AI will also be to suggest the driver to take a detour, Figure 6 In Figure 6 , what blocks the forklift route is an abnormal person squatting. The driver usually adopts a method of driving away the abnormal person to handle it. At this time, the suggestion given by the AI will also be for the driver to drive away the obstructive abnormal person by himself. The AI learning module also includes a record update module, which is used to sort the abnormal situation content and AI assistance handling suggestions collected by the AI according to time. When a driver operates the operation vehicle to come to Figure 5Before the scene shown, the AI gives the driver a suggestion to take a detour in advance, and the newly added abnormal situation content and the newly generated AI assistance handling suggestions have a higher priority. The abnormal situation content that does not appear again is updated and replaced, and the new handling operation methods made by the driver for the old abnormal situation content generate new AI assistance handling suggestions to replace the old ones;

[0037] The monitoring content of the vehicle-mounted camera module is the driver's operation behavior and the driving situation of the operating vehicle. In particular, it monitors whether the abnormal personnel entering the operating vehicle meet the driving qualifications. The specific identification method is face recognition. The driver with driving qualifications pre-records their face information into the system. When personnel without driving qualifications break into the operating vehicle, the voice alarm unit and the light alarm unit of the alarm module give an on-site alarm and send an alarm message to the mobile terminal module to facilitate the management personnel to quickly solve the problem.

[0038] The second implementation method:

[0039] Compared with the first implementation method, a driver's seat ejection module is mainly added. The specific added structure is as follows, and the rest of the structure is the same as that of the first implementation method.

[0040] Figures 7 to 9 As shown, the alarm module is also connected to a driver's seat ejection module. The driver's seat ejection module includes a cab 101 and a seat 102 inside the cab 101. The seat 102 is a foldable seat, and a storage box 103 is fixedly connected to the back of the seat 102. A preventive cover 104 is slidably connected to the outside of the storage box 103. The front of the preventive cover 104 correspondingly covers the steering wheel of the cab 101, and a power cylinder 105 is hinged between the bottom of the preventive cover 104 and the bottom of the cab 101. A cover groove 106 is opened in the middle of the front of the preventive cover 104, and a tooth 107 is fixedly connected to the inside of the cover groove 106. The tooth 107 corresponds to the steering wheel of the cab 101. The steering wheel of the cab 101 is covered by using the cover groove 106 and locked by the tooth 107, effectively preventing personnel without driving qualifications from operating the steering wheel. A telescopic column 108 is movably inserted into the tooth 107, and a telescopic spring is fixedly connected to the root of the telescopic column 108. The length of the tooth 107 is increased by the telescopic column 108, thereby effectively improving the locking effect on the steering wheel;

[0041] When a person without driving qualifications breaks into the operating vehicle, that is, when an abnormal person enters the cab, while the warning system uses the alarm module to give an alarm, the driver's seat displacement module also displaces the abnormal person. During the displacement operation, the power cylinder 105 pushes the bottom of the preventive cover plate 104, causing the seat 102 in the cab 101 to flip and fold. The preventive cover plate 104 slides and extends in the storage box 103, and finally the preventive cover plate 104 covers the steering wheel of the cab 101. The cover groove 106 covers the steering wheel of the cab 101 and locks the steering wheel through the locking teeth 107, horizontally occupying the space of the cab 101. That is, a person without driving qualifications cannot sit up normally in the cab and cannot operate the steering wheel in the cab, so as to displace the person without driving qualifications from the cab 101, effectively avoiding safety accidents caused by the misoperation of the person without driving qualifications.

[0042] Figures 10 to 12 As shown, drive-away support arms 2 are hinged at both ends of the back surface of the preventive cover plate 104. Closing piles 201 are fixedly connected to both ends of the top of the storage box 103. The end of the drive-away support arm 2 away from the hinge part contacts the closing pile 201, and a torsion spring 202 is sleeved on the hinge part of the drive-away support arm 2 and the preventive cover plate 104. When the preventive cover plate 104 extends out of the storage box 103, the drive-away support arm 2 extends outwards, thus occupying the left and right spaces of the steering wheel in the cab 101, further preventing a person without driving qualifications from operating the steering wheel, and displacing the person without driving qualifications from the cab 101. Fixed connection extension pages 203 are provided at the opposite ends of the two drive-away support arms 2. The two groups of extension pages 203 are parallel to each other up and down. The extension pages 203 are used to increase the space occupation area of the drive-away support arm 2, effectively improving the displacement effect on a person without driving qualifications;

[0043] When the preventive cover plate 104 is pushed out of the storage box 103, the drive-away support arm 2 gradually extends out of the closing pile 201, and the closing constraint position of the closing pile 201 on the drive-away support arm 2 gradually moves towards the hinge part of the drive-away support arm 2. The drive-away support arm 2 automatically turns outwards and unfolds by using the elastic force of the torsion spring 202, and increases the space occupation area through the extension pages 203, that is, occupies the left and right spaces of the cab 101, and further squeezes the person without driving qualifications out of the cab 101.

[0044] Figures 10 to 13It is shown that a tribulus nail 204 is fixedly connected to the middle of the extension page 203. The tribulus nail 204 penetrates the upper and lower surfaces of the extension page 203. The tribulus nail 204 is used to restrict the behavior of personnel without driving qualifications by the extension page 203, effectively preventing personnel without driving qualifications from continuing to operate the operating equipment of the cab 101. The tribulus nails 204 are annularly and equidistantly distributed about the hinge part of the driving-away support arm 2. The tribulus nails 204 on two groups of parallel corresponding extension pages 203 are arranged in a front-back staggered manner. A rubber head 205 is fixedly connected to the tip of the tribulus nail 204. Through the special arrangement of the tribulus nails 204, it is effectively avoided that the tribulus nails 204 affect the mutual unfolding of two parallel extension pages 203, so as to facilitate stacking longer extension pages 203 on the back of the prevention cover plate 104;

[0045] Tribulus nails 204 are arranged on both the upper and lower surfaces of the extension page 203 to enhance the restriction and driving-away effect of the upper and lower spaces of the extension page 203. During the process of the driving-away support arm 2 unfolding outwards, the tribulus nails 204 arranged in a ring on the extension page 203 are staggered back and forth during rotation, so as to avoid the tribulus nails 204 affecting the rotation operation of the extension page 203. At the same time, longer tribulus nails 204 can be arranged in the narrow space between the two groups of extension pages 203, so as to improve the deterrent effect of the tribulus nails 204 on personnel without driving qualifications and prevent personnel without driving qualifications from staying in the cab 101 forcefully.

[0046] Combined with the current actual requirements, the above-mentioned implementation manner adopted in this application, the protection scope is not limited to this. Within the scope of knowledge possessed by those skilled in the art, various changes made without departing from the concept of this application still fall within the protection scope of the present invention.

Claims

1. An AI scene monitoring and early warning system for work sites, characterized by: It includes a central control server, wherein the connection end of the central control server is connected to a monitoring module, an early warning module and a mobile terminal module, the mobile terminal module is used for management personnel to receive and process alarm information, and the early warning module includes a scene model module and an alarm module; The monitoring module includes a scene camera module and a vehicle camera module. Both the scene camera module and the vehicle camera module use cameras and are fixedly installed at the work site and the work vehicle respectively. The scene camera module monitors the work scene environment in real time, especially the scene with the work vehicle. The vehicle camera module detects the travel environment of the work vehicle and the operating status of the driver in real time. The scenario model module includes an abnormal analysis module and a model supplement module. The connection end of the scenario model module is connected to an AI learning module. The connection end of the abnormal analysis module is connected to a behavior recording module, a driving recording module and a scenario recording module. The scenario model module is used to generate a model for the data collected by the monitoring module. The abnormal analysis module is used to analyze the abnormal model that appears in the newly constructed scenario model, and to determine whether an early warning is required for the abnormal situation, to distinguish abnormal obstacles, and to distinguish whether the scene obstacles are abnormal personnel. Then the early warning module issues an early warning to the driver. The model supplement module is used to generate a new model for the newly appeared monitoring situation and add it to the scenario model module. The AI ​​learning module includes an AI interaction module, which is used to summarize the analysis and processing results of the abnormal analysis module, as well as the operating methods of the operating vehicle driver in dealing with such problems, and to obtain processing suggestions for dealing with such abnormal problems. The AI ​​interaction module is used to interact and communicate with the driver of the operating vehicle, collect the driver's instructions, and give the driver recommended processing methods based on such early warnings. The AI ​​learning module summarizes the analysis and processing results of the abnormal analysis module, and collects the driver's operating methods for dealing with such problems, and provides AI assistance and processing suggestions for drivers who encounter such problems in the future.

2. According to claim 1, the AI ​​scene monitoring and early warning system for the work site is characterized by: The AI ​​learning module also includes a record update module, which is used to sort the abnormal situation content and AI assistance processing suggestions recorded by AI according to time, and the newly added abnormal situation content and the newly generated AI assistance processing suggestions have a higher priority. The abnormal situation content that has not appeared again is updated and replaced, and the new processing operations made by the driver on the old abnormal situation content for many times generate new AI assistance processing suggestions to replace the old AI assistance processing suggestions.

3. According to claim 1, the AI ​​scene monitoring and early warning system for the work site is characterized in that: The alarm module includes a voice alarm unit and a light alarm unit, and the alarm module is also connected to a driver's seat expulsion module, which is used to automatically alarm and expel abnormal personnel entering the cab, thereby preventing abnormal personnel from staying in the cab for a long time.

4. The AI ​​scene monitoring and early warning system for a work site according to claim 3 is characterized in that: The driving position driving away module comprises a cab (101) and a seat (102) in the cab (101); the seat (102) is a foldable seat, and a storage box (103) is fixedly connected to the back of the seat (102); a preventive cover plate (104) is slidably connected to the outside of the storage box (103); the front of the preventive cover plate (104) corresponds to covering the steering wheel of the cab (101); and a power cylinder (105) is hinged between the bottom of the preventive cover plate (104) and the bottom of the cab (101).

5. The AI ​​scene monitoring and early warning system for a work site according to claim 4 is characterized in that: A cover groove (106) is provided in the middle of the front side of the preventive cover plate (104), and a latching tooth (107) is fixedly connected inside the cover groove (106), and the latching tooth (107) corresponds to the steering wheel of the cab (101).

6. The AI ​​scene monitoring and early warning system for a work site according to claim 5 is characterized in that: Both ends of the back side of the preventive cover (104) are hinged with a drive-away bracket arm (2), and both ends of the top of the storage box (103) are fixedly connected with a folding pile (201), and the end of the drive-away bracket arm (2) away from the hinged part is in contact with the folding pile (201), and the hinged part between the drive-away bracket arm (2) and the preventive cover (104) is provided with a torsion spring (202), and the ends of the two drive-away bracket arms (2) facing each other are fixedly connected with an extension page (203), and the two groups of the extension pages (203) correspond to each other in parallel up and down.

Images