Intelligent garbage collection device

Abstract

The invention discloses an intelligent garbage collection device, and belongs to the technical field of garbage collection devices. An intelligent garbage collecting device comprises a main body frame and further comprises four independent box doors and an electric throwing door which are arranged on the main body frame and correspond to four garbage recycling boxes of kitchen garbage, recyclable garbage, other garbage and harmful garbage respectively. An electric sealing cover is arranged at the top of each garbage recycling bin; the two sets of independently operating moving mechanisms are arranged on side plates of the main body frame; the two common trays are respectively driven by the two sets of moving mechanisms; the central visual identification unit is arranged above the top plate of the main body frame and covers the areas of the four garbage recycling bins; the control system of the integrated edge computing unit is fixed in a protection box at the top of the main body frame and is connected with each execution component through a CAN (Controller Area Network) bus; according to the invention, through central visual identification and mixed garbage grading disposal protocols, the error throwing rate is obviously reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of garbage collection devices, and in particular to an intelligent garbage collection device. BACKGROUND

[0002] With the acceleration of urbanization, the production of household garbage continues to rise. China produces nearly 300 million tons of municipal household garbage every year (data source: China Urban-Rural Construction Statistical Yearbook 2023). Although the garbage classification policy is fully implemented, the actual implementation effect is still not ideal: Low classification accuracy: the average accuracy rate in residential areas is less than 40%, and the misplacement rate in schools / public places is more than 50% (Urban Household Garbage Management Blue Book 2024); Low efficiency of cleaning and transportation: traditional garbage cans rely on manual inspection, and 33% of the cleaning and transportation vehicles run empty or half-loaded (Ministry of Housing and Urban-Rural Development 2023 survey); Mixed garbage disposal is difficult: mixed garbage containing harmful substances such as batteries accounts for 12%, and existing equipment cannot automatically sort, leading to frequent environmental pollution and safety accidents.

[0003] The current mainstream solutions have significant defects: Intelligent classification barrel: guided by image recognition, but cannot correct misplaced garbage, and mixed garbage directly enters the "other garbage" category, increasing the burden of backend sorting; Overflow monitoring system: based on ultrasonic sensor warning, but lacks predictive ability and still relies on fixed frequency cleaning and transportation; Hazardous substance identification device: sound and light alarm when detecting batteries, but no forced isolation mechanism, users continue to mix and throw after ignoring.

[0004] Especially in high-traffic scenarios such as universities and business districts, the above problems are more prominent: Students mix electronic garbage with batteries into kitchen waste, causing heavy metal pollution of leachate; The janitor needs to spend an additional 2.5 hours per day manually sorting misplaced garbage (Tsinghua University property survey report).

[0005] Therefore, it is urgent to develop a garbage collection device that integrates intelligent correction, mixed garbage classification disposal, and accurate prediction and scheduling, to fundamentally reduce manual intervention costs and improve classification efficiency and safety. SUMMARY

[0006] The present application relates to the technical field of garbage collection devices, and in particular to an intelligent garbage collection device.

[0007] In order to achieve the above purpose, the present application adopts the following technical solutions: The utility model provides an intelligent garbage collection device, including main body frame, still include: four independent box doors and electric delivery door are correspondingly arranged on the main body frame four garbage recycling boxes such as kitchen garbage, recyclable, other and harmful garbage, every garbage recycling box top is equipped with electric sealing cover, two sets of independent operation's mobile mechanism sets up on the main body frame side plate, two common trays are driven by two sets of mobile mechanism, central visual identification unit is arranged above the main body frame top plate, covers four garbage recycling box area, the control system of integrated edge calculation unit is fixed in the main body frame top protection box, is connected each executive component through CAN bus.

[0008] Preferably, the main body frame is provided with a guide plate corresponding to the common tray, a partition plate is arranged between every two adjacent garbage recycling boxes, and a reinforcing plate is arranged on the main body frame.

[0009] Preferably, the outer surface of the electric delivery door is provided with a classification mark and a touch switch, and the electric delivery door is turned inwards by greater than or equal to 90 degrees when the touch switch is turned on.

[0010] Preferably, a built-in prediction scheduling engine is used to predict the filling rate based on historical data and real-time sensor data, and the function is as follows: , Wherein, Predicted filling rate in future t hours, Human flow density weight coefficient, Real-time human flow density (from camera), Weather influence weight coefficient, Weather factor (rain and snow = 1.5, sunny = 1.0), Historical period weight coefficient, Same period data 24 hours ago, Random disturbance term; The prediction scheduling engine outputs a cleaning task sequence and pushes a navigation path to a cleaner terminal according to the urgency.

[0011] Preferably, the odor monitoring unit is a metal oxide semiconductor gas sensor installed on the top of each electric sealing cover, which can detect the concentration of H2S and NH3 in real time. The capacity monitoring unit is a 3DToF sensor installed on the top of each electric sealing cover, which emits laser to the surface of the garbage and calculates the filling height.

[0012] Preferably, when the gas sensor detects that the concentration of H2S is greater than 15ppm, the negative pressure deodorization system is automatically started. When the 3DToF sensor detects that the filling rate is greater than 95%, the electric sealing cover of the corresponding garbage can is locked.

[0013] Preferably, after the user puts the garbage into any common tray, the central visual recognition unit starts recognition; After the recognition is completed, the control system drives the common tray to move to the delivery position above the target garbage recycling bin, and the remaining garbage recycling bins remain stationary and sealed; After the garbage is delivered, the electric sealing cover on the target garbage recycling bin is reset and closed.

[0014] Preferably, the mixed garbage disposal protocol is: When it is recognized that the garbage contains hazardous garbage mixed with other garbage: a) Start a voice warning: "Please put the hazardous garbage into the hazardous garbage bin"; b) If no hazardous garbage bin delivery action is detected within 120 seconds: Control the current common tray to move to the coordinate (X0, Y0) directly above the hazardous garbage bin and lock; Send an emergency alarm containing the device ID to the management platform; c) The non-disposal tray continues to serve other users.

[0015] Preferably, if there is no hazardous garbage in the mixed garbage, the normal mixed distribution rule is enabled: If the kitchen waste ratio is > 60%, distribute to the kitchen waste bin; If the recyclable material ratio is > 50%, distribute to the recyclable bin; In other cases, distribute to the other garbage bin.

[0016] Preferably, a method for operating an intelligent garbage collection device, comprising the following steps: Step S1, the user opens any electric delivery door and places the garbage on the common tray; Step S2, the electric delivery door is closed, and the central visual unit recognizes the garbage type, which can be classified as pure garbage, hazardous mixed garbage, and ordinary mixed garbage, and different measures are taken accordingly; Step S3, the electric sealing cover on the garbage recycling bin is opened in advance, and the common tray moves to the delivery position on the target garbage recycling bin, and the remaining garbage recycling bins remain sealed; Step S4, after the garbage is delivered, the common tray is reset, and the electric sealing cover on the garbage recycling bin is closed.

[0017] Compared with the prior art, the present application provides an intelligent garbage collection device, which has the following beneficial effects: 1. The intelligent garbage collection device, through the cooperation of the central visual recognition unit and the two sets of moving mechanisms, realizes dynamic garbage distribution: after the user places the garbage arbitrarily, the system automatically recognizes and accurately transfers it to the corresponding recycling bin, improving the classification accuracy.

[0018] 2、The intelligent garbage collection device optimizes the cleaning and transportation efficiency through the cooperation of the prediction scheduling engine and the capacity and odor sensor: through the prediction filling rate, combined with real-time monitoring by the 3DToF sensor, a cleaning and transportation task sequence is generated to reduce empty vehicle trips.

[0019] 3、The intelligent garbage collection device forcibly isolates the pollution source through the cooperation of the mixed garbage disposal protocol and the moving mechanism: when detecting harmful mixtures, the shared tray is locked above the harmful box, and the fireproof liner is started synchronously to solve the problem of harmful mixed garbage disposal.

[0020] The parts not involved in the intelligent garbage collection device are the same as or can be realized by the prior art, and the present application significantly reduces the misplacement rate through central visual recognition and mixed garbage classification disposal protocol. BRIEF DESCRIPTION OF DRAWINGS

[0021] Figure 1 A structure diagram of the overall structure of the intelligent garbage collection device proposed by the present application Figure One ; Figure 2 A structure diagram of the overall structure of the intelligent garbage collection device proposed by the present application Figure Two ; Figure 3 A structure diagram of the internal structure of the intelligent garbage collection device proposed by the present application Figure One ; Figure 4 A structure diagram of the internal structure of the intelligent garbage collection device proposed by the present application Figure Two ; Figure 5 A structure diagram of the internal structure of the intelligent garbage collection device proposed by the present application Figure Three ; Figure 6 A flowchart of the intelligent garbage collection device proposed by the present application Figure One ; Figure 7 A flowchart of the intelligent garbage collection device proposed by the present application Figure Two .

[0022] In the figure: 1, main frame; 101, box door; 102, electric feeding door; 2, garbage collection box; 3, moving mechanism; 4, shared tray; 5, central visual recognition unit; 6, electric sealing cover; 7, guide plate; 8, reinforcing plate; 9, partition plate. DETAILED DESCRIPTION

[0023] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all.

[0024] In the description of the application, it should be understood that the terms "upper", "lower", "front", "back", "left", "right", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the application.

[0025] Reference Figures 1-7 , including the main frame 1, further comprising: four independent box doors 101 and an electric throwing door 102 are arranged on the main frame 1, which correspond to four garbage recycling boxes 2 of kitchen waste, recyclable, other and hazardous waste; each garbage recycling box 2 is provided with an electric sealing cover 6 at the top; two sets of independently operated moving mechanisms 3 are arranged on the side plate of the main frame 1; two shared trays 4 are driven by the two sets of moving mechanisms 3; a central visual recognition unit 5 is arranged above the top plate of the main frame 1, covering the area of the four garbage recycling boxes 2; a control system integrated with an edge calculation unit is fixed in the top protection box of the main frame 1, and each executing component is connected through a CAN bus.

[0026] In the application, the main frame 1 is welded and formed by 304 stainless steel, the size is 1.5m*0.8m*1.2m (length* width*height), and the surface is sprayed with an anti-corrosion coating. Four garbage recycling boxes 2 are injection molded by high-density polyethylene material, with a volume of 120L / box, and are designed to be detachable for easy cleaning. The specific structural features include: Moving mechanism 3: Adopting a screw rod mechanism (attached Figure 3 and attached Figure 4 ), The driving motor is a 57-step motor (torque 1.2N*m), The transmission accuracy is ±0.1mm, The maximum moving speed is 2m / s (recyclable) and 0.5m / s (kitchen waste splash-proof); The central visual recognition unit 5 is configured with: 2000 pixel RGB camera (Sony IMX586) Near-infrared spectrum sensor (wavelength 900-1700nm) NVIDIA Jetson Orin NX edge calculation module.

[0027] Control system integration: Main control chip: STM32H743VI (ARM Cortex-M7), Communication protocol: CAN 2.0B (baud rate 500kbps), Protection level: IP65 (dustproof and waterproof), Power supply system: 220V AC main power supply, 48V / 20Ah iron lithium backup battery, 100W solar panel (attached Figure 1 Top); In one embodiment, the garbage classification identification is sprayed or pasted on the electric drop door 102.

[0028] Referring to Figures 1-7 , the main frame 1 is provided with a guide plate 7 corresponding to the shared tray 4, a partition plate 9 is arranged between each adjacent two garbage recycling bins 2, and the main frame 1 is provided with a reinforcing plate 8; The outer surface of the electric drop door 101 is provided with a classification identification and a touch switch, which is turned inwards by ≥90° when opened through a hinge mechanism.

[0029] In the present application, after the electric drop door 102 is opened, the user drops garbage into the inside of the collecting device, the guide plate 7 can guide the garbage to flow onto the shared tray 4; The reinforcing plate 8 cooperates with the threaded block on the lead screw to serve as a limiting component of the threaded block, so that the shared tray 4 can be moved, and the reinforcing plate 8 enhances the overall strength of the device.

[0030] Referring to Figures 1-7 , a built-in prediction scheduling engine predicts the filling rate based on historical data and real-time sensor data, and the function is: , wherein, — future t-hour predicted filling rate, — human flow density weight coefficient, — real-time human flow density from a camera, — weather influence weight coefficient, Weather factor Rain and snow = 1.5, Sunny = 1.0, — historical period weight coefficient, — same period data 24 hours ago, — random disturbance term; The prediction scheduling engine outputs a cleaning task sequence and pushes a navigation path to the cleaner terminal according to the urgency.

[0031] In the filling rate prediction function, the weight coefficients α, β and γ are dynamically updated through historical data regression analysis, and the specific steps are as follows: (a) Collect filling rate, human flow density, weather and historical rate data for 30 consecutive days; (b) Solve the optimal coefficient matrix [α, β, γ]^T with the objective of minimizing the mean square error; (c) retrain the coefficients once every 7 days to adapt to the changing activity cycle of the campus or other living space.

[0032] In the present invention, data collection: Real-time crowd density P(t): Counted by overhead wide-angle camera, Weather factor W(t): Access to central weather station API (rain and snow = 1.5), Historical data H(t-24): Local storage of 30-day time series data; In one embodiment, a feasible assumption scheme is provided: The built-in prediction scheduling engine predicts that the No. 2 kitchen waste bin will be full in 1.5 hours, so it pushes the cleaning task to the APP:

Emergency

[0033] Referring to Figures 1-7 odor monitoring unit: each electric sealing cover 6 top mounted metal oxide semiconductor gas sensor, real-time detection of H2S, NH3 concentration; Capacity monitoring unit: 3DToF sensor is arranged on the top of each electric sealing cover 6, which emits laser to the surface of the garbage and calculates the filling height; When the gas sensor detects that the concentration of H2S is greater than 15ppm, the negative pressure deodorization system is automatically started; When the 3DToF sensor detects that the filling rate is greater than 95%, the electric sealing cover 6 corresponding to the garbage can is locked.

[0034] In the present invention, the negative pressure deodorization system is integrated on the top of the electric sealing cover 6, and the electric sealing cover 6 is installed on the main frame 1.

[0035] Referring to Figures 1-7 After the user puts the garbage into any shared tray 4, the central visual recognition unit 5 starts to recognize; After the recognition is completed, the control system drives the shared tray 4 to move to the delivery position above the target garbage recycling bin 2, and the remaining garbage recycling bins 2 remain stationary and sealed; After the garbage is delivered, the electric sealing cover 6 on the target garbage recycling bin 2 is reset and closed.

[0036] Mixed garbage disposal protocol: When it is identified that the garbage contains hazardous waste mixed with other garbage: a) Start voice warning: "Please put the hazardous waste into the hazardous waste bin"; b) If no hazardous waste bin delivery action is detected within 120 seconds: Control the current common tray 4 to move to the coordinate X0, Y0 directly above the hazardous waste bin and lock; Send an emergency alarm containing the device ID to the management platform; c) Non-disposal tray continues to serve other users.

[0037] If there is no hazardous waste in the mixed garbage, enable the normal mixed garbage distribution rule: If the kitchen waste ratio is > 60%, distribute it to the kitchen waste bin; If the recyclable waste ratio is > 50%, distribute it to the recyclable waste bin; In other cases, distribute it to other garbage bins.

[0038] In the present application, the mixed garbage disposal process is Scenario: A student throws a toy car containing a button cell into the recyclable tray Disposal process, Recognition stage: The central vision unit 5 recognizes a mixture of metal (battery) + plastic (toy car) The system determines that it is hazardous mixed garbage (confidence 92%) User warning: Voice broadcast: "Hazardous waste detected! Please remove the battery and put it in the red hazardous bin" The screen outside the electric drop door 102 displays an animation guide Timeout disposal (after 120 seconds): If the user has handled it (the hazardous bin is open, and the hazardous bin has been disposed of), the device operates normally; if the user has not handled it, the common tray 4 containing the hazardous mixed garbage is moved to the top of the hazardous waste bin and locked, and the janitor is notified; Janitor handling: Unlock the common tray 4 by swiping the card (NFC authentication) Remove the battery and put it in the hazardous bin (trigger the weight sensor verification) The remaining plastic car is disposed of in the recyclable bin according to the pure garbage rule.

[0039] In one embodiment, the hazardous waste recycling bin is made of fireproof metal material, with a temperature sensor and a fire-retardant gel sprayer built-in. When the temperature is detected to be > 70℃, the fire-retardant gel is automatically sprayed.

[0040] Referring to Figures 1-7 A method for operating an intelligent garbage collection device, comprising the following steps: Step S1, the user opens any electric drop door 102 and places the garbage on the common tray 4; Step S2, the electrically operated door 102 is closed, the central vision unit 5 identifies the type of garbage, and the identification result can be classified as pure garbage, hazardous mixed garbage, and ordinary mixed garbage, and different measures are taken accordingly; Step S3, the electrically operated sealing cover 6 on the garbage recycling bin 2 is opened in advance, the shared tray 4 is moved to the delivery position on the target garbage recycling bin 2, and the remaining garbage recycling bins 2 remain sealed; Step S4, after the garbage is delivered, the shared tray 4 is reset, and the electrically operated sealing cover 6 on the garbage recycling bin 2 is closed.

[0041] In the present application, more specifically, step S1: the user delivers The user touches the touch switch of the electrically operated door 102, and the door is opened by ≥90° through the hinge; Place the garbage in any idle shared tray 4 (double tray supports parallel operation); After delivery, the door is automatically closed, triggering the infrared sensor to start the system.

[0042] Step S2: intelligent identification and classification decision The central vision unit 5 starts multi-modal identification: The RGB camera scans the shape of the garbage (such as plastic bottles, batteries); Near-infrared spectrum analyzes the material composition (such as organic matter, metal); Classification determination logic: Pure garbage → directly matched to the corresponding garbage bin (kitchen waste / recyclable / other / hazardous) Hazardous mixed garbage → voice warning → 120-second countdown → if not handled, lock the tray and alarm Ordinary mixed garbage → (kitchen waste ratio > 60% → kitchen waste bin; recyclable > 50% → recyclable bin) Step S3: precise delivery and sealing control Dynamic linkage mechanism: The electrically operated sealing cover 6 of the target garbage recycling bin 2 is opened in advance (the covers of the remaining bins remain sealed to prevent odor spread); The moving mechanism 3 drives the shared tray 4 to move along the guide plate 7 to the top of the target bin (positioning accuracy ±0.1mm); The tray tilting mechanism is started (a motor is provided on the connecting rod of the shared tray 4 and the moving mechanism 3), and the garbage slides into the target bin (the kitchen waste bin slows down to 0.5m / s to prevent splashing).

[0043] Step S4: reset and abnormality handling After delivery: The shared tray 4 returns to the initial position standby; The electrically operated sealing cover 6 of the target bin is closed; Abnormal scenario disposal: Full box lock: if 3DToF detects fill rate > 95%, the box cover is locked and an alarm is reported. Hazard retention: after the tray is locked above the hazardous box, only the janitor's NFC is authorized to unlock.

[0044] The above description is only the preferred embodiment of the present application, but the protection scope of the present application is not limited to this. Any person skilled in the art can make equivalent replacements or changes within the technical range disclosed by the present application according to the technical solution and the inventive concept of the present application, which should be covered within the protection scope of the present application.

Claims

1. An intelligent garbage collection device comprising a main frame (1), characterized in that, Also includes: The main frame (1) is provided with four independent box doors (101) and an electrically operated drop door (102) corresponding to four garbage recycling bins (2) of kitchen waste, recyclable, other and hazardous waste, etc. Each of the garbage recycling bins (2) is provided with an electrically operated sealing cover (6) at the top; Two sets of independently operated moving mechanisms (3) are arranged on the side plates of the main frame (1); Two shared trays (4) are driven by the two sets of moving mechanisms (3); A central visual recognition unit (5) is arranged above the top plate of the main frame (1) and covers the area of the four garbage recycling bins (2); A control system integrated with an edge computing unit is fixed in a protection box at the top of the main frame (1) and connected to each executing component through a CAN bus.

2. The intelligent garbage collection device according to claim 1, characterized in that, The main frame (1) is provided with a guide plate (7) corresponding to the shared tray (4), and a partition plate (9) is arranged between each adjacent two garbage recycling bins (2), and the main frame (1) is provided with a reinforcing plate (8).

3. The intelligent garbage collection device according to claim 2, characterized in that, The outer surface of the electrically operated drop door (101) is provided with classification marks and a touch switch, which is turned inwards by ≥90° through a hinge mechanism when opened.

4. The intelligent garbage collection device according to claim 1, wherein, Built-in prediction scheduling engine, based on historical data and real-time sensor data to predict fill rate, function is: , wherein, — future t hours prediction fill rate, — human flow density weight coefficient, — real-time human flow density (from camera), — weather influence weight coefficient, weather factor (rain / snow = 1.5, sunny = 1.0), — historical period weight coefficient, — same period data 24 hours ago, — random disturbance term; The prediction scheduling engine outputs a cleaning task sequence and pushes a navigation path to the cleaner terminal according to the urgency.

5. The intelligent garbage collection device according to claim 1, wherein, Odor monitoring unit: A metal oxide semiconductor gas sensor is installed at the top of each electrically operated sealing cover (6) to detect the concentrations of H2S and NH3 in real time. Capacity monitoring unit: A 3DToF sensor is arranged at the top of each electrically operated sealing cover (6) to emit laser to the surface of the garbage and calculate the filling height.

6. The intelligent garbage collection device according to claim 5, wherein, When the gas sensor detects that the concentration of H2S is >15ppm, the negative pressure deodorization system is automatically started. When the 3DToF sensor detects that the filling rate is >95%, the electrically operated sealing cover (6) of the corresponding garbage bin is locked.

7. The intelligent garbage collection device according to claim 1, wherein, After the user drops garbage into any shared tray (4), the central visual recognition unit (5) starts to recognize. After the recognition is completed, the control system drives the shared tray (4) to move to the drop position above the target garbage recycling bin (2), and the remaining garbage recycling bins (2) remain stationary and sealed. After the garbage is delivered, the electrically operated sealing cover (6) on the target garbage recycling bin (2) is reset and closed.

8. The intelligent garbage collection device according to claim 1, wherein, Mixed garbage disposal protocol: When it is recognized that the garbage contains mixed hazardous waste and other garbage: a) Start voice warning: "Please put the hazardous waste into the hazardous waste bin"; b) If no hazardous waste bin delivery action is detected within 120 seconds: Control the current shared tray (4) to move to the hazardous waste bin directly above the coordinate (X0, Y0) and lock it; Send an emergency alarm containing the device ID to the management platform; c) The non-disposal tray continues to serve other users.

9. The intelligent garbage collection device according to claim 8, wherein, If there is no hazardous waste in the mixed garbage, the general mixed garbage distribution rule is enabled: If the proportion of kitchen waste is >60%, it is distributed to the kitchen waste bin; If the proportion of recyclable waste is >50%, it is distributed to the recyclable waste bin; In other cases, it is distributed to the other garbage bin.

10. A method of operating an apparatus as claimed in any one of claims 1 to 9, characterised by, The method comprises the following steps: Step S1, the user opens any electrically operated drop door (102) and places garbage on the shared tray (4); Step S2, the electrically operated door (102) is closed, the central vision unit (5) identifies the garbage type, the identification result can be classified as pure garbage, harmful mixed garbage and ordinary mixed garbage, and different measures are taken accordingly; Step S3, the electrically operated sealing cover (6) on the garbage recycling bin (2) is opened in advance, the shared tray (4) is moved to the target garbage recycling bin (2) on the delivery position, and the remaining garbage recycling bins (2) remain sealed; Step S4, after the garbage is delivered, the shared tray (4) is reset, and the electrically operated sealing cover (6) on the garbage recycling bin (2) is closed.

Images