Automatic central isolation belt garbage cleaning system based on machine vision

Abstract

The invention relates to the technical field of intelligent environmental sanitation equipment, and discloses a central isolation belt garbage automatic cleaning system based on machine vision. Comprising an autonomous moving main body comprising a containing space, wheels and a navigation module, a visual perception assembly, an intelligent flexible mechanical arm with a self-adaptive clamping jaw and a vacuum chuck, an intelligent classified storage assembly and a ventilation and heat dissipation assembly. According to the system, autonomous navigation is conducted along the central isolation belt based on machine vision, the position and type of garbage are recognized through visual perception, the mechanical arm grabs and sorts the garbage into the corresponding storage grids according to a strategy, and equipment stability is guaranteed through ventilation and heat dissipation. According to the automatic central isolation belt garbage cleaning system based on machine vision, all the assemblies are easy to maintain in design and high in protection performance, all the assemblies work cooperatively, full-process automatic operation of autonomous movement, garbage recognition, flexible grabbing and classified storage is achieved, and automatic garbage cleaning of a central isolation belt is effectively achieved.

Description

Technical Field

[0001] This invention relates to the field of intelligent sanitation equipment technology, and more specifically, to an automatic garbage cleaning system for central median strips based on machine vision. Background Technology

[0002] As an important component of road traffic safety facilities, the cleanliness of the central median strip directly affects the aesthetics of the road environment and driving safety. Currently, the cleaning of trash in the central median strip mainly relies on two methods: 1. Manual cleaning: Workers need to enter the central area of ​​the road with heavy traffic, bend over and pick up trash with tools, which is not only labor-intensive and inefficient (the average daily cleaning distance per person is less than 5 kilometers), but also faces a very high safety risk of being hit by passing vehicles, especially in high-speed and heavy traffic scenarios such as highways, where the safety hazards are more prominent; 2. Traditional mechanical assistance: In some scenarios, simple robotic arms are used in conjunction with manually operated cleaning equipment, but there are three major defects: First, it lacks accurate environmental perception capabilities, cannot effectively identify the type and location of trash, and is prone to accidentally grabbing non-trash targets such as guardrails and green plants; Second, the end effector of the robotic arm has poor adaptability, and the success rate of grabbing light and small trash (such as paper scraps and plastic bags) or irregular trash (such as beverage bottles and packaging bags) is less than 60%; Third, the equipment does not have a classification and storage function, and the trash needs to be sorted again afterward. Moreover, the internal power equipment (such as batteries and motors) lacks efficient heat dissipation design, which makes it prone to failure in high-temperature environments, affecting the continuous operation capability of the equipment.

[0003] Therefore, there is an urgent need for an automatic garbage cleaning system for central isolation zones with autonomous navigation, accurate identification, flexible grasping, intelligent classification and stable heat dissipation functions, in order to solve the technical pain points of existing cleaning methods such as "high safety risk, low efficiency, poor adaptability and insufficient stability". Summary of the Invention

[0004] In view of this, the present invention proposes an automatic garbage cleaning system for central isolation zones based on machine vision, which aims to solve the problems of the current technology, such as lack of autonomous navigation and risk avoidance capabilities, need for manual intervention or guidance, low garbage recognition accuracy, inability to distinguish garbage type and material and easy misidentification of non-garbage targets, poor adaptability of robotic arm grasping and difficulty in handling garbage of different shapes and weights, lack of immediate sorting and storage function, need for subsequent secondary processing, failure to meet environmental protection requirements, poor internal heat dissipation of equipment, lack of dustproof and waterproof design, and insufficient stability in complex outdoor environments.

[0005] This invention proposes an automatic garbage cleaning system for central median barriers based on machine vision, comprising: An autonomous mobile body has an internal storage space configured as a kinetic energy storage device; its exterior includes wheels, machine vision sensing components, and a rigid connecting seat. The wheels are configured with two at the front and two at the back of the autonomous mobile body, symmetrically connected from left to right. The machine vision sensing components are located on the sides of the autonomous mobile body. The rigid connecting seat is located below the autonomous mobile body. An intelligent flexible robotic arm is disposed on the side of the rigid connecting seat and connected to the autonomous moving body through the rigid connecting seat. The intelligent flexible robotic arm is configured to grasp garbage. An intelligent sorting and storage component is configured in the middle and rear of the autonomous mobile body for sorting and organizing. A ventilation and heat dissipation component is disposed above the intelligent sorting and storage component on the side and rear of the autonomous mobile body. The ventilation and heat dissipation component is connected to the autonomous mobile body and the accommodating space, respectively, and is configured to guide airflow into the accommodating space.

[0006] The radius of the wheel is greater than the length of the intelligent flexible robotic arm.

[0007] The machine vision perception component includes a high-resolution industrial camera, a depth camera, and an edge computing module running an improved recognition algorithm. The high-resolution industrial camera is used to acquire RGB images, the depth camera is used to acquire depth information, and the edge computing module is used to fuse the RGB images and depth information to identify the location, type, and material of waste, and avoid non-waste targets.

[0008] The rigid connecting seat is configured to connect to the autonomous moving body on the side. It has a hollow sleeve inside and is connected to the intelligent flexible robotic arm at the bottom via a connecting rod. The height of the hollow sleeve inside the rigid connecting seat is greater than the height of the connecting rod.

[0009] The intelligent flexible robotic arm is a degree-of-freedom collaborative robotic arm, with a flexible end effector consisting of an adaptive gripper and a vacuum suction cup at its end. The adaptive gripper is used to perform adaptive gripping of waste of different shapes, and the vacuum suction cup is used to perform adsorption operation on lightweight and small waste.

[0010] The intelligent sorting and categorizing component includes a multi-compartment sorting and storage bin, which is divided into recyclable waste bins and other waste bins according to the type of waste. The intelligent flexible robotic arm sorts the waste into the corresponding storage bins based on the recognition results of the machine vision perception component, and the height of the multi-compartment sorting and storage bin is lower than the length of the intelligent flexible robotic arm.

[0011] The ventilation and heat dissipation components include: An air intake channel is disposed on the side of the autonomous moving body and above the storage component, and is connected to the autonomous moving body and the internal accommodating space respectively. The air intake channel is configured to guide air into the accommodating space. An air outlet duct is disposed on the other side of the autonomous moving body. The air outlet duct is connected to the autonomous moving body and the internal accommodating space respectively. The air outlet duct is configured to guide air into the accommodating space. A dustproof and waterproof ventilation net is disposed outside the air inlet channel and the air outlet channel. The dustproof and waterproof ventilation net is configured to filter sand and water vapor in the air inside the air inlet channel. A temperature sensor is disposed on the inner side of the accommodating space inside the autonomous mobile body, and the temperature sensor is configured to monitor the temperature within the accommodating space.

[0012] The autonomous mobile entity uses lidar, GPS, and visual SLAM technologies to achieve autonomous navigation, obstacle avoidance, and lane-level positioning.

[0013] An application of a machine vision-based automated waste removal system for central median barriers includes the following steps: S1. The autonomous mobile platform navigates and moves along the central isolation zone, and the machine vision perception component collects image information of the central isolation zone area in real time. S2. The machine vision perception component processes the acquired image information, identifies the location, type, and material of the waste, and outputs three-dimensional coordinates and grasping strategy; S3. The intelligent flexible robotic arm plans its motion path according to the three-dimensional coordinates and the grasping strategy, and performs flexible grasping of the waste through the flexible end effector; S4. The intelligent flexible robotic arm sorts the captured waste into the corresponding storage bins of the intelligent sorting and storage unit.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention significantly improves safety performance, with fully automated operation that eliminates the need for manual entry into the central area of ​​the road, thus completely eliminating the traffic safety risks associated with manual cleaning. Workers only need to empty the garbage at the transfer point, improving safety and significantly reducing the safety risks and labor costs of cleaning the central median.

[0015] The machine vision technology based on the improved algorithm in this invention improves the accuracy of garbage identification and enables rapid identification and positioning of garbage, increasing efficiency by 3-5 times compared to manual cleaning. At the same time, lane-level positioning and obstacle avoidance functions ensure that the equipment moves accurately along the guardrail, leaving no area unattended. The multi-degree-of-freedom robotic arm, combined with a flexible end effector, adapts to the grasping needs of different types of garbage, resulting in high cleaning efficiency.

[0016] This invention integrates ventilation and heat dissipation, dustproof and waterproof protection designs, making it suitable for complex outdoor environments; navigation and obstacle avoidance technology ensures the stability of the movement process and ensures the continuous operation of the system.

[0017] This invention achieves both environmental protection and energy conservation. The intelligent sorting and storage component enables real-time waste sorting, reducing the cost of subsequent secondary sorting. The intelligent temperature control design of the ventilation and heat dissipation component reduces the failure rate of the equipment in high-temperature environments, while also reducing energy consumption. Attached Figure Description

[0018] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 This is a three-dimensional structural diagram of the present invention; Figure 2 This is a schematic diagram of the assembly of the intelligent flexible robotic arm and the rigid connecting seat of the present invention; Figure 3 This is a schematic diagram of the structure of the intelligent robotic arm of the present invention; Figure 4 This is a schematic diagram of the structure of the present invention, which includes a visual sensing component on the side near the central isolation zone. Figure 5 This is a cross-sectional structural diagram of the device of the present invention; Among them: 100. Autonomous moving main body, 101. Accommodation space, 102. Rigid connecting seat, 103. Wheel, 200. Mounting bracket, 300. Visual perception component, 400. Intelligent flexible robotic arm, 401. Adaptive gripper, 402. Vacuum suction cup, 403. Connecting rod, 500. Intelligent classification and storage component, 501. Storage box, 502. U-shaped card slot, 600. Ventilation and heat dissipation component, 601. Air inlet channel, 602. Dustproof and waterproof ventilation net, 603. Air outlet channel, 604. Temperature sensor, 700. Sensor combination, 701. LiDAR, 702. Positioning antenna, 703. Multifunctional sensor. Detailed Implementation

[0019] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0020] In the description of this application, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0021] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0022] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0023] Please see Figures 1-3 This invention provides a machine vision-based automatic garbage cleaning system for central median strips, comprising: The autonomous mobile main body 100 is made of high-strength lightweight alloy material to ensure stability and mobility during operation; The autonomous mobile body 100 has an internal accommodating space 101, in which a kinetic energy device is fixedly installed. The kinetic energy device includes a battery pack, a drive motor, and a control motherboard. The output shaft of the drive motor is connected to the transmission shaft of the wheel 103 through a linkage shaft. The cables are routed through metal corrugated pipes to avoid wear and provide power and command transmission support for the entire system. The autonomous mobile body 100 is symmetrically provided with wheel mounting seats on both the front and rear ends. The autonomous mobile body 100 is symmetrically connected to the left and right sides of the exterior, with two sets arranged at the front and rear, and connected to the wheel mounting seats. The wheels 103 are made of anti-slip rubber material, and the rims are made of aluminum alloy. They are connected to the drive shaft through deep groove ball bearings, and the shaft ends are limited by snap rings. The radius and height of the wheels 103 are strictly designed to be greater than the overall length of the intelligent flexible robotic arm 400 to avoid interference between the robotic arm and the ground or the edge of the isolation zone during the movement. A rigid connecting seat 102 is welded to the center of the bottom of the autonomous moving body 100, and a visual perception component 300 is fixedly installed on the side by bolts, preferably one set on each of the left and right sides. The autonomous mobile body 100 has a LiDAR 701 with a detection radius of 5m and an accuracy of ±2cm fixed to the top center with bolts. A GPS positioning antenna 702 is installed at the rear. A multi-functional sensor 703 is installed on each of the left and right sides (50cm away from the visual perception component 300). All navigation component cables are connected to the control motherboard to achieve data synchronization. The multi-functional sensor can be used to sense the distance to the central median or the distance to vehicles on the other side. It can also identify the edge line of the median, traffic signs, and vehicles. The LiDAR, GPS positioning module, and multi-functional sensor are integrated, and the data fusion of the three enables autonomous navigation, obstacle recognition and avoidance, and lane-level precise positioning, ensuring stable driving along the central median.

[0024] The autonomous mobile unit 100 has a compact overall structure, with a width adapted to the central median strip (usually 1-2m) and a height controlled within 1.2m to avoid interference with passing vehicles.

[0025] The rigid connecting seat 102 is integrally formed of stainless steel. Its side is fixedly connected to the bottom of the autonomous moving body 100 by high-strength bolts. Anti-slip washers are set on the connecting surface to enhance stability. The rigid connecting seat 102 is integrally milled from stainless steel and has an "L" shaped structure. The vertical part is a hollow sleeve and the horizontal part has 4 bolt holes. The rigid connecting seat 102 has a hollow sleeve structure inside, and the inner wall of the sleeve is provided with a lubricating coating. The lower part is detachably connected to the base of the intelligent flexible robotic arm 400 through the connecting rod 403. The height of the hollow sleeve is designed to be greater than the length of the connecting rod 403, so that the connecting rod 403 can be slightly adjusted along the axial direction of the sleeve to adapt to the garbage grabbing needs of different heights, while ensuring the rigidity of the robotic arm connection. The robotic arm base is welded with a connecting rod 403, which is inserted into the hollow sleeve of the rigid connecting seat. The sleeve side wall is provided with a top screw. Tightening the connecting rod allows for fine adjustment of the robotic arm height, adapting to garbage grabbing at different heights.

[0026] The visual perception component 300 includes a high-resolution industrial camera, a depth camera, and an edge computing module, all of which are integrated into a waterproof and dustproof housing. The housing is fixed to the outer side of the autonomous mobile body 100 near the central isolation zone via a bracket. The visual sensing component 300 is installed as follows: an L-shaped mounting bracket is provided on the side of the autonomous moving main body 100 near the central isolation zone. The visual sensing component 300 is fixed to the bracket with M6 bolts at a height of 1.2m. The lens is tilted at 15° towards the central isolation zone to ensure coverage of the isolation zone within a range of 1-3m. The tilted lens design of the visual sensing component 300 is also suitable for monitoring and sensing the pickup of the intelligent flexible robotic arm 400. The component cable is led out through a waterproof aviation plug, passes through a waterproof hole on the side of the main body, and connects to the control motherboard. The interface is sealed with sealant to prevent rainwater from seeping in.

[0027] The high-resolution industrial camera's lens is directed towards the central isolation zone area to acquire RGB color images in real time, with a resolution of no less than 1920×1080 pixels, ensuring clear capture of garbage details; the depth camera uses TOF technology to simultaneously acquire regional depth information and obtain the three-dimensional spatial location of the garbage. The edge computing module is equipped with an improved recognition algorithm that performs real-time fusion processing of RGB images and depth information to accurately identify the location, type (recyclable / other), and material (hard / lightweight) of waste. At the same time, it filters out non-waste targets such as vehicles and bollards, and outputs the three-dimensional coordinates of the waste and the corresponding grasping strategy.

[0028] The intelligent flexible robotic arm 400 adopts a degree-of-freedom collaborative robotic arm. The main body is made of carbon fiber and the surface is coated with an anti-corrosion coating. It has both rigidity and flexibility. It is detachably connected to the hollow sleeve of the rigid connecting seat 102 through the connecting rod 403. The robotic arm 400 cable is led out through the wire hole inside the connecting rod and connected to the control motherboard of the autonomous moving body to realize command transmission. The robotic arm 400 is equipped with a flexible end effector, which integrates an adaptive gripper 401 and a vacuum suction cup 402. The operating rod at the front end of the robotic arm 400 is a hollow sleeve, with its end connected to the vacuum suction cup 402. The adaptive gripper 401 uses flexible silicone finger pads and can adaptively adjust its opening and closing angle according to the size of the waste, with an opening and closing range of 0-15cm, enabling stable gripping of solid waste of different shapes such as beverage bottles and packaging bags. The vacuum suction cup 402 provides negative pressure through a micro vacuum pump, with an adsorption area of ​​5cm². 2 It is used to adsorb and grab lightweight, small pieces of waste such as paper scraps and foam fragments, preventing them from being missed.

[0029] The robotic arm has a built-in motion controller that can automatically plan the optimal motion path based on the three-dimensional coordinates and grasping strategy output by the vision sensing component 300. The motion repeatability positioning accuracy is ±0.1mm, ensuring that the grasping action is precise and smooth.

[0030] The autonomous mobile body 100 is provided with a U-shaped slot 502 at the rear. The intelligent sorting and storage component 500 is a sorting and storage box 501. The sorting and storage box 501 is embedded into the slot 502 through a slide rail. Positioning buckles are set at both ends of the slide rail to prevent the box from sliding when the vehicle is moving. It is also installed in the middle and rear of the autonomous mobile body 100 through the slide rail to facilitate subsequent garbage dumping and cleaning. The sorting and storage box 501 is made of high-strength plastic material. The sorting and storage bin 501 is divided into a recyclable waste compartment and a other waste compartment by a partition. The volume ratio of the two compartments is 1:2. The compartment opening is equipped with a guide ramp to facilitate the disposal of waste by the robotic arm. The overall height of the intelligent sorting and storage component 500 is designed to be lower than the length of the intelligent flexible robotic arm 400, ensuring that the end of the robotic arm 400 can smoothly extend into the slot to complete the placement, and the edge of the slot is designed with rounded corners to avoid scratching the robotic arm.

[0031] The ventilation and heat dissipation component 600 includes an air inlet channel 601, an air outlet channel 603, a dustproof and waterproof ventilation mesh 602, and a temperature sensor 604. The whole is an integrated design and is connected to the accommodating space 101 of the autonomous moving main body 100.

[0032] The air inlet channel 601 is a rectangular air duct, located on the left side of the autonomous moving body 100. The air duct has a built-in miniature cooling fan (with adjustable wind speed) to guide cooling air into the accommodating space 101. The air outlet channel 603 is located on the right side of the autonomous moving body 100, forming an oblique convection air duct with the air inlet channel 601 to exhaust hot air from the accommodating space 101.

[0033] Both the air inlet channel 601 and the air outlet channel 603 are fixedly installed with dustproof and waterproof ventilation nets 602. The ventilation net adopts a composite structure of stainless steel filter and waterproof membrane, with a filtration accuracy of 50μm, which can effectively block sand and water vapor from entering the accommodating space 101 and protect the internal electronic equipment.

[0034] The temperature sensor 604 is a patch sensor that is attached to the inner side of the accommodating space 101 to monitor the internal temperature in real time. The monitoring range is -20℃ to 80℃, and the temperature data is transmitted to the control motherboard. When the temperature exceeds 60℃, the cooling fan is automatically controlled to speed up to enhance heat dissipation.

[0035] The workflow of the machine vision-based automatic garbage cleaning system for central median barriers is as follows: 1. After the system is started, the autonomous moving body 100 moves at a constant speed (0.5-1m / s) along the central median strip according to the preset path or real-time navigation instructions. The visual perception component 300 is started simultaneously to collect RGB images and depth information of the central median strip area in real time.

[0036] 2. The edge computing module of the visual perception component 300 processes the collected information in milliseconds, identifies the three-dimensional coordinates, type and material of the waste, outputs a "gripping" strategy for hard, large pieces of waste, and outputs an "adsorption" strategy for light, small pieces of waste, and transmits the data to the controller of the intelligent flexible robotic arm 400.

[0037] 3. After receiving data, the intelligent flexible robotic arm 400 quickly plans a collision-free motion path and drives the joints to the garbage location: if it is a gripping strategy, the adaptive gripper 401 adjusts the opening and closing angle to clamp the garbage; if it is an adsorption strategy, the vacuum suction cup 402 starts negative pressure to adsorb the garbage, ensuring stable gripping.

[0038] 4. After the grabbing is completed, the intelligent flexible robotic arm 400 adjusts its movement posture and moves above the intelligent sorting and storage component 500. It places the waste into the corresponding storage box according to the type of waste. Then the robotic arm resets and the autonomous moving body 100 continues to move forward to enter the next round of waste cleaning cycle.

[0039] To make the technical solution of the present invention clearer, the structure and working process of the system are further described below with reference to specific embodiments: Example 1 1. Implementation scenario: The median strip of a main road in a city (1.2m wide and 5km long) contains garbage such as plastic bottles, paper scraps, and cigarette butts. The ambient temperature is 35℃ and the humidity is 60%.

[0040] 2. Equipment parameter settings Autonomous mobile unit 100 moving speed: 5km / h; Machine vision sensing component 300: recognition distance: 3-5m; Intelligent flexible robotic arm 400 clamping force: 0.5-3N (adjustable for the weight of urban road waste); The initial speed of the 600 ventilation and heat dissipation component fan is 1500 rpm.

[0041] 3. Work Process The autonomous mobile unit 100 moves along the isolation zone path, and the machine vision perception component 300 collects 20 pieces of trash (12 plastic bottles and 8 wads of paper) within 10 minutes. The edge computing module identifies plastic bottles as "recyclable waste" and generates a "gripping" strategy; it identifies paper scraps as "other waste" and generates an "adsorption" strategy. The intelligent flexible robotic arm can grip 400 pairs of plastic bottles (gripping force 1.5N) and use a vacuum suction cup to pick up paper scraps (vacuum degree -70kPa), with a 100% success rate in grasping them. 500 pieces of garbage were sorted into the corresponding collection bins. The cleaning of a 5-kilometer section of road took 60 minutes, which is 3 times more efficient than manual cleaning (which takes 5 hours). During operation, the temperature of the containment space reached a maximum of 38℃. The ventilation and heat dissipation components 600 automatically maintained the fan speed at 1500rpm, and the temperature remained stable at 35-38℃. The equipment operated without fault.

[0042] Example 2 1. Implementation Scenarios The median strip of a highway (0.8m wide and 10km long) contains trash such as aluminum cans and plastic bags. The ambient temperature is 40℃ and the wind speed is 3m / s.

[0043] 2. Equipment parameter settings Autonomous mobile unit 100 moving speed: 8km / h (prioritizing highway cleaning efficiency); Machine vision sensing component 300 recognition distance: 4-5m (to meet the needs of high-speed movement); Intelligent flexible robotic arm 400 clamping force: 2-5N (for heavier aluminum cans); Ventilation and heat dissipation component 600 fan initial speed: 2000rpm (preheating in high temperature environment).

[0044] 3. Work Process The autonomous mobile unit 100 achieves lane-level positioning (error ≤ 3cm) through GPS and LiDAR, moves stably along the median strip, and avoids guardrail posts (obstacle avoidance rate 100%). The machine vision sensing component 300 collected 35 pieces of trash (18 aluminum cans and 17 plastic bags) within 20 minutes, with an accuracy rate of 97% (only one small plastic bag was not identified due to obstruction). The robotic arm can grip 400 pairs of aluminum cans (gripping force 3N) and adsorb plastic bags (vacuum degree -80kPa), with a success rate of 94% (one plastic bag fell due to wind speed, but was successfully gripped on the second attempt). The 10-kilometer section of road was cleared in 75 minutes. After the 500-ton collection box was fully loaded, it automatically navigated to the service area transfer point. During operation, the temperature of the containment space reached a maximum of 42℃. The temperature sensor triggered the fan to speed up to 2200rpm, and the temperature dropped to 38℃ within 10 minutes. The equipment operated continuously without overheating.

[0045] In summary, the system of the present invention can achieve efficient, safe and stable cleaning of garbage in the central isolation zone in different scenarios, and has strong practicality and promotion value.

[0046] The above embodiments are merely preferred embodiments of the present invention and are not intended to limit the present invention. The scope of protection of the present invention is determined by the appended claims.

[0047] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A machine vision-based automatic garbage cleaning system for central median strips, characterized in that, include: An autonomous mobile body, which has an internal accommodating space configured as a kinetic energy storage device; Its exterior includes wheels, machine vision sensing components, and rigid connecting seats. The wheels are configured with two at the front and two at the back of the autonomous moving body, symmetrically connected from left to right. The machine vision sensing components are disposed on the sides of the autonomous moving body. The rigid connecting seats are disposed below the autonomous moving body of the main body. An intelligent flexible robotic arm is disposed on the side of the rigid connecting seat and connected to the autonomous moving body through the rigid connecting seat. The intelligent flexible robotic arm is configured to grasp garbage. An intelligent sorting and storage component is configured in the middle and rear of the autonomous mobile body for sorting and organizing. A ventilation and heat dissipation component is disposed above the intelligent sorting and storage component on the side and rear of the autonomous mobile body. The ventilation and heat dissipation component is connected to the autonomous mobile body and the accommodating space, respectively, and is configured to guide airflow into the accommodating space.

2. The automatic garbage cleaning system for a central median strip based on machine vision according to claim 1, characterized in that, The radius of the wheel is greater than the length of the intelligent flexible robotic arm.

3. The automatic garbage cleaning system for a central median strip based on machine vision according to claim 1, characterized in that, The machine vision perception component includes a high-resolution industrial camera, a depth camera, and an edge computing module running an improved recognition algorithm. The high-resolution industrial camera is used to acquire RGB images, the depth camera is used to acquire depth information, and the edge computing module is used to fuse the RGB images and depth information to identify the location, type, and material of waste, and avoid non-waste targets.

4. The automatic garbage cleaning system for a central median strip based on machine vision according to claim 1, characterized in that, The rigid connecting seat is configured to connect to the autonomous moving body on the side. It has a hollow sleeve inside and is connected to the intelligent flexible robotic arm at the bottom via a connecting rod. The height of the hollow sleeve inside the rigid connecting seat is greater than the height of the connecting rod.

5. The automatic garbage cleaning system for a central median strip based on machine vision according to claim 1, characterized in that, The intelligent flexible robotic arm is a degree-of-freedom collaborative robotic arm, with a flexible end effector consisting of an adaptive gripper and a vacuum suction cup at its end. The adaptive gripper is used to perform adaptive gripping of waste of different shapes, and the vacuum suction cup is used to perform adsorption operation on lightweight and small waste.

6. The automatic garbage cleaning system for a central median strip based on machine vision according to claim 1, characterized in that, The intelligent sorting and categorizing component includes a multi-compartment sorting and storage bin, which is divided into recyclable waste bins and other waste bins according to the type of waste. The intelligent flexible robotic arm sorts the waste into the corresponding storage bins based on the recognition results of the machine vision perception component, and the height of the multi-compartment sorting and storage bin is lower than the length of the intelligent flexible robotic arm.

7. The automatic garbage cleaning system for a central median strip based on machine vision according to claim 1, characterized in that, The ventilation and heat dissipation components include: An air intake channel is disposed on the side of the autonomous moving body and above the storage component, and is connected to the autonomous moving body and the internal accommodating space respectively. The air intake channel is configured to guide air into the accommodating space. An air outlet duct is disposed on the other side of the autonomous moving body. The air outlet duct is connected to the autonomous moving body and the internal accommodating space respectively. The air outlet duct is configured to guide air into the accommodating space. A dustproof and waterproof ventilation net is disposed outside the air inlet channel and the air outlet channel. The dustproof and waterproof ventilation net is configured to filter sand and water vapor in the air inside the air inlet channel. A temperature sensor is disposed on the inner side of the accommodating space inside the autonomous mobile body, and the temperature sensor is configured to monitor the temperature within the accommodating space.

8. The automatic garbage cleaning system for a central median strip based on machine vision according to claim 1, characterized in that, The autonomous mobile entity uses lidar, GPS, and visual SLAM technologies to achieve autonomous navigation, obstacle avoidance, and lane-level positioning.

9. An application of a machine vision-based automatic garbage cleaning system for central median strips according to claims 1-8, characterized in that, Includes the following steps: S1. The autonomous mobile platform navigates and moves along the central isolation zone, and the machine vision perception component collects image information of the central isolation zone area in real time. S2. The machine vision perception component processes the acquired image information, identifies the location, type, and material of the waste, and outputs three-dimensional coordinates and grasping strategy; S3. The intelligent flexible robotic arm plans its motion path according to the three-dimensional coordinates and the grasping strategy, and performs flexible grasping of the waste through the flexible end effector; S4. The intelligent flexible robotic arm sorts the captured waste into the corresponding storage bins of the intelligent sorting and storage unit.

Images