Intelligent residual material collecting vehicle based on visual detection
By using the visual detection and autonomous walking technology of the intelligent waste collection vehicle, the problem of low efficiency in traditional manual and mechanical collection has been solved, enabling accurate monitoring and data collection of waste, and improving the management efficiency and health level of dairy farms.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-05-21
- Publication Date
- 2026-06-19
AI Technical Summary
In existing technologies, manual or simple mechanical methods for collecting leftover feed are inefficient and cannot achieve accurate and efficient monitoring and data collection of leftover feed, resulting in data blind spots in dairy farm management and affecting feeding optimization and dairy cow health.
Design an intelligent waste material collection vehicle based on vision detection. It adopts a mobile chassis, a material collection mechanism, a material conveying mechanism, a storage bin, and a waste material detection module. It uses an image acquisition and processing module to monitor the distribution and quantity of waste material in real time, and achieves precise cleaning through autonomous walking and intelligent control.
It improves the efficiency of leftover feed cleaning, reduces the intensity of manual labor, realizes the quantitative monitoring and data collection of leftover feed, supports intelligent and precise feeding management, and provides data support for the digital operation of the ranch.
Smart Images

Figure CN122233182A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of aquaculture machinery technology, and specifically relates to an intelligent waste feed collection vehicle based on visual detection. Background Technology
[0002] With the rapid development of my country's livestock industry towards large-scale and intensive farming, the level of precision in feeding and management of large-scale dairy farms has become a key factor affecting breeding efficiency and animal welfare. In the dairy cow feeding process, Total Mixed Ration (TMR) technology has been widely applied. This involves precisely grinding, measuring, and mixing roughage, concentrates, and various additives according to a nutritional formula to create a nutritionally balanced feed. This well-mixed feed is then transported and spread in the dairy cow's feeding tunnel or trough for free access.
[0003] After feeding, uneaten feed inevitably remains in the feeding area, known as "leftovers," which need to be cleaned up. Currently, many dairy farms still rely on traditional manual methods for collecting leftovers. Operators must drive cleaning vehicles or use simple tools to patrol and clean along long feed troughs. This method is extremely inefficient and requires a continuous investment of manpower, especially in large barns where the cleaning work is time-consuming, severely encroaching on time and human resources that could be used for other important management tasks such as observing cow health and maintaining the environment. At the same time, it is difficult to ensure the timeliness and thoroughness of the collection of some leftovers, easily overlooking corners or feed residue mixed with manure, leading to the accumulation of leftovers in the feeding area. These organic residues ferment and mold rapidly in the warm and humid barn environment, not only causing secondary waste of feed but also potentially breeding pathogens, becoming a potential source of digestive system diseases or mastitis in dairy cows, directly threatening the health of the herd and the safety of milk.
[0004] To address the issue of manual labor efficiency, some preliminary mechanized collection devices, such as track-mounted or traction-type scraper conveyors, have emerged in the industry. However, these devices generally have a low level of intelligence. Their working mode is mostly repetitive scraping along preset routes, lacking the ability to perceive the actual distribution of leftover material. In the complex environment of actual cowsheds, the distribution of leftover material is extremely uneven, with concentrated accumulation points and areas that may have been licked clean by cows. The traditional "one-size-fits-all" operation method of machinery leads to wasted energy idling in areas with no material, while in areas with thick material, the effect may be reduced due to excessive load or incomplete cleaning, failing to achieve flexible, precise, and efficient on-demand operation.
[0005] Whether collected manually or using simple mechanical methods, the focus is on the physical act of cleaning itself, failing to quantify and collect data on leftover feed during the process. Managers cannot accurately know the specific quantity, distribution patterns, and trends of leftover feed each day and in each pen. This data is precisely the valuable information for assessing the palatability of TMR formulas, cow feeding behavior, and even the overall health of the herd. Current methods create data blind spots in the leftover feed stage, hindering data-driven, precision feeding optimization decisions. Summary of the Invention
[0006] Based on this, the present invention provides an intelligent waste material collection vehicle based on visual detection to solve the technical problems existing in the prior art. Whether it is manual or simple mechanical collection, it focuses on the physical cleaning action itself and cannot quantitatively monitor and collect data on the waste material during the cleaning process; managers cannot accurately know the specific quantity, distribution pattern and changing trend of waste material in each column every day.
[0007] The technical solution of the present invention to solve the above-mentioned technical problems is as follows: An intelligent waste collection vehicle based on vision detection, including The mobile chassis is capable of moving along a pre-set path on the ground. The material collection mechanism includes a material collection frame, a rolling main brush, a bidirectional auger, and a collection motor. The material collection frame is located at the front end of the mobile chassis. The rolling main brush is rotatably mounted in the material collection frame for brushing away residual material on the ground. The bidirectional auger is located in the material collection frame and behind the rolling main brush. The bidirectional auger is connected to the rolling main brush via a chain drive. The collection motor is connected to the bidirectional auger. The material conveying mechanism is connected at one end to the material collection frame and is inclined upwards behind the bidirectional auger; A storage bin is mounted on the mobile chassis, with its inlet corresponding to the output of the conveying mechanism; and a residual material detection module, including an image acquisition module and an image processing module. The image acquisition module is mounted on the collecting mechanism and is used to acquire images of the residual material in the area in front of the collecting mechanism and generate residual material image information. The image processing module is electrically connected to the image acquisition module and is used to identify the residual material image information within the image and control the collection motor to start based on the residual material image information.
[0008] Preferably, the material collection mechanism further includes a side auxiliary brush and a side motor. The side auxiliary brush is disposed on the material collection frame and located on both sides of the front end of the rolling main brush. The side motor is connected to the side auxiliary brush and is electrically connected to the image processing module.
[0009] Preferably, the material collection frame is provided with a lifting groove, which is located on both sides of the rolling main brush. The material collection mechanism also includes a main brush lifting component, which includes a rotating collar and a lifting cylinder. The rotating collar is sleeved on the rolling main brush, and one end of the lifting cylinder is connected to the rotating collar, while the other end is connected to the material collection frame. The rolling main brush moves within the lifting groove by retracting the lifting cylinder.
[0010] Preferably, the material conveying mechanism includes a material conveying scraper, a material conveying housing, and a material conveying motor. One end of the material conveying scraper is connected to the material collecting frame, and the other end is inclined upward and extends into the storage box. The material conveying motor is connected to the material conveying scraper and drives the material conveying scraper to rotate.
[0011] Preferably, the storage tank is provided with a plurality of stirring augers, which are distributed along the length of the storage tank, and the plurality of stirring augers are driven to rotate synchronously by a stirring motor installed on the outside of the storage tank.
[0012] Preferably, it further includes a lifting mechanism, which includes a hinged connecting rod and a lifting cylinder. The hinged connecting rod is hinged to the storage box. One end of the lifting cylinder is hinged to the movable chassis, and the other end is hinged to the conveying housing. The lifting cylinder drives the conveying mechanism to rise or fall.
[0013] Preferably, the image acquisition module is a depth camera, which uses point cloud data from the depth camera and, according to... This is used to convert the point coordinates Pc in the camera coordinate system to coordinates in the world coordinate system. ,in, The rotation matrix represents the rotation relationship between the camera coordinate system and the world coordinate system. It is a translation vector, which is a translation or offset vector between two coordinate systems; This is the inverse of the rotation matrix, used for inverse coordinate transformation; and it also includes the remaining image information. Transmit the data.
[0014] Preferably, the image processing module acquires image information of the remaining material. and the image information of the remaining material. The remaining material area is segmented, and the remaining material image information includes the volume and weight of the remaining material. The volume of the remaining material is obtained, where, This is the volume of the remaining material. Let be the height of the i-th pixel relative to the ground. and Where n is the actual physical size corresponding to a pixel, and n is the number of effective pixels; then through The weight of the remaining material is calculated, where, This is the weight of the remaining material. This represents the density of the remaining material.
[0015] Preferably, the rotational speed of the collecting motor is adjusted according to the weight of the remaining material obtained from the remaining material image, and according to... Adjust the speed of the collecting motor, wherein, The rotational speed of the collecting motor, The base speed of the collecting motor, This represents the weight of the remaining material obtained from the image of the remaining material. This represents the maximum remaining weight.
[0016] Preferably, the system also includes a ground cleanliness detection module, comprising a monocular camera and an image segmentation module. The monocular camera is used to capture images of the residual material area after the material collection mechanism has cleaned the area, forming clean area information. The image segmentation module is electrically connected to the monocular camera and acquires the clean area information, calculating the residual material rate of the clean area using an image segmentation algorithm. , .
[0017] Compared with the prior art, the present invention has at least the following advantages: This collection vehicle eliminates manual labor, especially in cleaning up leftover feed in large cow sheds, reducing labor intensity while improving cleaning efficiency and operational safety. Compared to traditional mechanical collection, this vehicle can quantitatively monitor and collect data on leftover feed during the cleaning process, allowing for accurate recording of leftover feed in different areas of the pen after the cows have eaten. This facilitates the assessment of TMR formula palatability, cow feeding behavior, and even the overall health of the herd, and is more conducive to intelligent and precise feeding management, providing reliable data support for the digital operation of the farm. Attached Figure Description
[0018] Figure 1 This is the first axonometric drawing of an intelligent waste material collection vehicle based on vision detection.
[0019] Figure 2 A first axonometric drawing (without part of the outer shell) of an intelligent waste collection vehicle based on vision detection.
[0020] Figure 3 for Figure 2 -A magnified view of a portion of the image.
[0021] Figure 4 This is the owner's view of an intelligent waste collection vehicle based on vision detection.
[0022] Figure 5 This is a left view of an intelligent waste collection vehicle based on vision detection.
[0023] Figure 6 This is a top view of an intelligent waste collection vehicle based on vision detection.
[0024] Figure 7 This is a schematic diagram of the remaining material detection module.
[0025] Figure 8 This is a schematic diagram of a floor cleanliness testing module.
[0026] The diagram shows: mobile chassis 100, material collection mechanism 200, material collection frame 210, lifting trough 211, rolling main brush 220, bidirectional auger 230, collection motor 240, side auxiliary brush 250, side motor 260, main brush lifting component 270, rotating collar 271, lifting cylinder 272, material conveying mechanism 300, material conveying scraper 310, material conveying housing 320, material conveying motor 330, storage bin 400, mixing auger 410, lifting mechanism 500, hinged connecting rod 510, lifting cylinder 520, residual material detection module 600, image acquisition module 610, image processing module 620, floor cleanliness detection module 700, monocular camera 710, and image segmentation module 720. Detailed Implementation
[0027] It should be noted that, unless otherwise specified, the embodiments and features described in the embodiments of the present invention can be combined with each other. The technical solutions of the present invention will be further described below with reference to the accompanying drawings of the embodiments of the present invention. The present invention is not limited to the following specific embodiments.
[0028] It should be understood that the same or similar reference numerals in the accompanying drawings of the embodiments correspond to the same or similar components. In the description of this invention, it should be understood that if terms such as "upper," "lower," "front," "rear," "left," "right," "top," and "bottom" indicate orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, they are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or component referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, the terms describing positional relationships in the accompanying drawings are for illustrative purposes only and should not be construed as limiting this patent. Those skilled in the art can understand the specific meaning of the above terms according to the specific circumstances.
[0029] Please refer to Figures 1 to 8 A vision-based intelligent waste collection vehicle, including The mobile chassis 100 is capable of moving along a pre-set route on the ground. The mobile chassis 100 provides stable support and a foundation for the entire collection vehicle, ensuring that the collection vehicle can accurately travel along the pre-set route. Employing autonomous navigation, it can automatically recognize pre-set markings on the ground or plan routes using built-in map information, avoiding the tediousness and errors of manual operation. Its built-in sensors and control system work together to adjust the direction and speed of travel in real time, ensuring comprehensive coverage of the work path.
[0030] The material collection mechanism 200 includes a material collection frame 210, a rolling main brush 220, a bidirectional auger 230, and a collection motor 240. The material collection frame 210 is located at the front end of the mobile chassis 100. The rolling main brush 220 is rotatably installed inside the material collection frame 210 and is used to roll and brush residual material on the ground, and to push the residual material to the rear by the rolling brush. The bidirectional auger 230 is disposed inside the material collection frame 210 and located behind the rolling main brush 220. The bidirectional auger 230 collects the residual material pushed up by the rolling main brush 220. The system serves to collect leftover materials. The bidirectional auger 230 and the rolling main brush 220 are connected via chain drive. The collecting motor 240 is connected to the bidirectional auger 230 and provides power to synchronously drive the bidirectional auger 230 and the rolling main brush 220, ensuring that they work together and improving the efficiency of leftover material collection. When the collecting motor 240 starts, the power is synchronously transmitted to the bidirectional auger 230 and the rolling main brush 220 via the chain drive system, realizing a continuous operation mode of brushing and collecting simultaneously, effectively avoiding material blockage or accumulation.
[0031] Specifically, the main rotating brush 220 rotates via a rotating spindle, driving the brushes wrapped around the spindle to perform floor cleaning. The brushes are relatively soft and can make full contact with the ground without damaging the surface material. The brushes collect the remaining material on the ground backward through the rolling action. At the same time, the brushes also peel off the remaining material stuck to the ground from the surface, improving cleaning efficiency.
[0032] Specifically, the bidirectional auger 230 is composed of left-handed and right-handed blades symmetrically welded onto a drive shaft. The intersection of the two blades forms a V-shaped guide zone that contracts towards the center. The rotation cycle of the bidirectional auger 230 is 30 seconds. The material conveying mechanism 300 is connected at one end to the material collecting frame 210 and is inclined upward behind the bidirectional auger 230; it is used to convey the collected residual material upward.
[0033] Specifically, the material conveying mechanism 300 can be used to convey material by rotary lifting or by screw conveying.
[0034] A storage bin 400 is mounted on the mobile chassis 100, and the inlet of the storage bin 400 corresponds to the output end of the conveying mechanism 300; it is used to receive the leftover material conveyed by the conveying mechanism 300 and temporarily store the collected leftover material to prevent spillage during transportation.
[0035] The residual material detection module 600 includes an image acquisition module 610 and an image processing module 620. The image acquisition module 610 is installed on the material collection mechanism 200 and is used to acquire images of the residual material area in front of the material collection mechanism 200 and generate residual material image information. The image processing module 620 is electrically connected to the image acquisition module 610 and is used to identify the residual material image information in the image and control the collection motor 240 to start according to the residual material image information.
[0036] Specifically, the image processing module 620 will also temporarily store the image information of the remaining material. After manual extraction, the image information of the remaining material will be sent to the terminal server, such as a computer or mobile phone, to facilitate the later analysis of the distribution pattern of the remaining material in the construction area and the efficiency of the cleaning operation.
[0037] Specific working principle: The collection vehicle operates autonomously, without manual driving. Following a pre-planned route, the mobile chassis 100 automatically travels to the work area. The image acquisition module 610 captures real-time images of the remaining feed distribution on the ground (primarily the feeding areas near the sides of the cattle pen), periodically taking images of the leftover feed and transmitting this image information to the image processing module 620 for analysis. When the image processing module 620 detects that the area of the leftover feed reaches a preset threshold, it immediately sends a start command to the control system of the collection motor 240. The collecting motor 240 is then powered on and starts running, driving the bidirectional auger 230 and the rolling main brush 220 to work synchronously. The rolling main brush 220 gathers the remaining material on the ground backward, while the bidirectional auger 230 gathers the material from both sides to the central V-shaped guide area through left-hand and right-hand rotating blades. The V-shaped guide area concentrates the remaining material to the inlet of the conveying mechanism 300. The conveying mechanism 300 receives the instruction from the image processing module 620 and starts up to transport the remaining material to the storage box 400 for temporary storage, and then transfers it for unified processing after the operation cycle is completed. This collection vehicle eliminates manual labor, especially in cleaning up leftover feed in large cow pens, reducing labor intensity while improving cleaning efficiency and operational safety. Compared to traditional mechanical collection, this vehicle can quantitatively monitor and collect data on leftover feed during the cleaning process. It can accurately record the leftover feed situation in different areas of the pen after milk consumption, which is beneficial for assessing the palatability of TMR formula, cow feeding behavior, and even the health status of the herd. It is also conducive to intelligent and precise feeding management, providing reliable data support for the digital operation of the farm.
[0038] In a preferred embodiment, since leftover material is also distributed on both sides of the cattle pen, the material collection mechanism 200 further includes a side auxiliary brush 250 and a side motor 260 to improve cleaning coverage. The side auxiliary brush 250 is disposed on the material collection frame 210 and located on both sides of the front end of the rolling main brush 220. The side motor 260 is connected to the side auxiliary brush 250 and electrically connected to the image processing module 620. When the image processing module 620 sends a start command to the control system of the collection motor 240, the image processing module 620 also simultaneously sends a start command to the control system of the side motor 260. The side motor 260 also starts simultaneously, driving the side auxiliary brush 250 to clean the leftover material in the near-dead corners on both sides of the cattle pen, improving cleaning coverage and preventing the fermentation of leftover material and the growth of bacteria.
[0039] In a preferred embodiment, as the rolling main brush 220 continuously cleans up residual material, the brush bristles of the rolling main brush 220 will wear down, causing an increase in the gap between the end face of the rolling main brush 220 and the ground, affecting the rolling main brush 220's ability to clean the ground. Therefore, the material collection frame 210 is provided with lifting grooves 211, which are located on both sides of the rolling main brush 220. The rotating shaft of the rolling main brush 220 is sleeved in the lifting grooves 211. The material collection mechanism 20 The 0 also includes a main brush lifting component 270, which includes a rotating collar 271 and a lifting cylinder 272. The rotating collar 271 is sleeved on the rolling main brush 220 and can rotate relative to the rolling main brush 220. One end of the lifting cylinder 272 is connected to the rotating collar 271 and the other end is connected to the material collection frame 210. The rolling main brush 220 is moved in the lifting groove 211 by the contraction of the lifting cylinder 272. In the initial state, when the main rolling brush 220 is not worn, the lifting cylinder 272 retracts to a shorter range, and the main rolling brush 220 is located at the upper end of the lifting groove 211. When the main rolling brush 220 is worn, the lifting cylinder 272 extends, pushing the main rolling brush 220 downward along the lifting groove 211, so that the brush at the end of the main rolling brush 220 is in contact with the ground, thereby improving the cleaning efficiency of residual materials.
[0040] In a preferred embodiment, the material conveying mechanism 300 includes a material conveying scraper 310, a material conveying housing 320, and a material conveying motor 330. One end of the material conveying scraper 310 is connected to the material collecting frame 210, and the other end is inclined upward and extends into the storage bin 400. The material conveying scraper 310 lifts the remaining material through the action of conveying and conveying. The material conveying housing 320 is disposed on both sides of the material conveying scraper 310 to protect the material conveying scraper 310. The material conveying motor 330 is connected to the material conveying scraper 310 and drives the material conveying scraper 310 to rotate. The image processing module 620 sends a start command to the control system of the collecting motor 240, and at the same time, the image processing module 620 also sends a start command to the control system of the conveying motor 330. The conveying motor 330 starts, driving the conveying scraper 310 to rotate. The conveying scraper 310 lifts the remaining material collected by the bidirectional auger 230 upward and conveys it to one side of the storage box 400. It is then discharged from the discharge end of the conveying scraper 310 and falls into the storage box 400. The remaining material can be quickly transferred through the conveying scraper 310.
[0041] In a preferred embodiment, the storage tank 400 is equipped with a plurality of agitators 410, which are distributed along the length of the storage tank 400. A stirring motor installed on the outside of the storage tank 400 drives the agitators 410 to rotate synchronously. When there is a significant amount of leftover material in the storage tank 400, the stirring motor starts, driving the agitators 410 to rotate. A weight sensor is installed inside the storage tank 400, and the stirring motor starts when the weight of the leftover material in the storage tank 400 reaches or exceeds a threshold. By installing agitators 410 inside the storage tank 400, the accumulation and fermentation of leftover material can be effectively prevented, facilitating subsequent processing of the leftover material.
[0042] In a preferred embodiment, the bottom of the storage bin 400 is also provided with a discharge port, and the discharge port is equipped with a conveyor belt. When it is necessary to unload the remaining material in the storage bin 400, the discharge port of the storage bin 400 is opened, and the remaining material is discharged from the discharge port by the stirring auger 410. Then, by the rotation of the conveyor belt, the remaining material flowing out from the discharge port is transported out along the conveyor belt, so as to achieve the purpose of unloading the remaining material.
[0043] In a preferred embodiment, a lifting mechanism 500 is further included. The lifting mechanism 500 includes a hinged connecting rod 510 and a lifting cylinder 520. The hinged connecting rod 510 is hinged to the storage box 400. One end of the lifting cylinder 520 is hinged to the movable chassis 100, and the other end is hinged to the conveying housing 320. The lifting cylinder 520 drives the conveying mechanism 300 to rise or fall. Before or after the collection vehicle enters or leaves the cattle shed, the mobile chassis 100 is electrically connected to the control system of the lifting cylinder 520. When the mobile chassis 100 is not working inside the cattle shed according to a preset route, the lifting cylinder 520 is in the extended state, the hinged connecting rod 510 rises, lifting the collecting mechanism 200 and the conveying mechanism 300, so that the collecting mechanism 200 is away from the ground to avoid collision with ground obstacles when not in operation. When the mobile chassis 100 enters the cattle shed and reaches the designated working position, the lifting cylinder 520 retracts, the hinged connecting rod 510 descends, driving the collecting mechanism 200 and the conveying mechanism 300 to descend, so that the collecting mechanism 200 is close to the ground for easy collection of leftover materials. Subsequently, the image acquisition module 610 operates, and the collecting mechanism 200 operates. The lifting mechanism 500 enables flexible lifting and lowering of the collecting mechanism 200 and the conveying mechanism 300, ensuring safe movement of the equipment in non-operating state and efficient collection in operating state.
[0044] In a preferred embodiment, the image acquisition module 610 is a depth camera, a visual sensor capable of acquiring three-dimensional point cloud data in real time, which can accurately capture the spatial distribution and terrain features of leftover material in the cowshed. It projects a speckle pattern using an infrared light source, calculates depth information using the principle of binocular parallax, and outputs a point cloud data frame containing XYZ coordinates and RGB colors. Through the point cloud data from the depth camera, the pixel coordinates in the point cloud data are converted into three-dimensional spatial coordinates in the world coordinate system, thereby achieving accurate identification and positioning of the leftover material accumulation area. And according to... Used to convert point coordinates in the camera coordinate system Convert to world coordinate system coordinates The world coordinate system refers to a fixed coordinate system with the ground as a reference plane, used to uniformly describe the spatial relationships of leftover feed within the cowshed; among which, The rotation matrix represents the rotation relationship between the camera coordinate system and the world coordinate system. It is a translation vector, which is a translation or offset vector between two coordinate systems; This is the inverse of the rotation matrix, used for inverse coordinate transformation; and it also includes the remaining image information. Transmit the data.
[0045] In a preferred embodiment, the image processing module 620 acquires image information of the remaining material. and the image information of the remaining material. The remaining material area is segmented into several equal-volume units, and an integral algorithm is used to calculate the remaining material image information, including the volume and weight of the remaining material. The volume of the remaining material is obtained, where, The volume of remaining material is the total volume of remaining material within the captured image of the remaining material. Let be the height of the i-th pixel relative to the ground. and Where n is the actual physical size corresponding to a pixel, and n is the number of effective pixels; then through The weight of the remaining material is calculated, where, This is the weight of the remaining material. The density of the remaining material is calculated as an average density value based on a preset density value of the remaining material on the ground, combined with historical sampling data and real-time detection results. Compared to traditionally captured images, existing images can only provide a rough image of the remaining material based on grayscale and shadow areas, resulting in a significant difference between the obtained weight of the remaining material and the actual weight. However, through the image acquisition module 610 and the image processing module 620 described in this application, the three-dimensional volume of the remaining material in the image can be accurately determined, leading to a more accurate weight calculation based on the remaining material volume.
[0046] In a preferred embodiment, the rotational speed of the collecting motor 240 is adjusted according to the weight of the remaining material obtained from the remaining material image, and according to... Adjust the speed of the collecting motor 240, wherein, The collecting motor rotates at 240 rpm. The base speed of the collecting motor is 240 rpm, and the base speed is 60 rpm. This represents the weight of the remaining material obtained from the image of the remaining material. The maximum residual material weight is 20 kg. Based on the residual material weight obtained from the residual material image, the rotation speed of the collection motor 240 is adaptively adjusted according to the residual material weight. In areas with more residual material, the rotation speed of the main rolling brush 220, the side auxiliary brush 250, and the bidirectional auger 230 can be increased. Compared with the rated speed of a traditional motor, the adaptive adjustment of the rotation speed of the collection motor 240 in this application can improve the collection rate of residual material, improve the cleanliness of the ground, and avoid residual material.
[0047] In a preferred embodiment, the system further includes a floor cleanliness detection module 700, which comprises a monocular camera 710 and an image segmentation module 720. The monocular camera 710 is used to capture images of the residual material area after the material collection mechanism 200 has been cleaned, and to form clean area information. The image segmentation module 720 is electrically connected to the monocular camera 710 and acquires the clean area information, and calculates the residual material rate of the clean area using an image segmentation algorithm. , The residual material rate is fed back in real time by the ground cleanliness detection module 700. When the residual material rate is greater than a preset threshold, the mobile chassis 100 can be triggered to push backward, and the material collection mechanism 200 will perform a second cleaning of the residual area, thereby improving the collection efficiency and cleaning quality of the residual material, ensuring a low ground residue rate and avoiding the growth of bacteria.
[0048] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the implementation of the present invention. Those skilled in the art can make other variations or modifications based on the above description. It is neither necessary nor possible to exhaustively describe all embodiments here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.
Claims
1. An intelligent waste material collection vehicle based on vision detection, characterized in that, include The mobile chassis is capable of moving along a pre-set path on the ground. The material collection mechanism includes a material collection frame, a rolling main brush, a bidirectional auger, and a collection motor. The material collection frame is located at the front end of the mobile chassis. The rolling main brush is rotatably mounted in the material collection frame for brushing away residual material on the ground. The bidirectional auger is located in the material collection frame and behind the rolling main brush. The bidirectional auger is connected to the rolling main brush via a chain drive. The collection motor is connected to the bidirectional auger. The material conveying mechanism is connected at one end to the material collection frame and is inclined upwards behind the bidirectional auger; A storage bin is mounted on the mobile chassis, and the inlet of the storage bin corresponds to the output end of the conveying mechanism. as well as The residual material detection module includes an image acquisition module and an image processing module. The image acquisition module is installed on the material collection mechanism and is used to acquire images of the residual material in the residual material area in front of the material collection mechanism and generate residual material image information. The image processing module is electrically connected to the image acquisition module and is used to identify the residual material image information in the image and control the collection motor to start according to the residual material image information. The image acquisition module is a depth camera, which uses point cloud data from the depth camera and, based on... Used to convert the coordinates of points in the camera coordinate system Convert to world coordinate system coordinates ,in, The rotation matrix represents the rotation relationship between the camera coordinate system and the world coordinate system. It is a translation vector, which is a translation or offset vector between two coordinate systems; This is the inverse of the rotation matrix, used for inverse coordinate transformation; and it also includes the remaining image information. Transmit; The image processing module acquires image information of the remaining material. and the image information of the remaining material. The remaining material area is segmented, and the remaining material image information includes the volume and weight of the remaining material. The volume of the remaining material is obtained, where, This is the volume of the remaining material. Let be the height of the i-th pixel relative to the ground. and Where n is the actual physical size corresponding to a pixel, and n is the number of effective pixels; then through The weight of the remaining material is calculated, where, This is the weight of the remaining material. This represents the density of the remaining material.
2. The intelligent waste material collection vehicle based on vision detection as described in claim 1, characterized in that, The material collection mechanism also includes a side auxiliary brush and a side motor. The side auxiliary brush is disposed on the material collection frame and located on both sides of the front end of the rolling main brush. The side motor is connected to the side auxiliary brush and is electrically connected to the image processing module.
3. The intelligent waste material collection vehicle based on vision detection as described in claim 1 or 2, characterized in that, The material collection frame is provided with lifting grooves, which are located on both sides of the rolling main brush. The material collection mechanism also includes a main brush lifting component, which includes a rotating collar and a lifting cylinder. The rotating collar is sleeved on the rolling main brush, and one end of the lifting cylinder is connected to the rotating collar, while the other end is connected to the material collection frame. The rolling main brush moves within the lifting groove by retracting the lifting cylinder.
4. The intelligent waste material collection vehicle based on vision detection as described in claim 1, characterized in that, The material conveying mechanism includes a material conveying scraper, a material conveying housing, and a material conveying motor. One end of the material conveying scraper is connected to the material collecting frame, and the other end is inclined upward and extends into the storage box. The material conveying housing is located on both sides of the material conveying scraper. The material conveying motor is connected to the material conveying scraper and drives the material conveying scraper to rotate.
5. The intelligent waste material collection vehicle based on vision detection as described in claim 1, characterized in that, The storage tank is equipped with several agitators, which are distributed along the length of the storage tank. The agitators are driven to rotate synchronously by an agitator motor installed on the outside of the storage tank.
6. The intelligent waste material collection vehicle based on vision detection as described in claim 4, characterized in that, It also includes a lifting mechanism, which includes a hinged connecting rod and a lifting cylinder. The hinged connecting rod is hinged to the storage box. One end of the lifting cylinder is hinged to the movable chassis and the other end is hinged to the conveying housing. The lifting cylinder drives the conveying mechanism to rise or fall.
7. The intelligent waste material collection vehicle based on vision detection as described in claim 1, characterized in that, Adjust the rotation speed of the collecting motor according to the weight of the remaining material obtained from the image of the remaining material, and according to... Adjust the speed of the collecting motor, wherein, The rotational speed of the collecting motor, The base speed of the collecting motor, This represents the weight of the remaining material obtained from the image of the remaining material. This represents the maximum remaining weight.
8. The intelligent waste material collection vehicle based on vision detection as described in claim 1, characterized in that, It also includes a ground cleanliness detection module, which comprises a monocular camera and an image segmentation module. The monocular camera is used to capture images of the residual material area after the material collection mechanism has cleaned the area, forming clean area information. The image segmentation module is electrically connected to the monocular camera and acquires the clean area information, calculating the residual material rate of the clean area using an image segmentation algorithm. , .