CN-121989871-A - Intelligent cleaning method and system for vehicle tires based on fusion perception and intelligent spray gun

CN121989871ACN 121989871 ACN121989871 ACN 121989871ACN-121989871-A

Abstract

The invention relates to an intelligent cleaning method and system for vehicle tires based on fusion perception and an intelligent spray gun, and relates to the technical field of intelligent vehicle cleaning. The intelligent vehicle cleaning method comprises the steps of building a global perception subsystem, an intelligent spray gun execution subsystem and a central intelligent decision unit, wherein the global perception subsystem outputs real-time dynamic parameters when a vehicle is driven in, the central intelligent decision unit distributes tasks according to a cross tracking pre-cleaning mode, drives the intelligent spray gun to cross clean a key region of a contralateral tire and connect the force pre-cleaning of the same lateral tire, the global perception subsystem generates a fine three-dimensional tire model and a dirt distribution map after the vehicle is stopped, the central intelligent decision unit controls the intelligent spray gun to execute a profiling fine cleaning mode, the global perception subsystem identifies a cleaning platform and a drain dirt residual region and outputs contour data when the vehicle is driven in, and the central intelligent decision unit controls all the intelligent spray guns to switch a ground curtain self-cleaning mode to synchronously spray. The invention can clean the vehicle tyre efficiently, thoroughly and stably.

Inventors

LIU LIMING
LI SHUANG
CAO JINXUAN
CHEN MING
LI YU
WANG ZHENXING
LU QIULIN

Assignees

湖南九九智能环保股份有限公司

Dates

Publication Date: 20260508
Application Date: 20260202

Claims (10)

1. An intelligent cleaning method for vehicle tires based on fusion perception and intelligent spray guns is characterized by comprising the following steps: Step 1, building a global perception subsystem, an intelligent spray gun execution subsystem and a central intelligent decision unit; Step 2, when a vehicle is driven in, the global perception subsystem perceives the motion state of the vehicle in real time, performs fusion processing on the collected vehicle point cloud data and image data, extracts the contour and position characteristic information of the vehicle, and combines the vehicle running speed to output real-time dynamic parameters; step 3, after the vehicle is stopped, the global perception subsystem acquires multi-source data and fuses the multi-source data to generate a fine three-dimensional model of the tire and a dirt distribution diagram, the central intelligent decision unit discretizes the three-dimensional contour of the tire and plans a path according to the fine three-dimensional model and the dirt distribution diagram, and then the intelligent spray gun is controlled to execute a profiling fine washing mode; and 4, when the vehicle exits, the global perception subsystem identifies the cleaning platform and the drain ditch dirt residual area and outputs contour data, and the central intelligent decision unit plans a spraying path according to the contour data and controls all intelligent spray guns to switch the floor curtain self-cleaning mode to synchronously spray.
2. The intelligent cleaning method for vehicle tires based on the fusion perception and intelligent spray gun according to claim 1, wherein in the step 1, the global perception subsystem is an anti-interference fusion framework constructed by a laser radar, an industrial camera and an auxiliary sensor; the intelligent spray gun execution subsystem comprises intelligent spray guns symmetrically arranged on two sides of the cleaning channel, and each intelligent spray gun is provided with a lifting component and a high-degree-of-freedom cradle head; the central intelligent decision unit is internally provided with a self-adaptive cleaning model and a cooperative controller.
3. The intelligent cleaning method for vehicle tires based on the fusion perception and intelligent spray gun according to claim 1, wherein in the step 2, fusion processing is performed on collected vehicle point cloud data and image data, wheel contour and position characteristic information are extracted, and real-time dynamic parameters are output by combining with vehicle running speed, and the method comprises the following steps: After fusion processing is carried out on the collected vehicle point cloud data and the image data, tracking and estimating are carried out on continuous frame displacement through a Kalman filtering algorithm, and the tire rotation speed and the vehicle travelling speed are calculated according to a time sequence difference method, wherein the calculation formula is as follows: ; ; ; in the formula, Representing the average displacement variation of the axle center of the wheel between two adjacent sampling moments; Representation of The axle center position of the wheel at moment; Representation of The axle center position of the wheel at moment; representing a sampling period; representing a vehicle travel speed; representing the rotational angular velocity of the tire; Representing the amount of tire angle change between successive frames; Fitting the circle center coordinates of the wheel contour through Hough transformation, and calculating the three-dimensional coordinates of the wheel under a global coordinate system by combining the space-time calibration results of the auxiliary sensor; The three-dimensional coordinates and the vehicle running speed Rotational angular velocity of tire As a real-time dynamic parameter output.
4. The intelligent cleaning method for vehicle tires based on fusion perception and intelligent spray guns as set forth in claim 3, wherein in step 2, the central intelligent decision unit predicts the tire track according to the real-time dynamic parameters, and when tasks are distributed according to a cross tracking pre-cleaning mode, a space-time coupling scheduling algorithm is adopted, and the space-time window of each intelligent spray gun for intervention and exit cleaning is calculated by combining the real-time dynamic parameters of the wheels, wherein the calculation formula is as follows: ; ; in the formula, Indicating the intervention cleaning time of the intelligent spray gun; Representing a preset cleaning starting distance; Indicating the moment when the intelligent spray gun exits the cleaning; showing the cleaning time dynamically adjusted according to the degree of dirt of the tire.
5. The intelligent cleaning method for vehicle tires based on the fusion perception and intelligent spray gun according to claim 3, wherein in the step 2, when the tires on the same side are pre-washed in relay, the starting time of the downstream intelligent spray gun is as follows: ; in the formula, Indicating the starting time of the downstream intelligent spray gun; indicating the starting time of an upstream intelligent spray gun; Representing the tire radius.
6. The intelligent cleaning method for vehicle tires based on the fusion perception and intelligent spray gun according to any one of claims 2 to 5, wherein in step 3, the process of obtaining multi-source data by the global perception subsystem and fusing to generate a fine three-dimensional model of the tire and a dirt distribution diagram comprises the following steps: Step 301, acquiring multi-view point cloud data of the tire surface through laser radar scanning, and acquiring texture information of the tire surface through an industrial camera, wherein the multi-view point cloud data and the texture information are used as input, registering and fusing are carried out through an ICP algorithm, and fused point cloud data under a unified coordinate system are output; step 302, taking the fusion point cloud data as input, sequentially removing point cloud noise through voxel filtering, reconstructing and optimizing the point cloud surface geometry through poisson, and outputting a tire fine three-dimensional model; step 303, taking the fusion point cloud data as input, identifying attachment areas in the fusion point cloud data based on a deep learning semantic segmentation network, and carrying out quantization processing on pollution density of each area to generate a pixel-by-pixel pollution thermodynamic diagram; and 304, performing space coordinate alignment on the pixel-by-pixel soil thermodynamic diagram and the tire fine three-dimensional model, and outputting a soil distribution diagram containing tire geometric characteristics and soil distribution information.
7. The intelligent cleaning method for vehicle tires based on the fusion perception and intelligent spray gun according to claim 6, wherein in step 3, the central intelligent decision unit discretizes the three-dimensional contour of the tire and plans a path according to the fine three-dimensional model and the dirt distribution diagram, comprising: step 311, extracting the pollution density of each area according to the pollution distribution map Comparing the pollution density of each area according to a preset pollution density threshold value, and identifying a high pollution area and a conventional pollution area; step 312, based on the tire fine three-dimensional model, for a conventional contaminated area, at sampling intervals Extracting contour point coordinates along the circumferential direction of the tire to form an ordered point set For the high pollution area, shortening the sampling interval to carry out encryption extraction and outputting an encryption ordered point set; Step 313, after discretizing the ordered point set and the encrypted ordered point set, fitting and optimizing through a Bezier curve to generate a continuous and conductive spray gun motion track, and outputting a cleaning path planning result.
8. The intelligent cleaning method for vehicle tires based on the fusion perception and intelligent spray gun according to any one of claims 1 to 5, wherein in step 3, the method further comprises the step that after cleaning is completed, the global perception subsystem feeds back the cleanliness characteristic, and whether incremental or fixed-point intensified cleaning is executed is judged through a preset cleaning threshold.
9. The intelligent cleaning method for vehicle tires based on the fusion awareness and intelligent spray gun according to any one of claims 2 to 5, wherein in step 4, the global awareness subsystem identifies a cleaning platform and a drain dirt residual area and outputs contour data, comprising: step 401, collecting point cloud data of a cleaning platform and a drainage ditch through a laser radar, performing filtering processing, and outputting effective point cloud data of a ground area; step 402, based on the ground area effective point cloud data, dividing a high-humidity muddy water residual area according to a preset reflection intensity threshold value, and identifying the boundary of the high-humidity muddy water residual area by combining the difference of the ground height; And step 403, clustering the high-humidity muddy water residual areas through an European clustering algorithm, extracting the outline coordinates of each dirt area, and outputting the outline data of the dirt residual areas.
10. Vehicle tire intelligent cleaning system based on fusion perception and intelligent spray gun, characterized in that the system comprises: The cleaning system building module is used for building a global perception subsystem, an intelligent spray gun execution subsystem and a central intelligent decision unit; The intelligent vehicle cleaning system comprises a global perception subsystem, a central intelligent decision unit and a remote control unit, wherein the global perception subsystem is used for perceiving the motion state of a vehicle in real time when the vehicle is driven in, carrying out fusion processing on the acquired vehicle point cloud data and image data, extracting the contour and position characteristic information of the vehicle, and combining the vehicle running speed to output real-time dynamic parameters; The intelligent tire cleaning system comprises a profiling fine washing module, a central intelligent decision unit, an intelligent spray gun, a central intelligent decision unit and a control unit, wherein the profiling fine washing module is used for acquiring multi-source data by the global perception subsystem after a vehicle is stopped, and generating a fine three-dimensional model and a dirt distribution map of the tire by fusion; The ground curtain self-cleaning module is used for identifying the cleaning platform and the drain ditch dirt residual area and outputting contour data by the global perception subsystem when the vehicle is driven out, and the central intelligent decision unit is used for planning a spraying and sweeping path according to the contour data and controlling all intelligent spray guns to switch the ground curtain self-cleaning mode to synchronously spray and sweep.

Description

Intelligent cleaning method and system for vehicle tires based on fusion perception and intelligent spray gun Technical Field The invention relates to the technical field of intelligent car washing, in particular to an intelligent car tire cleaning method and system based on fusion perception and an intelligent spray gun. Background In the loading and unloading and transporting links of the freight vehicle, various dirt is extremely easy to adhere to the surface of the tire, especially the positions such as the tread grooves, the side surfaces, the double-wheel cracks and the like, and the complex structure becomes the important and difficult point of cleaning. There are still a number of significant shortcomings in the current cleaning methods in the industry, which make it difficult to meet the requirements of efficient, thorough, and automated cleaning. The traditional cleaning means rely on manual hand-held high-pressure water gun or fixed high-pressure nozzle array to wash, not only cleaning efficiency is low, but also serious water resource waste problem exists, and is limited by a fixed structure or a manual operation range, all areas of the complex three-dimensional curved surface of the tire cannot be covered, and a large number of cleaning blind areas exist. Although some prior art and related patents propose ideas for identifying dirt and adjusting cleaning parameters through vision and spectrum, the problem of cleaning blind areas is still outstanding and seriously affects the overall cleaning effect, and the problem of complex cleaning requirements of a tire, which is a key and difficult to clean, is still not solved in a targeted manner. Meanwhile, in an actual cleaning working condition, the environment of high-pressure water mist diffusion and water splashing frequently can greatly interfere the recognition performance of a traditional vision auxiliary sensor, so that a 'blind' phenomenon occurs in a perception system, stable tracking and accurate decision-making of a tire cannot be realized, and the realization of automatic closed-loop cleaning of the whole process is difficult to support. From the perspective of an executing mechanism, the conventional scheme either adopts a fixed nozzle, cannot adapt to the complex outline of the tire of the freight vehicle, or relies on a mechanical arm with complex structure, so that the cost is high, faults are easy to occur under severe engineering environment, adaptability and reliability are poor, and the subsequent maintenance cost is high. In addition, in the existing cleaning flow, all stages of vehicle entrance, stillness, exiting and the like are mutually isolated, a coherent and cooperative design is lacking, an integral solution with optimal resource utilization rate cannot be formed, and even if part of solutions are provided with intelligent decision models, a cleaning executing mechanism which can be adapted to a freight vehicle with a complex structure is lacking, so that an excellent cleaning strategy cannot be effectively landed, finally, the cleaning effect is greatly reduced, and the high-efficiency, thorough and stable cleaning requirement of the tires of the freight vehicle is difficult to meet. Disclosure of Invention Accordingly, it is necessary to provide a vehicle tire intelligent cleaning method and system based on a fusion sensing and intelligent spray gun, which can efficiently, thoroughly and stably clean a vehicle tire. An intelligent cleaning method for vehicle tires based on fusion perception and intelligent spray guns, the method comprising: Step 1, building a global perception subsystem, an intelligent spray gun execution subsystem and a central intelligent decision unit; Step 2, when a vehicle is driven in, the global perception subsystem perceives the motion state of the vehicle in real time, performs fusion processing on the collected vehicle point cloud data and image data, extracts the contour and position characteristic information of the vehicle, and combines the vehicle running speed to output real-time dynamic parameters; step 3, after the vehicle is stopped, the global perception subsystem acquires multi-source data and fuses the multi-source data to generate a fine three-dimensional model of the tire and a dirt distribution diagram, the central intelligent decision unit discretizes the three-dimensional contour of the tire and plans a path according to the fine three-dimensional model and the dirt distribution diagram, and then the intelligent spray gun is controlled to execute a profiling fine washing mode; and 4, when the vehicle exits, the global perception subsystem identifies the cleaning platform and the drain ditch dirt residual area and outputs contour data, and the central intelligent decision unit plans a spraying path according to the contour data and controls all intelligent spray guns to switch the floor curtain self-cleaning mode to synchronously spray. On the other hand, still provide a vehicle