Tire management method and system for port AGV, terminal and medium

Abstract

The invention belongs to the technical field of AGV equipment management, and particularly discloses a tire management method and system for a port AGV, a terminal and a medium. Comprising the steps that tire information of all AGVs of a motorcade is obtained, and the tire information comprises tire pressure, abrasion and damage information; when it is detected that tire abrasion exceeds the limit, and tire pressure is abnormal or damaged, alarm information is generated; establishing a tire wear prediction model based on the historical mileage and load information of the AGV, performing dynamic prediction and alarm judgment on the tire wear state, and determining the maintenance priority; and the target AGV is scheduled to a tire maintenance area according to the maintenance priority in combination with a port operation plan, inflation and deflation or tire replacement operation is executed, and updated data is synchronized to a motorcade tire management end after the operation is completed. According to the invention, real-time monitoring, intelligent prediction, automatic scheduling and unmanned maintenance of port AGV tires are realized, and equipment safety and port area operation efficiency are improved.

Description

Technical Field

[0001] This invention belongs to the field of AGV equipment management technology, specifically relating to a tire management method, system, terminal, and medium for port AGVs. Background Technology

[0002] With the large-scale construction of automated terminals, Automated Guided Vehicles (AGVs) have become essential equipment for container loading, unloading, and horizontal transportation in ports. AGVs typically operate for extended periods on closed port roads, undertaking container transport tasks, and their operational efficiency and reliability directly impact the overall port's operational capacity. AGVs mostly use pneumatic heavy-duty tires to meet the load-bearing and vibration-damping requirements under complex road conditions. However, tires are prone to wear, bulges, damage, and abnormal tire pressure under prolonged high loads, frequent starts and stops, high turning frequency, and salt spray and humid environments. Failure to detect and address these issues promptly can lead to vehicle downtime, or even delays in cargo flow and safety accidents.

[0003] Currently, tire management for AGVs in ports mainly relies on periodic manual inspections and passive alarm mechanisms. Although some automated terminals have installed tire pressure monitoring systems on their vehicles, enabling basic tire pressure abnormality alarms, this is not the case.

[0004] However, current technology still cannot identify tire wear and damage conditions online. Existing tire condition detection is mostly based on manual visual inspection or portable inspection instruments, lacking continuity and real-time capability. The detection results depend on personnel experience and have low accuracy. In addition, the tire changing and inflation / deflation operations of port AGVs still require manual execution and lack a linkage mechanism with the fleet dispatch system, resulting in low maintenance and dispatch efficiency and insufficient vehicle utilization. Summary of the Invention

[0005] This invention addresses the problems in the prior art by providing a tire management method, system, terminal, and medium for port AGVs, thereby solving the issues in the background art such as the inability to identify tire wear and damage conditions online, lack of continuity and real-time performance, reliance on human experience for detection results, and low accuracy.

[0006] The technical solution adopted in this invention is as follows: In a first aspect, this application provides a tire management method for port AGVs, the method comprising the following steps: Step S1: Obtain tire information for all AGVs in the fleet. Tire information includes tire pressure, wear, and damage information. Tire pressure information is collected by tire pressure sensors installed on the tires; Wear information is obtained by identifying the distance between the tire tread and the wear reference line using cameras installed in the automated operation area. Image recognition algorithms are used to determine the condition of the tire body to obtain information on damage. And upload the tire information to the fleet tire management terminal; Step S2: When the AGV passes through the automated operation area, the camera and the tire pressure sensor of the AGV will detect the AGV. When the tire wear exceeds the limit, the tire pressure is abnormal or the tire is damaged, a first-level alarm message will be generated. Otherwise, step S3 will be executed to predict the tires of the AGV. Step S3: Establish a tire prediction model based on the historical mileage and load information of the AGV, and perform alarm prediction on the tires according to the tire prediction model and tire information. When the time corresponding to the alarm prediction is reached, generate a secondary alarm message. The maintenance priority of each AGV tire is determined based on the degree of abnormality of the alarm information. Step S4: Based on the maintenance priority information and the port operation plan, dispatch the AGVs that need maintenance to the tire maintenance area. When the AGV requiring maintenance arrives at the tire maintenance area, it will perform tire inflation / deflation and / or tire replacement operations. Once completed, the updated tire status data will be synchronized to the fleet tire management terminal.

[0007] Furthermore, in step S3, the AGV operation process is divided into several continuous wear intervals with constant load according to the detection time sequence, and the triplets of each interval are obtained. ; in, Let be the mileage of the i-th interval. Let be the load constant for the i-th interval. This represents the measured wear amount within this range; Based on the aforementioned triplet, a functional relationship between wear amount and mileage and load is established, and the parameters are estimated using the least squares method. The functional model is as follows:

[0008] These are the preset model parameters; To obtain the cumulative wear and tear at the current moment. Wear threshold Then, based on the representative interval characteristics obtained from the most recent intervals... ; in, For the expected single-section driving distance, For the expected single-section load, Expected duration of a single interval, The weighted average or median of the corresponding values ​​from the most recent intervals can be used; Calculate the predicted wear amount for a single interval and determine that it meets the requirements. Find the smallest positive integer n to obtain the estimated time to reach the wear threshold:

[0009] in, The current moment; A level 2 alarm message is generated when the time corresponding to the alarm prediction is reached.

[0010] Furthermore, in step S1, a defect identification model is established based on the acquisition of damage information. The defect identification model is trained and used for online identification in the following manner: A dataset was constructed and labeled for tire images collected in the automated operation area. The defect categories included sidewall cracks, bulges, cuts, foreign object embedding, and shoulder defects. The image is subjected to distortion correction and scaling. The scaling is achieved based on the pixel distance between the tire tread and the wear reference line or the known geometric dimensions of the rim mounting hole. A supervised learning approach is used to train a probabilistic output model that integrates classification and localization to obtain the confidence scores for defect location and category.

[0011] Furthermore, in step S2, the alarm and maintenance priority determination includes: setting the classification thresholds for wear, tire pressure and puncture and comparing the thresholds respectively; When multiple indicators are triggered simultaneously on the same vehicle, the alarm and maintenance priority are determined in the order of damage, wear, and tire pressure. Within the same priority, the scores are ranked based on a weighted score, which is composed of wear proximity, tire pressure deviation, defect confidence or defect area ratio, and cumulative mileage or time since the last maintenance, and is obtained by summing after weight normalization. An alarm determination strategy is adopted, and an alarm is output only when continuous sampling meets the grade threshold or the duration of exceeding the limit is not less than the preset duration; and a cooling time window is set to suppress short-term repeated scheduling.

[0012] Furthermore, it also includes the following steps: Step S5: Each time a tire passes through the camera detection area of ​​the automated operation zone, the tire prediction model is corrected using the tire information. For model parameters The update process includes: forming a sample set from the K most recent wear intervals with unchanged load, and constructing a sample matrix:

[0013] in For the first The driving distance of each section, For the first The load constant of each interval No. Measured wear amount in each interval; The parameter vector is calculated using the weighted least squares method:

[0014] Here is the weight matrix, and the weights are:

[0015] The time decay factor, For the first The confidence coefficient of each interval detection; The corrected prediction is obtained using the updated parameters:

[0016] Used for subsequent wear prediction and alarm determination.

[0017] Secondly, this application provides a tire management system for port AGVs, used to implement the tire management method for port AGVs as described in the first aspect, the system comprising: The tire information acquisition unit is used to acquire tire information of all AGVs in the fleet. The tire information includes at least tire pressure information, wear information, and damage information. The tire information acquisition unit includes tire pressure sensors installed on the tires, cameras installed in the automated operation area, and an interactive interface for manual input. The acquired tire information is then uploaded to the back-end processing terminal through a communication network. The tire prediction unit constructs a tire wear prediction model based on the AGV's historical mileage, load information, and wear detection results. It divides the continuous time period with constant load into several wear intervals, obtains triplet data for each wear interval, and establishes a functional relationship between wear amount and mileage and load through a regression algorithm. When the latest wear data detected by the camera or by manual inspection is obtained, the updated samples are used to re-estimate the model parameters to complete the model correction. Based on the current cumulative wear amount and wear threshold, the model is called to calculate the estimated time required for the tire wear amount to reach the wear threshold, and the tire life end prediction result is output for the alarm and priority determination unit to perform subsequent alarm and scheduling control. The alarm and priority determination unit is used to comprehensively determine the results from the tire information acquisition unit and the tire prediction unit, set the classification thresholds for wear, tire pressure and damage and compare the thresholds respectively; when the same vehicle triggers multiple indicators at the same time, alarms are output according to the preset priority order and maintenance priorities are determined, and sorted based on weighted scores within the same priority; the alarm and priority determination unit is also used to implement continuous sampling duration and cooling time window constraints to suppress short-term repeated scheduling. The scheduling unit is used to generate scheduling instructions based on maintenance priorities and port operation plans, to schedule target AGVs to tire maintenance or inspection areas, and to synchronize the scheduling results to the production scheduling system. The maintenance execution unit is used to perform inflation / deflation operations and / or tire replacement operations after the AGV arrives at its position, and to generate maintenance result data after the operations are completed. The data synchronization unit is used to synchronize the updated tire status data and maintenance records to the backend processing end after the maintenance execution unit completes its work, and to provide the tire prediction unit with updates for subsequent prediction and alarm judgment.

[0018] Furthermore, after receiving the work instructions from the scheduling unit, the maintenance execution unit performs on-time verification and alignment confirmation of the AGV based on the target wheel position and status information provided by the tire information acquisition unit. After confirming successful alignment, it automatically selects and executes either inflation / deflation or tire replacement operations. During the operation, it completes the work steps in a preset order and verifies the work results in real time. If an operation abnormality is detected, it reports the abnormality information to the alarm and priority determination unit to trigger a re-inspection or manual review process.

[0019] Furthermore, after completing the inflation / deflation or tire replacement operation, the maintenance execution unit transmits the operation type, operation time, operation result, and corresponding tire status parameters back to the backend processing unit through the data synchronization unit. This data is then used by the tire prediction unit to update the prediction input dataset and alarm judgment criteria. When the deviation between the transmitted data and the prediction result of the tire prediction unit exceeds a preset threshold, the alarm and priority judgment unit generates a model calibration prompt and notifies the tire prediction unit to execute the parameter re-estimation process.

[0020] Thirdly, this application provides a terminal, including: Memory for storing tire management programs for port AGVs; A processor is configured to perform the steps of the tire management method for a port AGV as described in the first aspect when executing the tire management device for a port AGV.

[0021] Fourthly, this application provides a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes the tire management method for port AGVs as described in the first aspect.

[0022] As can be seen from the above technical solutions, the advantages of the present invention are: This invention constructs an intelligent management process for the entire lifecycle of AGV tires in ports, achieving closed-loop control of real-time tire status perception, dynamic prediction, hierarchical alarms, and automatic maintenance. This method comprehensively utilizes tire pressure sensors, visual recognition, and historical operational data to simultaneously acquire multi-dimensional tire information such as tire pressure, wear, and damage, overcoming the problems of low frequency, large errors, and strong lag in traditional manual inspections. The tire wear prediction model, based on mileage and load information, can be parameter-corrected after tire inspection, ensuring that prediction results are adaptively updated according to operational status, thereby improving the accuracy and stability of tire life prediction. By setting multi-level alarm thresholds for wear, tire pressure, and damage, and establishing a comprehensive scoring mechanism to prioritize maintenance, maintenance resources can be rationally allocated in the high-density operation environment of ports, reducing concentrated vehicle downtime and scheduling conflicts. At the system level, by linking tire prediction results with the fleet scheduling system, scheduling instructions are automatically generated and AGVs are guided to the maintenance area, achieving autonomous allocation and execution of maintenance tasks. The maintenance execution unit employs a hydraulic lifting device and a tire changing device in conjunction with a slide rail assembly, enabling automated lifting, disassembly, alignment, locking, and inflation / deflation operations, significantly reducing manual labor intensity and operational risks. Overall, this invention achieves digital, intelligent, and unmanned operation of port AGV tire management, improving vehicle uptime and tire lifespan, and enhancing the safety and economic benefits of automated port operations. Attached Figure Description

[0023] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the description will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a step diagram of the tire management method for port AGVs in the embodiment; Figure 2 This is a schematic diagram of the tire structure; Figure 3 This is a structural diagram of the maintenance execution unit; Figure 4 A schematic diagram of the slide rail assembly for the maintenance execution unit; Figure 5 A schematic diagram of a tire changing device; Figure 6 This is a schematic diagram of the lateral movement mechanism in a tire changing device. Figure 7 This is a partial structural diagram of a tire changing device.

[0025] In the diagram: 11. Tire tread; 12. Tire wear reference line; 13. Tire wear warning line; 14. Tire mounting bolt holes; 15. Press-fit tire inflation / deflation valve; 21. Hydraulic lifting device; 22. Tire changing device; 23. Used tire storage area; 24. Slide rail assembly; 25. New tire storage area; 31. Lateral slide rail; 32. Lateral base; 33. Lateral platform; 34. Lateral moving block; 35. Longitudinal slide rail; 36. Longitudinal base; 37. Longitudinal moving block; 38. Tire bracket stop; 41. Lateral movement mechanism; 42. Mechanism steering base; 43. Longitudinal platform; 44. Clamping cylinder; 45. Tilting cylinder; 46. Tire clamp; 47. Camera; 48. Bolt sleeve; 49. Press-fit tire inflation / deflation interface; 51. Hydraulic electrical cabinet; 52. Air compressor cabinet; 53. Lateral movement cylinder; 54. Rotary gear; 55. Rotary rack; 56. Rotary cylinder; 57. Rotary cable drag chain; 58. Fixed connecting rod; 59. Hydraulic center rotary joint. Detailed Implementation

[0026] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] Please see Figure 1 As shown, the present invention provides a tire management method for port AGVs, comprising the following steps: Step S1: Obtain tire information for all AGVs in the fleet. Tire information includes tire pressure, wear, and damage information. Tire pressure information is collected by tire pressure sensors installed on the tires; Wear information is obtained by identifying the distance between the tire tread and the wear reference line using cameras installed in the automated operation area. Image recognition algorithms are used to determine the condition of the tire body to obtain information on damage. And upload the tire information to the fleet tire management terminal; Step S2: When the AGV passes through the automated operation area, the camera and the tire pressure sensor of the AGV will detect the AGV. When the tire wear exceeds the limit, the tire pressure is abnormal or the tire is damaged, a first-level alarm message will be generated. Otherwise, step S3 will be executed to predict the tires of the AGV. Step S3: Establish a tire prediction model based on the historical mileage and load information of the AGV, and perform alarm prediction on the tires according to the tire prediction model and tire information. When the time corresponding to the alarm prediction is reached, generate a secondary alarm message. Step S4: Based on the maintenance priority information and the port operation plan, dispatch the AGVs that need maintenance to the tire maintenance area. When the AGV requiring maintenance arrives at the tire maintenance area, it will perform tire inflation / deflation and / or tire replacement operations. Once completed, the updated tire status data will be synchronized to the fleet tire management terminal.

[0028] In some embodiments, in step S3, the AGV operation process is divided into several continuous wear intervals with constant load according to the detection time sequence, and the triplet of each interval is obtained. ; in, Let be the mileage of the i-th interval. Let be the load constant for the i-th interval. This represents the measured wear amount within this range; Based on the aforementioned triplet, a functional relationship between wear amount and mileage and load is established, and the parameters are estimated using the least squares method. The functional model is as follows:

[0029] These are the preset model parameters; To obtain the cumulative wear and tear at the current moment. Wear threshold Then, based on the representative interval characteristics obtained from the most recent intervals... ; in, For the expected single-section driving distance, For the expected single-section load, Expected duration of a single interval, The weighted average or median of the corresponding values ​​from the most recent intervals can be used; Calculate the predicted wear amount for a single interval and determine that it meets the requirements. Find the smallest positive integer n to obtain the estimated time to reach the wear threshold:

[0030] in, The current moment; A level 2 alarm message is generated when the time corresponding to the alarm prediction is reached.

[0031] In some embodiments, in step S1, a defect identification model is established based on the acquisition of damage information. The defect identification model is trained and used for online identification in the following manner: A dataset was constructed and labeled for tire images collected in the automated operation area. The defect categories included sidewall cracks, bulges, cuts, foreign object embedding, and shoulder defects. The image is subjected to distortion correction and scaling. The scaling is achieved based on the pixel distance between the tire tread and the wear reference line or the known geometric dimensions of the rim mounting hole. A supervised learning approach is used to train a probabilistic output model that integrates classification and localization to obtain the confidence scores for defect location and category.

[0032] In some embodiments, step S2, the alarm and maintenance priority determination includes: setting classification thresholds for wear, tire pressure and puncture and comparing the thresholds respectively; When multiple indicators are triggered simultaneously on the same vehicle, the alarm and maintenance priority are determined in the order of damage, wear, and tire pressure. Within the same priority, the scores are ranked based on a weighted score, which is composed of wear proximity, tire pressure deviation, defect confidence or defect area ratio, and cumulative mileage or time since the last maintenance, and is obtained by summing after weight normalization. An alarm determination strategy is adopted, and an alarm is output only when continuous sampling meets the grade threshold or the duration of exceeding the limit is not less than the preset duration; and a cooling time window is set to suppress short-term repeated scheduling.

[0033] In some embodiments, the following steps are also included: Step S5: Each time a tire passes through the camera detection area of ​​the automated operation zone, the tire prediction model is corrected using the tire information. For model parameters The update process includes: forming a sample set from the K most recent wear intervals with unchanged load, and constructing a sample matrix:

[0034] in For the first The driving distance of each section, For the first The load constant of each interval No. Measured wear amount in each interval; The parameter vector is calculated using the weighted least squares method:

[0035] Here is the weight matrix, and the weights are:

[0036] The time decay factor, For the first The confidence coefficient of each interval detection; The corrected prediction is obtained using the updated parameters:

[0037] Used for subsequent wear prediction and alarm determination.

[0038] Please see Figure 2-7 As shown, in some embodiments, this application provides a tire management system for port AGVs to implement a tire management method for port AGVs, the system comprising: The tire information acquisition unit is used to acquire tire information of all AGVs in the fleet. The tire information includes at least tire pressure information, wear information, and damage information. The tire information acquisition unit includes tire pressure sensors installed on the tires, cameras installed in the automated operation area, and an interactive interface for manual input. The acquired tire information is then uploaded to the back-end processing terminal through a communication network. The tire prediction unit constructs a tire wear prediction model based on the AGV's historical mileage, load information, and wear detection results. It divides the continuous time period with constant load into several wear intervals, obtains triplet data for each wear interval, and establishes a functional relationship between wear amount and mileage and load through a regression algorithm. When the latest wear data detected by the camera or by manual inspection is obtained, the updated samples are used to re-estimate the model parameters to complete the model correction. Based on the current cumulative wear amount and wear threshold, the model is called to calculate the estimated time required for the tire wear amount to reach the wear threshold, and the tire life end prediction result is output for the alarm and priority determination unit to perform subsequent alarm and scheduling control. The alarm and priority determination unit is used to comprehensively determine the results from the tire information acquisition unit and the tire prediction unit, set the classification thresholds for wear, tire pressure and damage and compare the thresholds respectively; when the same vehicle triggers multiple indicators at the same time, alarms are output according to the preset priority order and maintenance priorities are determined, and sorted based on weighted scores within the same priority; the alarm and priority determination unit is also used to implement continuous sampling duration and cooling time window constraints to suppress short-term repeated scheduling. The scheduling unit is used to generate scheduling instructions based on maintenance priorities and port operation plans, to schedule target AGVs to tire maintenance or inspection areas, and to synchronize the scheduling results to the production scheduling system. The maintenance execution unit is used to perform inflation / deflation operations and / or tire replacement operations after the AGV arrives at its position, and to generate maintenance result data after the operations are completed. The data synchronization unit is used to synchronize the updated tire status data and maintenance records to the backend processing end after the maintenance execution unit completes its work, and to provide the tire prediction unit with updates for subsequent prediction and alarm judgment.

[0039] In some embodiments, after receiving the work instructions issued by the scheduling unit, the maintenance execution unit performs on-site verification and alignment confirmation of the AGV based on the target wheel position and status information provided by the tire information acquisition unit. After confirming successful alignment, it automatically selects and executes either inflation / deflation or tire replacement operations. During the operation, the operation steps are completed in a preset order and the operation results are checked in real time. If an operation abnormality is detected, the abnormality information is fed back to the alarm and priority determination unit to trigger a re-inspection or manual review process.

[0040] In some embodiments, after completing inflation / deflation or tire replacement, the maintenance execution unit transmits the operation type, operation time, operation result, and corresponding tire status parameters back to the backend processing unit via the data synchronization unit. The tire prediction unit then uses this data to update the prediction input dataset and alarm judgment criteria. When the deviation between the transmitted data and the prediction result of the tire prediction unit exceeds a preset threshold, the alarm and priority judgment unit generates a model calibration prompt and notifies the tire prediction unit to perform the parameter re-estimation process.

[0041] In some embodiments, the maintenance execution unit is located in the tire maintenance area and is used to complete the automatic disassembly, assembly, inflation, and deflation of AGV tires; the maintenance execution unit includes: The system includes a hydraulic lifting device 21, a tire changing device 22, a used tire storage area 23, a slide rail assembly 24, and a new tire storage area 25. When the AGV needs tire maintenance, the vehicle management system dispatches the AGV to the tire maintenance area. Once the AGV is in position, the hydraulic lifting device 21 acts on the structural support points at the four corners of the AGV to lift the entire vehicle, lifting the tires off the ground and providing space for the tire changing operation. Subsequently, the tire changing device 22 automatically performs tire removal, installation, and inflation / deflation operations. Used tires are transported to the used tire storage area 23 via the slide rail assembly 24, while spare tires are retrieved from the new tire storage area 25 and installed onto the AGV axle, achieving automated tire changing.

[0042] The slide rail assembly 24 guides the movement of the tire changing device 22 in the lateral and longitudinal directions. Its structure comprises a lateral slide rail 31, a lateral base 32, a lateral platform 33, a lateral movement stop 34, a longitudinal slide rail 35, a longitudinal base 36, a longitudinal movement stop 37, and a tire support stop 38. The lateral slide rail 31 is fixed to the lateral base 32, which is fixed to the ground foundation. The lateral platform 33 is mounted on the lateral slide rail 31 and supports the tire changing device 22. The lateral slide rail platform 33 can move along the lateral slide rail 31 to the lateral movement stop 34 to assist the tire changing device 22 in moving between the used tire storage area 23 and the new tire storage area 25. The longitudinal slide rail 35 is fixed to both the lateral platform 33 and the longitudinal base 36, and the longitudinal slide rails 35 are connected end-to-end, allowing the tire changing device 22 to move back and forth within the range of the longitudinal slide rail 35. The longitudinal movement stop 37 limits the longitudinal travel and prevents overtravel. Tire bracket blocks 38 are installed at both ends of the waste tire storage area 23 and the new tire storage area 25. They are telescopically driven by an electronic control system to prevent tires from slipping when picking up or placing them.

[0043] The tire changing device 22 is used for automatic disassembly, installation, and inflation / deflation of AGV tires. Its structure consists of a lateral movement mechanism 41, a mechanism steering base 42, a longitudinal platform 43, a clamping cylinder 44, a tilting cylinder 45, a tire clamp 46, a camera 47, a bolt sleeve 48, and a press-fit tire inflation / deflation interface 49. The longitudinal platform 43 is mounted on a longitudinal slide rail 35 and guides the tire changing device 22 to move longitudinally. The mechanism steering base 42 is mounted on the longitudinal platform 43 and uses a hydraulic steering mechanism to adjust the direction of the tire changing device. The lateral movement mechanism 41 is mounted on the mechanism steering base 42 and drives the mounting frame to move laterally, achieving tire alignment. The clamping cylinder 44 drives the tire clamp 46 to perform clamping and releasing actions, and the tilting cylinder 45 drives the tire clamp 46 to perform tilting operations, thereby achieving posture adjustment during tire disassembly and installation. A camera 47 is mounted on the front end of the tire clamp 46 to detect the installation deviation between the tire and the axle. It calculates the alignment error using an image recognition algorithm and outputs a correction signal. A bolt sleeve 48 is used to remove and install the tire fixing bolts. A press-fit tire inflation / deflation interface 49 presses against the tire inflation nozzle 15 through a lateral movement to realize inflation and deflation operations.

[0044] The lateral rotation mechanism is installed inside the tire changing device 22 and consists of a hydraulic electrical cabinet 51, an air compressor cabinet 52, a lateral cylinder 53, a rotating gear 54, a rotating rack 55, a rotating cylinder 56, a rotating cable chain 57, a fixed connecting rod 58, and a hydraulic center rotary joint 59. The hydraulic electrical cabinet 51 is mounted on the top of the mechanism's steering base 42 and provides electrical and hydraulic power to the tire changing device 22; the air compressor cabinet 52 is fixed in the same position and provides compressed air for tire inflation / deflation operations. The lateral cylinder 53 drives the tire clamp 46 to move laterally, achieving tire alignment and changing station conversion. The rotating cylinder 56 drives the rotating rack 55 to move, which in turn drives the rotating gear 54 to rotate, and via the fixed connecting rod 58, drives the tire clamp 46 and the clamped tire to rotate, thus achieving precise alignment between the tire locking hole 14 and the axle mounting hole. The rotating cable drag chain 57 and the hydraulic center rotary joint 59 are used to achieve tangle-free transmission of hydraulic oil and electrical signals during rotation, ensuring continuous power supply and stable transmission of control signals.

[0045] During automated operation, after the AGV hydraulic lifting device 21 lifts the vehicle, the tire changing device 22 moves to the target tire position via the slide rail assembly 24. The tire clamp 46 automatically aligns under the guidance of the camera 47, and the clamping cylinder 44 and the tilting cylinder 45 work together to clamp and tilt the tire. The bolt sleeve 48 performs the installation and removal of the fixing bolts. The removed used tire moves to the used tire storage area 23 via the slide rail assembly 24 and is positioned by the tire bracket stop 38. Subsequently, the tire changing device 22 retrieves a new tire from the new tire storage area 25 and precisely installs it via the action of the lateral movement mechanism 41 and the rotating gear 54. After installation, the press-fit tire inflation / deflation interface 49 performs the inflation operation, and finally, the hydraulic lifting device 21 lowers, completing the automated tire changing operation.

[0046] In some embodiments, this application provides a terminal, including: Memory for storing tire management programs for port AGVs; A processor is configured to implement the steps of the tire management method for port AGVs when executing the tire management system for port AGVs.

[0047] In some embodiments, this application provides a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes the tire management method for port AGVs.

[0048] The above description is merely a preferred embodiment of one or more embodiments of this specification and is not intended to limit the scope of one or more embodiments of this specification. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of one or more embodiments of this specification should be included within the protection scope of one or more embodiments of this specification.

Claims

1. A tire management method for port AGVs, characterized in that, Includes the following steps: Step S1: Obtain tire information for all AGVs in the fleet. Tire information includes tire pressure, wear, and damage information. Tire pressure information is collected by tire pressure sensors installed on the tires; Wear information is obtained by identifying the distance between the tire tread and the wear reference line using cameras installed in the automated operation area. Image recognition algorithms are used to determine the condition of the tire body to obtain information on damage. And upload the tire information to the fleet tire management terminal; Step S2: When the AGV passes through the automated operation area, the camera and the tire pressure sensor of the AGV will detect the AGV. When the tire wear exceeds the limit, the tire pressure is abnormal or the tire is damaged, a first-level alarm message will be generated. Otherwise, step S3 will be executed to predict the tires of the AGV. Step S3: Establish a tire prediction model based on the historical mileage and load information of the AGV, and perform alarm prediction on the tires according to the tire prediction model and tire information. When the time corresponding to the alarm prediction is reached, generate a secondary alarm message. The maintenance priority of each AGV tire is determined based on the degree of abnormality of the alarm information. Step S4: Based on the maintenance priority information and the port operation plan, dispatch the AGVs that need maintenance to the tire maintenance area. When the AGV requiring maintenance arrives at the tire maintenance area, it will perform tire inflation / deflation and / or tire replacement operations. Once completed, the updated tire status data will be synchronized to the fleet tire management terminal.

2. The tire management method for port AGVs according to claim 1, characterized in that, In step S3, the AGV operation process is divided into several continuous wear intervals with constant load according to the detection time sequence, and the triplet of each interval is obtained. ; in, Let be the mileage of the i-th interval. Let be the load constant for the i-th interval. This represents the measured wear amount within this range; Based on the aforementioned triplet, a functional relationship between wear amount and mileage and load is established, and the parameters are estimated using the least squares method. The functional model is as follows: These are the preset model parameters; To obtain the cumulative wear and tear at the current moment. Wear threshold Then, based on the representative interval characteristics obtained from the most recent intervals... ; in, For the expected single-section driving distance, For the expected single-section load, Expected duration of a single interval, The weighted average or median of the corresponding values ​​from the most recent intervals can be used; Calculate the predicted wear amount for a single interval and determine that it meets the requirements. Find the smallest positive integer n to obtain the estimated time to reach the wear threshold: in, The current moment; A level 2 alarm message is generated when the time corresponding to the alarm prediction is reached.

3. The tire management method for port AGVs according to claim 1, characterized in that, In step S1, a defect identification model is established based on the acquisition of damage information. The defect identification model is trained and used for online identification in the following way: A dataset was constructed and labeled for tire images collected in the automated operation area. The defect categories included sidewall cracks, bulges, cuts, foreign object embedding, and shoulder defects. The image is subjected to distortion correction and scaling. The scaling is achieved based on the pixel distance between the tire tread and the wear reference line or the known geometric dimensions of the rim mounting hole. A supervised learning approach is used to train a probabilistic output model that integrates classification and localization to obtain the confidence scores for defect location and category.

4. The tire management method for port AGVs according to claim 1, characterized in that, In step S3, the alarm and maintenance priority determination includes: setting the classification thresholds for wear, tire pressure and puncture and comparing the thresholds respectively; When multiple indicators are triggered simultaneously on the same vehicle, the alarm and maintenance priority are determined in the order of damage, wear, and tire pressure. Within the same priority, the scores are ranked based on a weighted score, which is composed of wear proximity, tire pressure deviation, defect confidence or defect area ratio, and cumulative mileage or time since the last maintenance, and is obtained by summing after weight normalization. An alarm determination strategy is adopted, and an alarm is output only when continuous sampling meets the grade threshold or the duration of exceeding the limit is not less than the preset duration; and a cooling time window is set to suppress short-term repeated scheduling.

5. The tire management method for port AGVs according to claim 2, characterized in that, It also includes the following steps: Step S5: Each time a tire passes through the camera detection area of ​​the automated operation zone, the tire prediction model is corrected using the tire information. For model parameters The update process includes: forming a sample set from the K most recent wear intervals with unchanged load, and constructing a sample matrix: in For the first The driving distance of each section, For the first The load constant of each interval No. Measured wear amount in each interval; The parameter vector is calculated using the weighted least squares method: Here is the weight matrix, and the weights are: The time decay factor, For the first The confidence coefficient of each interval detection; The corrected prediction is obtained using the updated parameters: Used for subsequent wear prediction and alarm determination.

6. A tire management system for port AGVs, used to implement the tire management method for port AGVs as described in any one of claims 1-5, characterized in that, The system includes: The tire information acquisition unit is used to acquire tire information of all AGVs in the fleet. The tire information includes at least tire pressure information, wear information, and damage information. The tire information acquisition unit includes tire pressure sensors installed on the tires, cameras installed in the automated operation area, and an interactive interface for manual input. The acquired tire information is then uploaded to the back-end processing terminal through a communication network. The tire prediction unit constructs a tire wear prediction model based on the AGV's historical mileage, load information, and wear detection results. It divides the continuous time period with constant load into several wear intervals, obtains triplet data for each wear interval, and establishes a functional relationship between wear amount and mileage and load through a regression algorithm. When the latest wear data detected by the camera or by manual inspection is obtained, the updated samples are used to re-estimate the model parameters to complete the model correction. Based on the current cumulative wear amount and wear threshold, the model is called to calculate the estimated time required for the tire wear amount to reach the wear threshold, and the tire life end prediction result is output for the alarm and priority determination unit to perform subsequent alarm and scheduling control. The alarm and priority determination unit is used to comprehensively determine the results from the tire information acquisition unit and the tire prediction unit, set the classification thresholds for wear, tire pressure and damage and compare the thresholds respectively; when the same vehicle triggers multiple indicators at the same time, alarms are output according to the preset priority order and maintenance priorities are determined, and sorted based on weighted scores within the same priority; the alarm and priority determination unit is also used to implement continuous sampling duration and cooling time window constraints to suppress short-term repeated scheduling. The scheduling unit is used to generate scheduling instructions based on maintenance priorities and port operation plans, to schedule target AGVs to tire maintenance or inspection areas, and to synchronize the scheduling results to the production scheduling system. The maintenance execution unit is used to perform inflation / deflation operations and / or tire replacement operations after the AGV arrives at its position, and to generate maintenance result data after the operations are completed. The data synchronization unit is used to synchronize the updated tire status data and maintenance records to the backend processing end after the maintenance execution unit completes its work, and to provide the tire prediction unit with updates for subsequent prediction and alarm judgment.

7. The tire management system for port AGVs according to claim 6, characterized in that, After receiving the work instructions from the scheduling unit, the maintenance execution unit performs on-time verification and alignment confirmation of the AGV based on the target wheel position and status information provided by the tire information acquisition unit. After confirming successful alignment, it automatically selects and executes either inflation / deflation or tire replacement operations. During the operation, it completes the work steps in a preset order and verifies the work results in real time. If an operation abnormality is detected, it reports the abnormality information to the alarm and priority determination unit to trigger a re-inspection or manual review process.

8. The tire management system for port AGVs according to claim 7, characterized in that, After completing inflation / deflation or tire replacement operations, the maintenance execution unit transmits the operation type, operation time, operation results, and corresponding tire status parameters back to the backend processing unit via the data synchronization unit. This data is then used by the tire prediction unit to update the prediction input dataset and alarm judgment criteria. When the deviation between the transmitted data and the prediction results of the tire prediction unit exceeds a preset threshold, the alarm and priority judgment unit generates a model calibration prompt and notifies the tire prediction unit to execute the parameter re-estimation process.

9. A terminal, characterized in that, include: Memory for storing tire management programs for port AGVs; A processor is configured to implement the steps of the tire management method for a port AGV as described in claim 1 when executing the tire management device for a port AGV.

10. A computer-readable storage medium, characterized in that, The storage medium stores computer instructions. When the computer reads the computer instructions from the storage medium, the computer executes the tire management method for port AGVs as described in claim 1.

Images