Full-automatic folding and unfolding and vehicle weighing method for portable wagon balance

Abstract

The invention discloses a portable wagon balance full-automatic folding and unfolding and vehicle weighing method, and relates to the technical field of intelligent weighing equipment and automatic control, and the method comprises the following steps: S1, recognizing a license plate, checking vehicle parameters, and assigning AGV trolleys with corresponding axle numbers; s2, enabling the trolleys to carry the wagon balance formation to pre-position at the bottom of the vehicle; s3, accurately aligning the trolley and placing a wagon balance; s4, guiding the vehicle to roll the wagon balance, and acquiring and processing data to obtain axle load and total weight; and S5, after weighing, guiding the vehicle to drive away from the area, recovering the wagon balance by the trolley, and waiting for the system. Vehicle parameters are obtained by means of roadside visual recognition, wagon balance alignment placement and recovery are cooperatively completed in combination with the AGV, manual operation is not needed, multi-axle vehicle weighing can be completed by a single person, single-time operation time is greatly shortened, manual weighing data errors can be reduced, meanwhile, the vehicle parameters are complemented through the laser radar, the AGV adapts to the complex environment, and the working efficiency is improved. Safety is guaranteed through wagon balance self-checking leveling, vehicle speed monitoring and the like, and the system is suitable for multiple scenes.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of intelligent weighing equipment and automation control technology, in particular to a portable full-automatic receiving and releasing of a weighbridge and a vehicle weighing method. BACKGROUND

[0002] In the fields of logistics transportation, highway oversize detection, and construction site material transportation, vehicle load weighing is a key link to ensure transportation safety, standardize loading standards, and account for logistics costs. Currently, the industry mainly relies on two types of equipment, fixed weighbridges and traditional portable weighbridges, to carry out weighing operations. However, both types of equipment have significant technical defects and scene adaptation limitations in actual application, making it difficult to meet the modern weighing requirements of efficiency, precision, and safety.

[0003] For fixed weighbridges, they need to be fixedly installed on pre-planned hardened ground, which not only has a long construction period and high initial investment cost, but also cannot be moved due to fixed equipment location. This completely cannot adapt to the mobility requirements of temporary freight stations, temporary weighing at construction sites, and emergency material transportation weighing. At the same time, fixed weighbridges usually require vehicles to completely drive onto the weighbridge to complete the weighing. For long-wheelbase, multi-axle large vehicles, sufficient access channels need to be reserved, which has strict requirements on the space and is greatly limited in narrow working environments.

[0004] While traditional portable weighbridges have some mobility, their operation process completely relies on manual operation, which has multiple pain points. First, the efficiency of weighbridge transportation and deployment is extremely low. The weight of a single portable weighbridge usually reaches dozens of kilograms, and multiple weighbridges are needed for multi-axle vehicle weighing, which requires 2-3 people to cooperate to carry and deploy the weighbridges to the designated location. The distance between the weighbridges needs to be adjusted manually to match the wheelbase and track of the vehicle, and the preparation time for single multi-axle vehicle weighing often exceeds 40 minutes, which seriously affects the transportation efficiency. Second, the weighing accuracy cannot be guaranteed. Manual adjustment of the weighbridges cannot accurately align the vehicle tire rolling track, which may cause weighing data deviation due to the deviation of the weighbridges. Manual recording of weighing data may also result in human errors, which cannot meet the high-precision weighing requirements. Third, the adaptability to vehicles without pre-set information is poor. Key parameters such as the number of vehicle axles and the wheelbase need to be measured manually before operation. If the vehicle has complex types and unknown parameters, the measurement process is tedious and prone to errors, further prolonging the operation time. Fourth, the safety risk is prominent. During the vehicle rolling process, there is no effective vehicle speed monitoring and guiding mechanism, making it difficult for the driver to determine the optimal driving speed and parking position, which may cause damage to the weighbridge due to vehicle overspeed rolling. There is no levelness detection and calibration mechanism for the weighbridge after deployment. If the ground is uneven, the weighbridge may tilt, which not only increases the weighing error but also may cause safety accidents such as tire skidding and weighbridge overturning during vehicle rolling. Manual deployment of the weighbridge around the vehicle also poses a safety risk of being scratched by the vehicle.

[0005] Furthermore, existing technologies lack solutions that enable automatic weighbridge loading and unloading, precise alignment, and collaborative operation. They cannot replace manual labor in core processes such as weighbridge handling, positioning, and calibration. They also lack integrated designs for automatic vehicle parameter recognition, dynamic monitoring of the weighing process, and equipment safety protection. As a result, traditional weighing methods have significant shortcomings in efficiency, accuracy, safety, and scenario adaptability, which seriously restricts the modernization and upgrading of vehicle weighing operations. There is an urgent need for a new weighing technology solution that can overcome these limitations. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a portable, fully automated weighbridge deployment and vehicle weighing method. This invention uses roadside visual recognition to obtain vehicle parameters and combines it with AGV carts to complete the deployment and alignment of the weighbridge. No manual operation is required, and a single person can complete the weighing of multiple axles of vehicles. The time for a single operation is greatly shortened, and the error of manual weighing data is reduced. At the same time, it uses LiDAR to supplement vehicle parameters, the AGV carts adapt to complex environments, and the weighbridge self-checking and leveling, vehicle speed monitoring and other functions ensure safety. It is suitable for multiple scenarios.

[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution: On one hand, a fully automatic portable weighbridge deployment and retrieval method for vehicle weighing, the method comprising the following steps:

[0008] S1. Vehicle Recognition and Task Planning: The roadside vision recognition unit captures images of the vehicle to be tested, identifies its license plate information, and queries a pre-set vehicle information database to obtain the vehicle's number of axles, wheelbase, and key wheelbase parameters. The central control unit dynamically calculates and assigns the number of AGVs (Automated Guided Vehicles) required to perform the weighing task based on the key parameters, and generates a collaborative task instruction set containing the target position sequence. The number of AGVs is equal to the number of axles of the vehicle, and each AGV is responsible for the weighbridge operation at the position corresponding to one axle.

[0009] S2, Weighbridge Collaborative Transport and Coarse Positioning: Multiple AGVs are automatically driven out from the base station. Each AGV uses its onboard gripping robotic arm to grab and fix a portable weighbridge. The multiple AGVs autonomously navigate in a longitudinal formation, moving from the side or rear of the vehicle to be tested to the area directly below it, forming an initial working array.

[0010] S3. Precise Alignment and Placement of Weighbridge: Based on the vehicle parameters obtained in S1, each AGV first performs coarse positioning using an odometer and inertial measurement unit. Then, the stereo vision modules mounted on both sides of each AGV are activated to perform real-time image acquisition and 3D reconstruction of the ground contact features of the front and rear tires on the same side of the vehicle, accurately calculating the real-time center coordinates and orientation of the tires. According to the calculation results, the central control unit coordinates and schedules each AGV to fine-tune its posture, so that the weighbridge it carries moves precisely to the preset weighing preparation position. The weighing preparation position meets the following requirements: the upper surface of the weighbridge is flush with the ground, and its center point is located at a predetermined safe distance directly in front of the theoretical rolling trajectory of the corresponding tire. After all the weighbridges are in place, the robotic arms of each AGV perform the weighbridge lowering action, placing it stably on the ground. Then, the robotic arms are decoupled from the weighbridges and retracted.

[0011] S4. Vehicle Guidance and Weighing Data Acquisition: The central control unit issues a visual and / or audible forward movement command to the driver of the vehicle under test; the vehicle under test moves forward at a low and constant speed, so that all its tires successively roll over the corresponding weighbridges; the weighing sensors built into each weighbridge collect the weight signal under the load of that axle, and send the weight data in time series to the central control unit through the wireless transmission module; the central control unit filters and integrates the received weight data of each axle to obtain and record the weight of each axle and the total weight of the vehicle;

[0012] S5. Weighbridge Cooperative Retrieval and Return: After weighing is completed, the central control unit issues a reversing command to the driver; after the vehicle tires have completely exited the weighbridge bearing surface, the robotic arms of each AGV trolley extend again and clamp and fix the corresponding weighbridge; under the scheduling of the central control unit, all AGV trolleys carry the weighbridge and autonomously exit from under the vehicle in a formation, returning to the designated parking position at the base station; after the vehicles leave, the system returns to standby status.

[0013] Furthermore, in step S1, the vehicle information database is a relational database stored in the cloud or locally, which is accessed through license plate number association and at least pre-stores the vehicle's number of axles, number of tires per axle, standard wheelbase, and wheelbase information; if the query is unsuccessful, the central control unit starts the backup lidar point cloud scanning process, and calculates the number of axles, wheelbase, and wheelbase of the vehicle under test in real time through point cloud clustering and geometric feature extraction algorithms.

[0014] Furthermore, the precise alignment and placement described in step S3 are specifically implemented as follows:

[0015] S31. Based on the stereo vision module, continuously acquire image pairs of the vehicle tire sidewall and the ground contact area;

[0016] S32. Extract feature points and perform stereo matching on the image pairs. Combine the calibrated camera intrinsic and extrinsic parameters to calculate the three-dimensional coordinates of the feature points in the robot coordinate system.

[0017] S33. By fitting the geometric center of the tire contact patch and combining it with the tire radius parameter, the precise horizontal coordinates (X, Y) of the tire center point and the normal vector of the tire plane are calculated.

[0018] S34. The central control unit calculates the target position for placing the weighbridge based on the tire center coordinates, normal vector, and preset vehicle movement direction; the target position ensures the center point coordinates of the weighbridge ( , The following relationship must be satisfied:

[0019]

[0020] in, The forward offset is set according to the vehicle type and safety margin;

[0021] S35. Each AGV trolley moves autonomously to its position according to the received target location, and completes the final placement of the weighbridge through the micro-operation capability of the robotic arm, ensuring that the upper surface of the weighbridge is flush with the ground and the longitudinal center line of the weighbridge is perpendicular to the vehicle's driving direction.

[0022] Furthermore, the forward offset in step S34 Its value is dynamically adjusted based on the vehicle wheelbase queried from the database or measured in real time. The adjustment strategy is defined by the following formula:

[0023]

[0024] in:

[0025] A proportional coefficient set according to vehicle type, and 0 < <1;

[0026] The wheelbase of the vehicle under test is expressed in meters (m).

[0027] This is a fixed constant set based on a safety margin, and the unit is meters (m).

[0028] Furthermore, the lurking autonomous mobile robot (AGV) adopts an omnidirectional mobile chassis, and its driving method is a combination of Mecanum wheels or omnidirectional wheels; the gripping robotic arm is an electric servo robotic arm with at least 4 degrees of freedom, and its end effector is an adaptive gripper, which can be compatible with gripping portable weighbridges with different thicknesses and frame structures; the stereo vision module includes two symmetrically arranged global shutter cameras, and their baseline distance has been optimized to meet the requirements of accurate distance measurement in low-light environments under the vehicle.

[0029] Furthermore, in step S4, the central control unit, while issuing the forward command, initiates a vehicle speed monitoring subprocess based on the roadside vision recognition unit or the AGV vehicle vision module; if the vehicle's forward speed is detected to exceed a preset threshold, a deceleration warning is immediately issued to the driver, and valid weighing data is only recorded after the speed drops to a safe range.

[0030] Furthermore, after the weighbridge is placed in step S3 and before the vehicle moves forward in step S4, a weighbridge self-check and level calibration step is inserted: the tilt sensors built into each weighbridge detect the levelness of the weighbridge after it is placed. If the levelness deviation exceeds the allowable range, the nearest AGV trolley will re-clamp the weighbridge and perform a second leveling and placement, or the central control unit will calculate the levelness deviation compensation amount and perform software compensation in the subsequent weighing data.

[0031] On the other hand, an automatic weighing system for fully automatic loading and unloading of portable weighbridges and vehicle weighing methods includes:

[0032] A central control unit, acting as the brain of the system, is responsible for task scheduling, data fusion, and decision-making;

[0033] A roadside vision recognition unit is fixedly installed above the entrance to the detection area for vehicle capture and license plate recognition;

[0034] A vehicle information database used to store and provide vehicle structural parameters;

[0035] At least two stealthy autonomous mobile robots (AGVs) constitute the main execution body of the system;

[0036] A portable weighbridge matching the number of AGV vehicles;

[0037] A base station provides parking, charging, and weighbridge storage functions for the AGV vehicles; wherein, the central control unit establishes real-time and reliable wired and / or wireless communication connections with the roadside vision recognition unit, vehicle information database, all AGV vehicles, and all portable weighbridges.

[0038] Furthermore, the portable weighbridge has undergone adaptive structural improvements, including:

[0039] A standardized gripping mechanism matching the gripper of the AGV robot arm is provided on its side or corner;

[0040] It has a built-in tilt sensor, wireless communication module and high-precision weighing sensor;

[0041] The weighbridge has a high-friction, wear-resistant coating on its upper surface to increase the adhesion of tires when they roll over it.

[0042] Compared with existing technologies, this fully automatic portable weighbridge deployment and vehicle weighing method has the following advantages:

[0043] I. This invention automatically acquires vehicle parameters through roadside visual recognition and, in conjunction with AGV carts, completes the weighbridge placement and alignment. There is no need for manual handling of the weighbridge, manual positioning, or operation of the weighing equipment. A single person can complete the weighing process for multi-axle vehicles. Compared with the traditional manual assisted weighing mode, the time for a single weighing operation is greatly shortened, while avoiding the problem of inaccurate weighing data caused by human operation errors.

[0044] Second, this invention supplements vehicle parameter acquisition through lidar point cloud scanning, solving the weighing problem of vehicles without pre-set information. The AGV trolley adopts an omnidirectional mobile chassis, which can operate flexibly in narrow spaces and complex road environments. At the same time, the weighbridge's self-checking and leveling mechanism ensures the stability of the weighing benchmark, and the vehicle speed monitoring and safety distance design prevent collisions or rollovers when vehicles run over the weighbridge. In addition, the high-friction cover layer of the weighbridge prevents tire slippage, comprehensively ensuring the safety of personnel, equipment and vehicles during the weighing process. It is suitable for various scenarios such as logistics parks, construction sites, and highway checkpoints.

[0045] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0047] Figure 1 A flowchart of a fully automated portable weighbridge deployment and vehicle weighing method;

[0048] Figure 2 This is a flowchart illustrating the precise placement and self-inspection process of a portable weighbridge with fully automated loading, unloading, and vehicle weighing capabilities. Detailed Implementation

[0049] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0050] I. System Overall Architecture and Component Parameter Settings

[0051] The automatic weighing system in this embodiment is centered on a central control unit, and works in conjunction with roadside vision recognition units, vehicle information databases, AGV (Automated Guided Vehicle) robots, portable weighbridges, and base stations to form a complete fully automatic weighing closed loop. The specific parameters and structural designs of each component are as follows:

[0052] (a) Central control unit

[0053] The central control unit adopts an industrial-grade embedded controller, equipped with a multi-core processor with a main frequency of no less than 3.0GHz, and features no less than 16GB of RAM and a 512GB high-speed solid-state drive. This ensures the computing speed and data storage capacity during multi-task parallel processing. It can simultaneously process motion control commands for multiple AGV vehicles, collect and analyze multiple sets of weighing data, and coordinate communication between units.

[0054] This controller supports Ethernet, 4G / 5G, and Wi-Fi 6 wireless communication protocols, enabling real-time communication with roadside vision recognition units, vehicle information databases, all AGVs, and portable weighbridges. Communication latency is controlled within 100ms, ensuring timely command transmission and data feedback. It features a built-in, self-developed weighing task scheduling system based on the Linux operating system, integrating modules for task planning, motion control, data acquisition and processing, and anomaly alarms.

[0055] The task planning module can dynamically generate collaborative operation instructions for AGVs based on parameters such as the number of axles, wheelbase, and track width, including the driving path, formation method, and weighbridge placement location.

[0056] The motion control module supports formation navigation and pose fine-tuning of multiple AGVs, ensuring the motion accuracy of the vehicles by correcting driving deviations in real time.

[0057] The data processing module uses the Kalman filter algorithm to reduce noise in the weight signal and combines it with integral calculation to obtain accurate axle weight and total weight, with the weight data error controlled within ±0.5%.

[0058] The abnormal alarm module can monitor the equipment's operating status in real time. When problems such as communication interruption, abnormal weighing data, or excessive deviation of the trolley movement occur, it will immediately trigger an audible and visual alarm and record the fault information.

[0059] (ii) Roadside visual recognition unit

[0060] The roadside visual recognition unit is fixedly installed above the entrance of the detection area, at a height of 4.0-5.0m above the ground. Its core components include a high-resolution global shutter industrial camera, a fixed-focus lens, and an adaptive lighting module. The camera resolution is no less than 1920×1080, with a frame rate set to 25-30fps, enabling it to clearly capture vehicle images in both natural daylight and low-light conditions. The lens focal length is matched to the installation height to ensure coverage of the critical field of view when a vehicle enters the detection area, with no distortion in the license plate recognition area.

[0061] The supplementary lighting module uses an LED array light source with a color temperature adjustment range of 5000-6000K. The light intensity can automatically switch according to the ambient brightness. At night or in rainy weather, it can accurately supplement the license plate area, avoiding overexposure or underexposure, and ensuring a license plate recognition accuracy of no less than 99.5%. This unit has a built-in deep learning-based license plate recognition algorithm that supports the recognition of various license plate types, including blue plates, yellow plates, and new energy vehicle plates. The time from image acquisition to outputting the license plate number is less than 0.3 seconds. It also integrates a vehicle contour detection function, which determines whether a vehicle has entered the detection area by recognizing the vehicle's front contour. When the front of the vehicle crosses the entrance detection line, the vehicle recognition process is automatically triggered without manual intervention.

[0062] (III) Vehicle Information Database

[0063] The vehicle information database uses a MySQL relational database, supporting both local server storage and cloud backup modes to ensure data security and availability. The local server is equipped with dual processors, at least 64GB of RAM, and over 4TB of hard drive space to ensure fast database read and write speeds, with a single data retrieval response time of no more than 0.1 seconds. Cloud storage uses object storage services from mainstream cloud providers to achieve off-site data backup and regular automatic synchronization of local data to prevent data loss due to local hardware failures.

[0064] Each vehicle record in the database contains key parameters such as license plate number, number of axles, number of tires per axle, standard track width, and wheelbase, covering common vehicle models from 2 to 6 axles. Track widths range from 1.8 to 2.5 meters, and wheelbases range from 3.2 to 12 meters, meeting the parameter query needs of different vehicle types, including sedans, buses, and freight vehicles. The database supports precise searching by license plate number. When a license plate number is entered, the corresponding vehicle parameters are quickly matched and returned. If the query is unsuccessful (e.g., the license plate number is not entered into the database), a backup LiDAR scanning process is automatically triggered to obtain the vehicle parameters.

[0065] (iv) AGV trolley

[0066] The AGV (Automated Guided Vehicle) trolley, as the main actuator of the system, adopts an omnidirectional mobile chassis and is driven by Mecanum wheels or a combination of omnidirectional wheels. The wheel load capacity is no less than 500 kg / wheel, and the overall rated load of the trolley is no less than 1000 kg, meeting the transportation requirements of portable weighbridges. The trolley's dimensions are controlled within 800mm × 600mm × 350mm (length × width × height), with a minimum turning radius of 0mm. It can rotate in place and translate in any direction, flexibly adapting to the movement needs of narrow spaces under the vehicle and avoiding collisions with vehicle chassis components.

[0067] The robotic arm mounted on the vehicle is an electrically driven servo arm with at least 4 degrees of freedom, a maximum load of no less than 5 kg, and a repeatability of no less than ±0.02 mm, ensuring precise gripping and placement of the weighbridge. The end effector of the robotic arm is an adaptive gripper with an opening range of 0-150 mm, compatible with portable weighbridges with a thickness of 50-100 mm and a frame width of 30-80 mm. The inner side of the gripper is equipped with a rubber anti-slip pad with a friction coefficient of no less than 0.8 to prevent the weighbridge from slipping during gripping and to avoid scratching the surface of the weighbridge.

[0068] The vehicle is equipped with a stereo vision module consisting of two symmetrically arranged global shutter cameras. These cameras have a resolution of at least 2592×1944 and a frame rate of at least 14fps. The baseline distance is optimized (typically 250-350mm), and after calibration, the ranging accuracy reaches ±2mm within the 0.5-3m range. The camera lenses are fixed-focus lenses equipped with infrared filters, enabling normal operation in low-light conditions under the vehicle (light intensity <50 lux), clearly capturing images of the tire sidewalls and the contact patch area. In addition, the vehicle is also equipped with a high-precision odometer (resolution at least 0.1mm) and an inertial measurement unit (IMU). Its acceleration measurement range is ±16g, and its angular velocity measurement range is ±2000° / s, used for coarse positioning to lay the foundation for subsequent precise alignment.

[0069] (v) Portable weighbridge

[0070] The portable weighbridge is made of high-strength aluminum alloy, balancing structural strength and lightweight design. Its overall dimensions are 1200mm × 600mm × 80mm (length × width × height), and its weight does not exceed 50kg, facilitating gripping and transport by AGV trolleys. It has a rated weighing capacity of no less than 20,000kg and a maximum overload capacity of 120% of the rated load, capable of withstanding the axle load pressure of freight vehicles and resisting deformation over long-term use. Standardized gripping mechanisms with U-shaped grooves are installed on the sides or corners of the weighbridge. The width and depth of the grooves are matched to the size of the AGV trolley's grippers, ensuring a secure grip without loosening.

[0071] The weighbridge has multiple sets of high-precision sensors and communication modules built in:

[0072] Inclination sensor: Measurement range ±15°, accuracy ±0.1°, real-time detection of the levelness of the weighbridge after placement, providing data support for level calibration;

[0073] Wireless communication module: Supports LoRa or other low-power wide area network protocols, with a communication distance of over 100m and a transmission rate of no less than 50kbps, ensuring stable transmission of weight data and status information to the central control unit;

[0074] Weighing sensors: C3-level high-precision sensors with a weighing range of no less than 5000kg are used. Four points are evenly distributed at the four corners of the bottom of the weighbridge to ensure uniform weight signal acquisition and reduce errors caused by uneven local force.

[0075] The upper surface of the weighbridge is covered with a high-friction coefficient wear-resistant coating made of polyurethane or rubber, with a thickness of not less than 5mm and a friction coefficient of not less than 1.2. This effectively increases the adhesion of tires when they roll over the weighbridge, prevents vehicles from slipping, and improves the wear resistance of the weighbridge surface, thus extending its service life.

[0076] (vi) Base stations

[0077] The base station is constructed with a steel frame, and its overall dimensions are designed according to the number of AGVs and weighbridges, typically 5m × 3m × 2.5m (length × width × height). Internally, it is divided into an AGV parking area, a weighbridge storage area, and a charging area, with clear functional zoning for easy equipment management and maintenance. The parking area has multiple parking spaces (the number matching the number of AGVs), each equipped with a positioning guidance device (such as a QR code or infrared locator) to guide the AGVs to park accurately, with a parking deviation of no more than 10mm. The weighbridge storage area has a corresponding number of storage racks, each equipped with a fixing device to prevent the weighbridge from shifting or tipping over during storage. The charging area uses wireless charging, with a wireless charging transmitter coil installed under each parking space, with an output power of no less than 1000W. Charging automatically starts after the AGV stops, taking approximately 2 hours. A fully charged AGV can operate continuously for more than 8 hours, meeting the needs of long-term operation.

[0078] The base station is also equipped with an environmental monitoring module to monitor the internal temperature, humidity and dust concentration in real time. When the temperature exceeds 40°C or the humidity exceeds 80%, the ventilation and heat dissipation or dehumidification equipment will be automatically activated. When the dust concentration exceeds the set threshold, the dust removal device will be activated to ensure the stable operation environment of the equipment and avoid environmental factors from affecting the performance and lifespan of the equipment.

[0079] II. Detailed Operating Procedures for the Fully Automatic Deployment and Retraction of Portable Weighbridges and Vehicle Weighing Methods

[0080] This implementation method uses a 4-axle freight vehicle as an example to illustrate in detail the operation steps of a portable weighbridge's fully automatic deployment and retrieval and vehicle weighing method, covering the entire process of vehicle identification, weighbridge transport, precise alignment, weighing data acquisition, and weighbridge retrieval, as detailed below:

[0081] (a) Step S1: Vehicle recognition and task planning

[0082] When a 4-axle freight vehicle drives to the entrance of the inspection area, the industrial camera of the roadside vision recognition unit starts capturing images as the front of the vehicle crosses the entrance inspection line. The supplementary lighting module automatically adjusts the light intensity according to the current ambient brightness to ensure that the captured vehicle image is clear and that the license plate area is unobstructed and glare-free.

[0083] The license plate recognition algorithm extracts the license plate area from the image, removes image noise and interference information through preprocessing steps such as grayscale conversion, edge detection, and character segmentation, and then determines the license plate number through a character recognition model and sends the recognition result to the central control unit.

[0084] After receiving the license plate number, the central control unit immediately sends a query request to the vehicle information database. The request includes the target license plate number and the required parameter types (number of axles, track width, wheelbase, etc.). Upon receiving the request, the database quickly retrieves the corresponding vehicle record using an index. If a matching record is found, it extracts parameters such as the number of axles (4 axles), number of tires per axle (2), track width (2.2m), and wheelbase (6.5m) and sends them back to the central control unit. If no matching record is found, it returns a "query unsuccessful" signal to the central control unit.

[0085] The central control unit determines the number of AGVs based on the number of axles, following the principle of "one AGV per axle." This task requires the assignment of four AGVs, each responsible for the weighbridge operation at the corresponding axle position. Simultaneously, combining wheelbase and track width parameters, a kinematic model calculates the target position sequence for each AGV, including the convoy driving path after leaving the base station (avoiding obstacles and ensuring driving safety), the initial working array position under the vehicle (lateral deviation from the corresponding axle not exceeding 100mm), and the initial coordinates of the weighbridge placement. This generates a collaborative task instruction set containing driving speed, steering angle, and action timing, which is then transmitted wirelessly to the four AGVs.

[0086] If a "query unsuccessful" signal is received from the database, the central control unit immediately initiates the backup LiDAR point cloud scanning process. The LiDAR is installed next to the roadside visual recognition unit, with a 360° scanning range, a ranging accuracy of ±3cm, and a scanning frequency of no less than 10Hz. The LiDAR performs a full-range scan of the vehicle, acquiring its 3D point cloud data. A point cloud clustering algorithm separates the vehicle's point cloud from the background point cloud, removing interfering point clouds such as those from the ground and trees. Then, a geometric feature extraction algorithm is used to identify the geometric shape and positional relationship of the wheels and axles, calculating the vehicle's number of axles, wheelbase, and track width. The entire scanning and calculation process takes no more than 10 seconds, ensuring the smooth progress of subsequent task planning and not affecting weighing efficiency.

[0087] (II) Step S2: Weighbridge Co-transportation and Coarse Positioning

[0088] After receiving the collaborative task instruction set, the four AGVs automatically drive out of their corresponding parking spaces in the base station parking area. During the driving process, the vehicles record the distance traveled by the odometer, and combine the acceleration and angular velocity data collected by the inertial measurement unit to calculate their own position and attitude in real time. The PID control algorithm is used to adjust the wheel speed and steering to ensure that the deviation of the driving path does not exceed 50mm, so as to avoid collisions between vehicles or deviation from the preset route.

[0089] After the trolley drives to the weighbridge storage area, it stops in front of the corresponding weighbridge storage rack according to the instructions. The robotic arm starts according to the preset program: First, the joint angle of the robotic arm is adjusted so that the end gripper is aligned with the standardized clamping mechanism on the side of the weighbridge. The gripper slowly closes. When the clamping force reaches the preset value (50N), it stops closing and maintains the clamping state to ensure that the weighbridge and the robotic arm are firmly connected without any loosening or shaking. After clamping is completed, the robotic arm drives the weighbridge to lift slightly (the lifting height does not exceed 10mm) to avoid friction between the weighbridge and the storage rack.

[0090] After the four AGVs are clamped by the weighbridge, they travel in a longitudinal formation specified by the central control unit: a safe distance of 1.5-2.0m is maintained between each AGV. The leading AGV transmits its position and speed information to the following AGVs via wireless communication, and the following AGVs adjust their driving status based on the information from the leading AGV to ensure a neat formation. The driving route extends from the side of the vehicle being tested (0.3-0.5m away from the vehicle body to avoid collision) to the area directly under the vehicle, with the driving speed controlled at 0.3-0.5m / s to ensure smooth movement.

[0091] After reaching the area under the vehicle, the four AGVs adjust their positions according to the instructions of the central control unit to form an initial working array: the lateral deviation of the position of each AGV from the theoretical position of the corresponding axis does not exceed 100mm and the longitudinal deviation does not exceed 200mm, leaving room for adjustment for the subsequent precise alignment of the weighbridge; after the array is formed, the AGVs send a "positioned" signal to the central control unit and wait for the next instruction.

[0092] (III) Step S3: Precise alignment and placement of the weighbridge

[0093] 1. Coarse positioning stage

[0094] The four AGVs first perform coarse positioning based on odometers and inertial measurement units: the odometers record the number of wheel rotations in real time and convert them into travel distance; the inertial measurement units collect the acceleration and angular velocity of the vehicles and obtain the displacement and attitude changes of the vehicles through integration calculations; the central control unit fuses the two sets of data, uses the Kalman filter algorithm to correct measurement errors, calculates the real-time position of each vehicle, and controls the vehicles to adjust to a range of no more than 50mm from the target position to complete the coarse positioning.

[0095] 2. Precise Alignment Stage

[0096] S31: The central control unit sends a "start visual positioning" command to each AGV. The AGV activates the stereo vision modules on both sides of the vehicle body. Two global shutter cameras simultaneously acquire real-time images of the ground contact features of the front and rear tires on the same side of the vehicle. The acquisition frequency is consistent with the camera frame rate (14fps) to ensure that continuous and complete tire image data is obtained, covering the tire ground contact area and key features of the sidewall.

[0097] S32: The robot preprocesses the acquired image pairs: first, it performs grayscale conversion, transforming color images into grayscale to reduce data volume; then, it uses a Gaussian filtering algorithm to remove image noise, preventing noise from affecting feature extraction; finally, it uses the Canny edge detection algorithm to extract edge features of the tire sidewall and ground contact area, highlighting the tire contour. After preprocessing, it uses SIFT or SURF feature point extraction algorithms to extract unique feature points (such as tire tread inflection points, sidewall markings, etc.) from the images; then, it uses the FLANN matching algorithm to perform stereo matching of feature points in the two images. During the matching process, it uses the Ratio check to eliminate incorrect matching pairs, ensuring that the matching accuracy is not less than 95%. Combining the calibrated camera intrinsic and extrinsic parameters (intrinsic parameters include focal length and principal point coordinates, extrinsic parameters include rotation matrix and translation vector), it calculates the three-dimensional coordinates of the feature points in the robot coordinate system using the principle of triangulation, and establishes a three-dimensional model of the tire.

[0098] S33: Based on the three-dimensional coordinates of the feature points, the car fits the geometry of the tire contact patch: Since the tire contact patch is approximately rectangular, the four sides of the rectangle are fitted using the least squares method, and the center coordinates of the rectangle are calculated. These center coordinates are the tire contact patch center coordinates. Combining the preset tire radius parameters (obtained from a database or through visual measurement depending on the vehicle type; in this embodiment, the tire radius is 0.5m), the precise horizontal coordinates (X, Y) of the tire center point are calculated through geometric relationships. The distance the tire radius is extended upwards in a direction perpendicular to the ground, using the tire contact patch center as a reference, is the Z coordinate of the tire center point. The horizontal coordinates (X, Y) are calculated through the projection relationship between the contact patch center coordinates and the tire radius in the horizontal direction. Simultaneously, by fitting the plane equation of the tire sidewall, the normal vector of the tire plane is solved, determining the tire's orientation. The final accuracy of the tire center point horizontal coordinates can reach ±2mm, and the accuracy of the normal vector direction can reach ±0.5°.

[0099] S34: After receiving the tire center point coordinates and normal vectors sent by each vehicle, the central control unit calculates the target position for placing the weighbridge, based on the preset vehicle movement direction (in this embodiment, the vehicle movement direction is horizontal forward). The target position must satisfy the coordinates of the weighbridge's center point ( , Relationship: , ,in The X-axis coordinate of the center point of the weighbridge. X is the Y-axis coordinate of the weighbridge center point, X is the X-axis coordinate of the tire center point, and Y is the Y-axis coordinate of the tire center point. This represents the forward offset.

[0100] Forward offset The adjustment is dynamically based on the vehicle's wheelbase, and the adjustment strategy is based on the formula. Define, where:

[0101] This is a proportional coefficient, set according to vehicle type, for freight vehicles. The value is typically 0.2-0.4 (k=0.3 in this example), for small cars. The value is usually 0.1-0.3 to ensure that the weighbridge placement is compatible with different vehicle models;

[0102] The wheelbase of the vehicle under test (in this embodiment) =6.5m);

[0103] To maintain a constant value, a safety margin is set, typically 0.1-0.3m (in this embodiment). =0.2m), used to prevent the tires from directly colliding with the weighbridge when the vehicle starts.

[0104] Substituting into the formula, we can obtain =0.3×6.5+0.2=2.15m, that is, the center point of the weighbridge is located 2.15m in front of the center point of the tire, to ensure that the tires can accurately and smoothly roll on the weighbridge when the vehicle is driving, while avoiding the risk of collision during the starting phase.

[0105] S35: After receiving the target position sent by the central control unit, the four AGVs move autonomously via an omnidirectional mobile chassis. During the movement, the AGVs monitor their deviation from the target position in real time through a stereo vision module, collecting deviation data every 50ms. The motion control algorithm adjusts the wheel speed and steering to gradually reduce the deviation. When the AGVs reach the target position, the robotic arm starts the micro-operation mode, precisely controlling the position and posture of the grippers through servo motors, and finely adjusting the lateral and longitudinal positions and horizontal angles of the weighbridge to ensure that the upper surface of the weighbridge is flush with the ground (height difference not exceeding 0.5mm) and that the longitudinal centerline of the weighbridge is perpendicular to the vehicle's driving direction (vertical deviation not exceeding 0.1°), thus completing the precise alignment of the weighbridge.

[0106] 3. Decentralization and dismantling of weighbridge connections

[0107] After all weighbridges have been precisely aligned, the robotic arms of each AGV gradually reduce their clamping force: from an initial 50N to 10N. During this process, pressure sensors monitor changes in clamping force to ensure a smooth change in force and prevent the weighbridge from shifting due to sudden changes in force. When the clamping force drops to 10N, the grippers slowly open at a speed controlled within 5mm / s until they are completely detached from the weighbridge's clamping mechanism. After the grippers open, the robotic arms slowly retract to their initial position (during the retraction process, the distance between the robotic arms and the weighbridge is no less than 50mm) to avoid collisions with the weighbridge or vehicle chassis.

[0108] 4. Weighbridge self-inspection and level calibration

[0109] After the weighbridge is placed in step S3 and before the vehicle moves forward in step S4, insert the weighbridge self-check and level calibration steps to ensure that the weighbridge's placement meets the weighing requirements:

[0110] Self-inspection process: The tilt sensors built into each weighbridge start detection, collect the tilt angles of the weighbridge in the X and Y axis directions, and send the detection data to the central control unit through the wireless communication module; the central control unit compares the tilt angles with the allowable range (±0.5°) to determine whether the weighbridge is level.

[0111] Calibration process:

[0112] If the levelness deviation is within the allowable range (≤±0.5°), the weighbridge sends a "normal" signal to the central control unit, and the system proceeds to the next weighing step.

[0113] If the levelness deviation exceeds the allowable range, two situations should be handled:

[0114] When the deviation is small (0.5° < deviation ≤ 1°): The central control unit calculates the levelness deviation compensation amount based on the tilt angle collected by the tilt sensor. By establishing a mapping relationship between the deviation and the weight error, the collected weight data is compensated by software in the subsequent weighing data processing to offset the impact of the levelness deviation on the weighing accuracy.

[0115] When the deviation is large (deviation > 1°): The central control unit sends a "secondary leveling" command to the nearest AGV. The robotic arm of the AGV extends again, the gripper aligns with the weighbridge clamping mechanism and applies a clamping force of 50N. Then, the position and attitude of the weighbridge are finely adjusted by the omnidirectional moving chassis (lateral, longitudinal translation or angular rotation). At the same time, the tilt sensor provides real-time feedback on the levelness data until the levelness deviation of the weighbridge is reduced to the allowable range. After leveling is completed, the robotic arm is released from the fixed connection and retracted. The weighbridge re-enters the self-inspection process to confirm that the levelness is qualified.

[0116] (iv) Step S4: Vehicle guidance and weighing data collection

[0117] After the weighbridge completes its self-inspection and level calibration, the central control unit confirms that all equipment is in normal condition. It then issues a visual driving instruction to the driver of the vehicle being tested via the LED display screen at the entrance of the testing area. The display screen shows "Please proceed at low speed, speed ≤ 5km / h", with a font size of no less than 50 points to ensure that the driver can clearly see it from the driver's seat. At the same time, an audible driving instruction is emitted through a high-volume speaker, repeatedly playing "Please proceed at low speed, pay attention to safety", with the volume controlled at 60-80 decibels to ensure that the driver can hear it while avoiding noise interference with the surrounding environment.

[0118] While issuing the forward command, the central control unit initiates the vehicle speed monitoring subprocess: the industrial camera based on the roadside visual recognition unit continuously captures motion images of the vehicle, calculates the vehicle's displacement in adjacent frames using the inter-frame difference method, and calculates the vehicle's real-time speed by combining the camera frame rate; if the vehicle's forward speed is detected to exceed the preset threshold (5km / h), the LED display immediately switches to display "Speed ​​too fast, please slow down", the font color changes to red, and the loudspeaker simultaneously plays the warning sound "Speed ​​too fast, please slow down" until the vehicle speed drops to below 5km / h before recording valid weighing data, thus avoiding weighing data fluctuations or damage to the weighbridge due to excessive vehicle speed.

[0119] The driver of the vehicle under test, following instructions, controls the vehicle to move forward at a low and constant speed of 2-5 km / h, with all the vehicle's tires sequentially rolling over the corresponding weighbridges. When the first axle tire rolls over the first weighbridge, the weighbridge's built-in load cell begins to collect weight signals at a frequency of 100Hz to ensure that the weight changes during the tire rolling process are captured. The weight signal first passes through an amplification circuit (1000x magnification) to amplify the weak sensor signal to a processable range, and then passes through a low-pass filter circuit (10Hz cutoff frequency) to remove high-frequency noise, resulting in a stable weight signal. The processed weight signal is then transmitted to the central control unit in real time via a wireless communication module in the form of a time sequence. CRC check is used during transmission to ensure data integrity.

[0120] The central control unit further processes the received weight data for each axle: First, it uses a Kalman filter algorithm, combined with the changing patterns of vehicle speed and weighing data, to remove random noise from the data and obtain more accurate weight data; then, it performs an integral calculation on the filtered weight data and takes the average of the integral results as the final weight value of that axle; for example, during the period when the first axle tires are rolling over the weighbridge, the integral calculation yields a weight of 5000 kg for the first axle, 6000 kg for the second axle, 5500 kg for the third axle, and 5800 kg for the fourth axle; the central control unit adds up the weights of each axle to obtain the total vehicle weight of 22300 kg, and stores the weights of each axle, the total vehicle weight, and the weighing timestamp in the local database and the cloud backup database. At the same time, it displays the weighing results to the driver through an LED display screen, showing the license plate number, the weights of each axle, the total weight, and the weighing status (qualified / exceeding the standard).

[0121] (V) Step S5: Weighbridge Co-recovery and Return

[0122] After weighing is completed, the central control unit confirms that the weighing data has been stored completely and issues a visual reversing command to the driver through the LED display screen. The display screen shows "Weighing complete, please reverse out of the detection area". At the same time, the loudspeaker emits an audible reversing command and plays "Weighing complete, please reverse out" repeatedly to remind the driver to reverse in time.

[0123] The driver of the vehicle under test controls the vehicle to reverse slowly. The central control unit monitors the vehicle's position in real time through the industrial camera of the roadside vision recognition unit: by recognizing the position changes of the vehicle's rear wheels, it determines whether the tires have completely exited the weighbridge bearing surface; when it is detected that all tires have exited the weighbridge bearing surface, the central control unit sends a "weighbridge recovery" command to the four AGV trolleys to prevent the trolleys from starting the recovery action before the vehicle has completely exited, thus preventing equipment collision.

[0124] After receiving the "weighbridge recovery" command, the four AGVs travel in a longitudinal formation (1.5m apart) to the corresponding weighbridge position. The AGVs use a stereo vision module to identify the weighbridge's clamping mechanism, adjust their position, and align the robotic arm's grippers with the clamping mechanism. The grippers slowly close, applying a 50N clamping force to ensure the weighbridge is securely clamped. Then, the robotic arm slightly lifts the weighbridge (lifting height not exceeding 10mm) to prevent the weighbridge from rubbing against the ground.

[0125] After all AGVs have finished gripping the weighbridge, under the control of the central control unit, they carry the weighbridge out from under the vehicle at a speed of 0.3-0.5 m / s. The exit route follows the original travel route, and the odometer and inertial measurement unit ensure accurate travel path to avoid collisions with vehicles or surrounding obstacles. Upon reaching the base station, the AGV first travels to the weighbridge storage area, places the weighbridge on the corresponding storage rack, opens and retracts the grippers, and completes the weighbridge storage. Then, the AGV travels to the corresponding parking space in the parking area and stops precisely using the positioning guidance device. After stopping, wireless charging is automatically started, and the charging coil aligns with the receiving coil on the bottom of the AGV to begin charging. The charging status is fed back to the central control unit via wireless communication, and the central control unit monitors the charging progress in real time.

[0126] After all AGVs have returned to their positions and started charging, the central control unit sends a "drive away" command to the driver via the LED display screen. The display screen shows "Equipment has been retrieved, please drive away from the detection area." The driver then controls the vehicle to drive away from the detection area. The central control unit organizes and archives all data from this weighing task (vehicle information, weighing results, equipment operating status), and the system returns to standby mode, waiting for the next weighing task.

[0127] III. Performance Testing and Validation of Systems and Methods

[0128] To ensure that the system and method in this embodiment meet the needs of practical applications, multiple rounds of performance tests and verifications were conducted for different vehicle models and environmental conditions to verify the system's weighing accuracy, operational efficiency, environmental adaptability, and safety. Specific test results are as follows:

[0129] (a) Weighing accuracy test

[0130] Ten standard-weight vehicles with different axle configurations (weights known, error ±10kg) were selected, including 2-axle sedans (standard gross weight 1.5t), 3-axle buses (standard gross weight 12t), 4-axle freight vehicles (standard gross weight 22t), 5-axle freight vehicles (standard gross weight 30t), and 6-axle freight vehicles (standard gross weight 49t), with two vehicles of each type. The system and method described herein were used to weigh each vehicle three times, and the deviation of each weighing result from the standard weight was calculated.

[0131] Test results show that the axle load deviations are all controlled within ±0.5%, with the smallest deviation being ±0.2% for 2-axle passenger cars and ±0.45% for 6-axle freight vehicles. The total vehicle weight deviations are all controlled within ±0.3%, meeting the accuracy requirements for axle load meters in the "JJG1014-2006 Verification Procedure for Axle (Wheel) Weight Meters for Motor Vehicle Testing" (axle load deviation ≤ ±1%, total weight deviation ≤ ±0.5%), proving that the system's weighing accuracy is reliable.

[0132] (ii) Operational efficiency test

[0133] The total time from when a vehicle enters the inspection area (the front of the vehicle crosses the inspection line) to when it completes weighing and leaves the inspection area was recorded. Ten tests were conducted for each vehicle type, and the average time was taken. The test results showed that the average total time was approximately 2 minutes for a 2-axle sedan, approximately 2.5 minutes for a 3-axle bus, approximately 3 minutes for a 4-axle freight vehicle, approximately 3.5 minutes for a 5-axle freight vehicle, and approximately 4 minutes for a 6-axle freight vehicle.

[0134] Compared to the traditional manual operation of weighbridges (manually moving the weighbridge, manually positioning it, and manually recording data, which takes an average of 10 minutes), this system improves operational efficiency by more than 60%, significantly reduces vehicle waiting time, reduces vehicle congestion in the inspection area, and is suitable for scenarios with high traffic volume.

[0135] (III) Environmental adaptability test

[0136] The system underwent continuous 24-hour stability testing under different environmental conditions, including high temperature (45℃, relative humidity 50%), low temperature (-10℃, relative humidity 60%), rainy weather (rainfall 20mm / h, ambient temperature 25℃), and dusty environment (dust concentration 50mg / m³, ambient temperature 25℃). Each type of environment was tested 5 times, and the operating status and weighing performance of each component of the system were recorded.

[0137] Test results show that: in high-temperature environments, the temperature of the central control unit and the AGV trolley was controlled below 60℃ (the maximum allowable temperature for the equipment is 70℃), and the weighing accuracy remained unchanged; in low-temperature environments, the battery capacity retention rate was above 85%, the equipment started normally, and there were no jamming issues; in rainy conditions, the camera lens of the roadside vision recognition unit had no water accumulation, the license plate recognition accuracy remained at 99.5%, and the weighbridge's wireless communication was uninterrupted; in dusty environments, the Mecanum wheels of the AGV trolley did not jam, the robotic arm moved flexibly, and the data collected by the weighing sensors was stable. In all environments, the system's weighing accuracy deviation was within the allowable range, and the operating efficiency did not decrease significantly, proving that the system has good environmental adaptability and can operate stably in complex outdoor environments.

[0138] (iv) Security Testing

[0139] Simulate abnormal scenarios in real-world applications to test the system's emergency response capabilities. Specific scenarios include vehicle speeding, weighbridge placement deviation, communication interruption, and robotic arm failure. Each scenario is tested 10 times, and the system's response is recorded.

[0140] Vehicle speeding scenario: When the vehicle speed exceeds 5km / h (up to 8km / h), the system can issue a deceleration warning within 0.5s. The LED display and the high-volume horn respond synchronously until the vehicle speed drops to a safe range. During this period, the weighbridge suspends data collection to avoid invalid data recording and prevents equipment damage.

[0141] Scenario of weighbridge placement deviation: When the level deviation of the weighbridge reaches 2° (exceeding the allowable range), the system can accurately identify the deviation. For cases with large deviations, the AGV can complete secondary leveling within 30 seconds. After leveling, the level deviation of the weighbridge is reduced to ±0.3%, and the weighing accuracy is not affected. For cases with small deviations, the software compensation algorithm can effectively offset the deviation, and the weight data error is controlled within ±0.4%.

[0142] Communication interruption scenario: When communication between the central control unit and the AGV is interrupted (interruption time not exceeding 10 seconds), the AGV immediately stops its current action and enters standby mode; after communication is restored, the central control unit resends the command, and the AGV continues to execute the unfinished task without any task disorder or equipment collision.

[0143] Robotic arm failure scenario: When the robotic arm of an AGV has insufficient clamping force (below 30N), the system immediately triggers an alarm, stops the movement of the AGV, and dispatches a backup AGV (if available) to take over the task; if there is no backup AGV, the central control unit records the fault information and notifies the maintenance personnel, while ensuring that other AGVs can complete the remaining tasks normally to avoid interruption of the entire weighing process.

[0144] Through the above tests and verifications, the portable weighbridge fully automatic loading and unloading and vehicle weighing method and system in this embodiment have demonstrated excellent performance in terms of weighing accuracy, operational efficiency, environmental adaptability and safety. It can be widely used in scenarios such as highway overload detection stations, logistics parks, ports, and mines where vehicles need to be weighed quickly, to achieve fully automated, efficient and accurate vehicle weighing, reduce manual intervention and lower operating costs.

[0145] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A fully automatic method for launching and retracting a portable weighbridge and weighing vehicles, characterized in that, The method includes the following steps: S1. Vehicle Recognition and Task Planning: The roadside vision recognition unit captures images of the vehicle to be tested, identifies its license plate information, and queries a pre-set vehicle information database to obtain the vehicle's number of axles, wheelbase, and key wheelbase parameters. The central control unit dynamically calculates and assigns the number of AGVs (Automated Guided Vehicles) required to perform the weighing task based on the key parameters, and generates a collaborative task instruction set containing the target position sequence. The number of AGVs is equal to the number of axles of the vehicle, and each AGV is responsible for the weighbridge operation at the position corresponding to one axle. S2. Weighbridge Cooperative Transportation and Pre-positioning: Multiple AGVs are automatically driven out from the base station. Each AGV uses a pair of gripping robotic arms to grab and fix a portable weighbridge. The multiple AGVs autonomously navigate in a longitudinal formation, moving from the side or rear of the vehicle to be tested to the area directly below it, forming an initial working array. S3. Precise Alignment and Placement of Weighbridge: Based on the vehicle parameters obtained in S1, each AGV first performs pre-positioning based on the odometer and inertial measurement unit; then, the stereo vision modules mounted on both sides of each AGV are activated to perform real-time image acquisition and 3D reconstruction of the ground contact characteristics of the front and rear tires on the same side of the vehicle, accurately calculating the real-time center coordinates and orientation of the tires; the central control unit coordinates and schedules each AGV to fine-tune its posture according to the calculation results, so that the weighbridge it carries is accurately placed in the preset weighing preparation position; the weighing preparation position meets the following requirements: the upper surface of the weighbridge is flush with the ground, and its center point is located at a predetermined safe distance directly in front of the theoretical rolling trajectory of the corresponding tire; after all weighbridges are in place, the robotic arm of each AGV performs the weighbridge lowering action, so that it is placed stably on the ground, and then the robotic arm is decoupled from the weighbridge and retracted to the AGV; S4. Vehicle Guidance and Weighing Data Acquisition: The central control unit issues a visual and / or audible forward movement command to the driver of the vehicle under test; the vehicle under test moves forward at a low and constant speed, so that all its tires simultaneously roll over the corresponding weighbridge; the weighing sensors built into each weighbridge collect the weight signal under the load of that axle, and send the weight data in time series to the central control unit through the wireless transmission module; the central control unit filters and integrates the received weight data of each axle to obtain and record the weight of each axle and the total weight of the vehicle; S5. Weighbridge Cooperative Retrieval and Return: After weighing is completed, the central control unit issues a reversing command to the driver; after the vehicle tires have completely exited the weighbridge bearing surface, the robotic arms of each AGV trolley extend again and clamp and fix the corresponding weighbridge; under the scheduling of the central control unit, all AGV trolleys carry the weighbridge and autonomously exit from under the vehicle in a formation, returning to the designated parking position at the base station; after the vehicles leave, the system returns to standby status.

2. The portable weighbridge fully automatic loading and unloading and vehicle weighing method according to claim 1, characterized in that, In step S1, the vehicle information database is a relational database stored in the cloud or locally, which is accessed through the license plate number and at least pre-stores the vehicle's number of axles, number of tires per axle, standard wheelbase, and wheelbase information; If the query is unsuccessful, the central control unit initiates the backup lidar point cloud scanning process, and calculates the number of axles, wheelbase, and track width of the vehicle under test in real time through point cloud clustering and geometric feature extraction algorithms.

3. The portable weighbridge fully automatic loading and unloading and vehicle weighing method according to claim 1, characterized in that, The precise alignment and placement described in step S3 are specifically implemented as follows: S31. Based on the stereo vision module, continuously acquire image pairs of the vehicle tire sidewall and the ground contact area; S32. Extract feature points and perform stereo matching on the image pairs. Combine the calibrated camera intrinsic and extrinsic parameters to calculate the three-dimensional coordinates of the feature points in the robot coordinate system. S33. By fitting the geometric center of the tire contact patch and combining it with the tire radius parameter, the precise horizontal coordinates (X, Y) of the tire center point and the normal vector of the tire plane are calculated. S34. The central control unit calculates the target position for placing the weighbridge based on the tire center coordinates, normal vector, and preset vehicle movement direction; the target position ensures the center point coordinates of the weighbridge ( , The following relationship must be satisfied: in, The forward offset is set according to the vehicle type and safety margin; S35. Each AGV trolley moves autonomously to its position according to the received target location, and completes the final placement of the weighbridge through the micro-operation capability of the robotic arm, ensuring that the upper surface of the weighbridge is flush with the ground and the longitudinal center line of the weighbridge is perpendicular to the vehicle's driving direction.

4. The portable weighbridge fully automatic loading and unloading and vehicle weighing method according to claim 3, characterized in that, The forward offset in step S34 Its value is dynamically adjusted based on the vehicle wheelbase queried from the database or measured in real time. The adjustment strategy is defined by the following formula: in: A proportional coefficient set according to vehicle type, and 0 < <1; The wheelbase of the vehicle under test is expressed in meters (m). This is a fixed constant set based on a safety margin, and the unit is meters (m).

5. The portable weighbridge fully automatic loading and unloading and vehicle weighing method according to claim 1, characterized in that, The lurking autonomous mobile robot (AGV) adopts an omnidirectional mobile chassis, and its driving method is Mecanum wheels or a combination of omnidirectional wheels; the gripping robotic arm is an electric servo robotic arm with at least 4 degrees of freedom, and its end effector is an adaptive gripper, which can be compatible with gripping portable weighbridges with different thicknesses and frame structures; the stereo vision module includes two symmetrically arranged global shutter cameras, and their baseline distance has been optimized to meet the requirements of accurate distance measurement in low-light environments under the vehicle.

6. The portable weighbridge fully automatic loading and unloading and vehicle weighing method according to claim 1, characterized in that, In step S4, the central control unit initiates a vehicle speed monitoring subprocess based on the roadside vision recognition unit or the AGV vehicle vision module while issuing a forward command. If the vehicle's forward speed is detected to exceed a preset threshold, a deceleration warning is immediately issued to the driver, and valid weighing data is only recorded after the speed drops to a safe range.

7. The portable weighbridge fully automatic loading and unloading and vehicle weighing method according to claim 1, characterized in that, After the weighbridge is placed in step S3 and before the vehicle moves forward in step S4, a weighbridge self-check and level calibration step is inserted: the tilt sensors built into each weighbridge detect the levelness of the weighbridge after it is placed. If the levelness deviation exceeds the allowable range, the nearest AGV trolley will re-clamp the weighbridge and perform a second leveling and placement, or the central control unit will calculate the levelness deviation compensation amount and perform software compensation in the subsequent weighing data.

8. An automatic weighing system for a fully automatic portable weighbridge for loading, unloading, and vehicle weighing, applicable to the fully automatic portable weighbridge for loading, unloading, and vehicle weighing method described in any one of claims 1-7, characterized in that, The system includes: A central control unit, acting as the brain of the system, is responsible for task scheduling, data fusion, and decision-making; A roadside vision recognition unit is fixedly installed above the entrance to the detection area for vehicle capture and license plate recognition; A vehicle information database used to store and provide vehicle structural parameters; At least two stealthy autonomous mobile robots (AGVs) constitute the main execution body of the system; A portable weighbridge matching the number of AGV vehicles; A base station provides parking, charging, and weighbridge storage functions for the AGV vehicles; wherein, the central control unit establishes real-time and reliable wired and / or wireless communication connections with the roadside vision recognition unit, vehicle information database, all AGV vehicles, and all portable weighbridges.

9. An automatic weighing system for a fully automatic loading and unloading of portable weighbridges and a method for weighing vehicles, as described in claim 8, is characterized in that... The portable weighbridge has undergone adaptive structural improvements, including: A standardized gripping mechanism matching the gripper of the AGV robot arm is provided on its side or corner; It has a built-in tilt sensor, wireless communication module and high-precision weighing sensor; The weighbridge has a high-friction, wear-resistant coating on its upper surface to increase the adhesion of tires when they roll over it.

Images