A heavy-duty frame assembly AGV intelligent parking system
By introducing servers and intelligent parking modules in the AGV system, using sensors to monitor the status of assembly equipment and dynamically adjusting the parking path, the problem of inefficient production efficiency caused by fixed paths in heavy-load frame assembly of the AGV system is solved, and more efficient Resource utilization and production optimization.
Patent Information
- Application Number
- CN202411252504.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-09
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-09-09
AI Technical Summary
During the assembly process of heavy-load frames, the existing AGV system has reduced production efficiency and cannot adapt to changes in production conditions or temporary adjustments.
A smart AGV parking system including server, assembly monitoring module and intelligent parking module was designed. Through communication with sensors, the operation information of each parking space is obtained, the status of assembly equipment is monitored and analyzed in real time, the parking path of AGV is dynamically regulated, and the assembly task is optimized. Assignment and path selection.
The intelligent parking and assembly path optimization of AGV is realized, avoiding the limitations of fixed routes, improving the resource utilization rate of the production system, and ensuring the efficient operation and production efficiency of AGV.
Smart Images

Figure CN118799650B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of intelligent parking, and in particular to an AGV intelligent parking system for heavy-duty vehicle frames. Background Art
[0002] Heavy-duty frame assembly refers to the process of assembling the frame (chassis or frame) of heavy vehicles or equipment. The frame is the main load-bearing structure of the vehicle or equipment, usually used to support and connect other key components, such as the engine, transmission system, suspension system and body. Heavy-duty frames are usually large in size and heavy in weight. Manual or traditional mechanical transportation is prone to inefficiency and safety issues. AGV (automatic guided vehicle) can automatically carry and transport these frames and move freely in the assembly workshop; therefore, AGV is particularly important for heavy-duty frame assembly.
[0003] Current AGVs (automated guided vehicles) usually park and assemble based on a predetermined route when assembling heavy-loaded frames. This fixed-route parking and assembly method lacks flexibility because once the production situation changes or the assembly order is temporarily adjusted, the fixed-path AGV system cannot adjust and respond quickly, resulting in reduced production efficiency. Summary of the invention
[0004] Based on this, it is necessary to provide a heavy-duty frame assembled AGV intelligent parking system to address the problems mentioned in the above background technology.
[0005] The object of the present invention can be achieved by the following technical solutions: A heavy-duty frame assembly AGV intelligent parking system, including a server, an assembly monitoring module and an intelligent parking module;
[0006] The server communicates with various sensors (specifically, laser radar, ultrasonic sensor, visual camera, temperature sensor, etc.) in the heavy-duty frame assembly workshop to obtain and save the operation information of each parking space;
[0007] The assembly monitoring module performs real-time monitoring and analysis on the assembly equipment of each parking space based on the operation information of each parking space to obtain an operation evaluation index, and sends it to the intelligent parking module;
[0008] The intelligent parking module dynamically adjusts the heavy-load frame assembly based on the operation evaluation index of each workstation, and controls the AGV to perform intelligent parking accordingly; specifically:
[0009] Step 1: Retrieve the delivery time and quantity of the order to which each AGV's assembly task belongs, calculate the time difference between the delivery time and the current system time to get the delivery time, and record the delivery time and delivery quantity as F1 and F2 respectively; substitute F1 and F into the set formula Calculate and obtain the delivery value Fb, where b1 and b2 are the set proportional coefficients respectively; sort the AGVs according to their delivery values, and select the AGV with the largest delivery value as the calibration AGV;
[0010] Step 2: Retrieve the assembly history records of each parking station, where the assembly history records include the number of assemblies, the three-dimensional assembly image corresponding to each assembly, the assembly start time and the assembly end time; calculate the time difference between the assembly start time and the assembly end time to obtain the assembly duration;
[0011] Assume that there is a standard three-dimensional image, and perform feature matching analysis between the standard three-dimensional image and the assembly three-dimensional image to obtain the assembly deviation value between the assembly three-dimensional image and the standard three-dimensional image; thereby, the assembly deviation value corresponding to each assembly in the assembly history record can be obtained, and the mean value can be calculated to obtain the deviation mean;
[0012] The average assembly time corresponding to each assembly is calculated and recorded as B, and the average assembly time is calculated and compared with the deviation mean. Substitute the set formula The assembly performance value CB of each parking space is calculated, where f4 and f5 are respectively the set proportional coefficients;
[0013] Step 3: Get the position of the AGV, and calculate the distance difference between it and the position of each parking space to get the transport distance, which is recorded as N1; get the number of assembly AGVs to be processed in each parking space, and record it as N2; retrieve the assembly performance value CB and operation evaluation index Ag of each parking space closest to the current time of the system, and use the set formula Calculate and obtain the assembly value Nk, where k1, k2, k3, and k4 are respectively the set proportional coefficients; select the parking space with the largest assembly value as the calibration parking space, and assign the assembly task of the calibration AGV to the corresponding calibration parking space, then the number of pending assembly AGVs in the corresponding calibration parking space increases by one;
[0014] Step 4: Repeat steps 1 to 3 until all AGV assembly tasks are assigned;
[0015] Step 5: Each AGV generates an assembly path according to its assigned parking space, and goes to the corresponding parking space for assembly along the assembly path; every time an AGV assembly task is completed in a parking space, the number of AGVs to be assembled in the parking space is reduced by one, and the assembly start time, assembly end time and assembly 3D image of this assembly are recorded; the assembly start time, assembly end time and assembly 3D image of this assembly are integrated into an assembly history record, and the assembly history record is updated to step 2.
[0016] In some embodiments, the specific process of feature matching analysis is:
[0017] 201: Set a number of feature points in the standard three-dimensional image, thereby obtaining the three-dimensional coordinates of each feature point in the standard three-dimensional image, and record it as P 标准 (X 标准 j, Y 标准 j, Z 标准 j), where j = 1, 2, 3...J, j is the serial number of any feature point, J represents the total number of feature points, and J is a positive integer;
[0018] 202: identifying feature points corresponding to those in the standard three-dimensional image in the assembled three-dimensional image, thereby forming feature point pairs, and calculating the position deviation of each pair of feature point pairs; when the position deviation of the feature point pair is greater than zero, the feature point pair is recorded as a deviation feature point pair, thereby obtaining a plurality of deviation feature point pairs;
[0019] 203: Compare and analyze the position deviation with the set deviation interval. When the position deviation is greater than the maximum value in the set deviation interval, a high deviation point pair is accumulated; when the position deviation is in the set deviation interval, a medium deviation point pair is accumulated; when the position deviation is less than the minimum value in the set deviation interval, a low deviation point pair is accumulated; the accumulated numbers of high deviation point pairs, medium deviation point pairs and low deviation point pairs are counted respectively, and they are recorded as C1, C2 and C3 respectively; the position deviations of the high deviation point pairs, medium deviation point pairs and low deviation point pairs are summed up to obtain the high deviation value, medium deviation value and low deviation value, and they are recorded as C4, C5 and C6 respectively;
[0020] Substitute C1, C2, C3, C4, C5, and C6 into the set formula The assembly deviation value Cf is calculated, where f1, f2, and f3 are respectively set proportional coefficients, and f1>f2>f3>0, thereby obtaining the assembly deviation value corresponding to each assembly in the assembly history record.
[0021] In some embodiments, the specific process of identifying feature points in the assembly three-dimensional image and calculating the position deviation is:
[0022] 301: Feature point recognition: In the assembly 3D image, use the algorithm to identify representative feature points;
[0023] 302: Extracting geometric information: extracting geometric information of feature points, using the nearest neighbor algorithm, preliminarily matching feature points in the assembled 3D image corresponding to feature points in the standard 3D image, and calculating the Euclidean distance of each pair of matched feature points, preliminarily screening out several feature points whose positions are close to feature points in the standard 3D image, thereby forming preliminary feature point pairs;
[0024] 303: Use the RANSAC algorithm to remove incorrectly matched point pairs, and determine the optimal point pair matching through random sampling and fitting;
[0025] 304: Use the ICP algorithm to perform fine alignment, repeatedly iteratively minimize the distance between the corresponding feature points in the assembled 3D image and the standard 3D image until convergence; thus, the feature points in the assembled 3D image corresponding to the feature points in the standard 3D image are recorded as P (Xj, Yj, Zj);
[0026] 305: Through the Euclidean space distance formula The position deviation ΔPj of each feature point is obtained.
[0027] In some embodiments, the specific process of real-time monitoring and analyzing the assembly equipment of each parking space based on the operation information of each parking space is as follows:
[0028] 401: Retrieve the operation information corresponding to each acquisition moment, wherein the operation information includes the noise intensity value, actual power value, vibration amplitude and temperature of each part of the assembly equipment; set a standard operation parameter corresponding to each assembly equipment, wherein the standard operation parameter includes standard power and standard amplitude;
[0029] 402: Comprehensively quantify the interference degree of each part of the assembly equipment according to the temperature thereof to obtain the temperature interference value of the assembly equipment at each collection time;
[0030] 403: The noise intensity value Si, actual power Wi, vibration amplitude Di, temperature interference value Mai, standard power HW and standard amplitude HD are calculated by the set formula Calculation is performed to obtain the state value STi, where h1, h2, h3, and h4 are respectively the set proportional coefficients; a two-dimensional rectangular coordinate system is constructed with time as the horizontal coordinate and the state value as the vertical coordinate, and the state value is placed at the second point on the coordinate axis according to its corresponding collection time to obtain the state point; a smooth curve is used to connect each state point in turn to obtain a relationship diagram of the state value of the assembly equipment of each parking space changing with time, and based on this, an in-depth analysis of the state changes of the assembly equipment of each parking space is carried out to obtain the operation evaluation value.
[0031] In some embodiments, the specific process of comprehensively quantifying the interference degree of each part of the assembly equipment according to the temperature thereof is as follows:
[0032] Set a temperature sensitivity coefficient corresponding to each part of the assembly equipment, and multiply the temperature of each part by the corresponding temperature sensitivity coefficient to obtain the sensitivity value of each part;
[0033] The sensitivity value of each part is compared and analyzed with the set sensitivity range to divide the parts corresponding to the sensitivity value into highly sensitive parts, moderately sensitive parts and low sensitive parts, and the number of highly sensitive parts, moderately sensitive parts and low sensitive parts in the assembly equipment is counted respectively, and they are recorded as M1, M2, and M3 respectively; the sensitivity values of the highly sensitive parts, moderately sensitive parts and low sensitive parts are summed up to obtain highly sensitive values, moderately sensitive values and low sensitive values, and they are recorded as M4, M5, and M6 respectively;
[0034] Substitute M1, M2, M3, M4, M5, and M6 into the set formula The temperature interference value Ma is obtained by calculation, wherein a1, a2, and a3 are respectively set proportional coefficients, a1>a2>0 and a3>a4>0, thereby the temperature interference value at each acquisition time can be obtained.
[0035] In some embodiments, the specific process of in-depth analysis of the status change of the assembly equipment in each parking space is as follows:
[0036] Draw a tangent line at each state point, and calculate the slope of each tangent line to obtain the state slope;
[0037] The state slopes greater than zero are summed up to obtain the state increase value, and the state slopes less than zero are summed up and then the absolute value is taken to obtain the state decrease value;
[0038] Substitute the state value STi, state slope Li, state increase value A1 and state decrease value A2 at each acquisition time into the set formula The operation evaluation index Ag is calculated, where g1, g2, and g3 are the set proportional coefficients. is the mean value of the state slope of each state point.
[0039] Compared with the prior art, the present invention has the following beneficial effects:
[0040] 1. By obtaining the operating status of the assembly equipment in each parking space (such as noise intensity, actual power, vibration amplitude, temperature, etc.) and analyzing it to obtain the operation evaluation index, the operating status of the assembly equipment in each parking space can be grasped in real time, so that the staff can find potential faults in time; using the operation evaluation index, the operating status of the equipment can be quantitatively evaluated, which can provide data support for the intelligent parking of AGV;
[0041] 2. By dynamically adjusting the AGV assembly task based on the real-time assembly situation (such as the urgency of the assembly task, the equipment status, the current load of the parking space, etc.), the intelligent parking and assembly path selection of the AGV can be optimized, avoiding the limitations of fixed routes; achieving dynamic optimal configuration of assembly resources, avoiding idle or overused resources, and improving the resource utilization of the production system;
[0042] 3. The system continuously updates the assembly history records. Through the long-term accumulation and analysis of the assembly history records, it not only provides data support for the current assembly tasks, but also provides a basis for future assembly optimization, promoting continuous optimization. BRIEF DESCRIPTION OF THE DRAWINGS
[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the conventional technology, the drawings required for use in the embodiments or the conventional technology descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0044] Figure 1 It is a principle block diagram of the present invention;
[0045] Figure 2 It is the transportation route and parking space layout diagram of the present invention. DETAILED DESCRIPTION
[0046] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without violating the connotation of the present invention, so the present invention is not limited by the specific embodiments disclosed below.
[0047] like Figure 1 As shown, a heavy-duty frame assembly AGV intelligent parking system includes a server, an assembly analysis module and an intelligent parking module;
[0048] like Figure 2 As shown, it is assumed that there are several transport paths in the heavy-duty frame assembly workshop, and there are several parking spaces on both sides of each transport path for the assembly of heavy-duty frames. The AGV can assemble the heavy-duty frame by driving to the corresponding assembly parking space without affecting the driving of other AGVs; the assembly tasks are randomly assigned to each AGV;
[0049] The server communicates with various sensors (specifically, laser radar, ultrasonic sensor, visual camera and other sensors) in the heavy-duty frame assembly workshop to obtain the operation information of each parking space and save it; the operation information refers to the parameters of the assembly equipment corresponding to the parking space during operation;
[0050] The assembly monitoring module performs real-time monitoring and analysis on the assembly equipment of each parking space based on the operation information of each parking space to obtain the operation evaluation index, and sends it to the intelligent parking module; specifically:
[0051] Retrieve the operation information corresponding to each collection moment, where the operation information includes the noise intensity value of the assembly equipment (abnormal noise indicates the risk of friction, looseness or other mechanical problems inside the equipment), actual power value (too high or too low power consumption indicates the risk of electrical failure or abnormal equipment load), vibration amplitude (excessive vibration may indicate wear or imbalance of mechanical parts of the equipment, which requires timely adjustment or maintenance) and the temperature of each part of the assembly equipment, and record the noise intensity value, actual power and vibration amplitude as Si, Wi, Di, where i=1,2,3...I, I is a positive integer, I represents the total number of collection moments, and i represents the sequence number of any collection moment;
[0052] Each assembly equipment is set to correspond to a standard operating parameter, where the standard operating parameters include standard power (usually refers to rated power), standard amplitude (usually refers to the maximum allowable amplitude of vibration displacement of the equipment under normal operating conditions, this amplitude is usually set by the manufacturer based on the mechanical structure and usage scenario of the equipment), and they are recorded as HW and HD respectively; standard operating parameters refer to the performance indicators or working conditions that the equipment should comply with under normal working conditions, which are determined during the design and manufacturing of the equipment; these parameters are usually determined by the equipment manufacturer during the product design and testing phase based on factors such as the function, structure and expected use environment of the equipment. Standard operating parameters set the ideal working state of the equipment to ensure that the equipment can achieve optimal performance and the longest service life when operating within these parameter ranges;
[0053] Different parts of the assembly equipment are set to correspond to a temperature sensitivity coefficient. It should be noted that the temperature sensitivity coefficient refers to the sensitivity of different parts of the equipment to temperature changes. It indicates the response degree of the performance or characteristics of a certain part to temperature changes. Due to the different materials, structures and functions of different parts of the equipment, their sensitivity to temperature changes will also be different. For example, the insulation material of the motor or motor winding is very sensitive to temperature. Too high temperature will cause the motor to overheat or the insulation layer to be damaged, thereby causing motor failure. Bearings are sensitive to temperature changes. Excessive temperature will affect the viscosity of the lubricating oil, reduce the lubrication effect, increase wear, and even cause bearing failure. The housing is generally made of metal or other materials and is not very sensitive to temperature changes. It usually only shows obvious thermal expansion or contraction at extreme temperatures (such as overcooling or overheating). It can be seen that the temperature sensitivity coefficient corresponding to the motor is greater than that corresponding to the bearing, and the temperature sensitivity coefficient corresponding to the bearing is greater than that corresponding to the housing. Multiply the temperature of each part by the corresponding temperature sensitivity coefficient to obtain the sensitivity value of each part.
[0054] Compare and analyze the sensitivity value of each part with the set sensitivity interval. When the sensitivity value is greater than the maximum value in the set sensitivity interval, a highly sensitive part is accumulated; when the sensitivity value is in the set sensitivity interval, a moderately sensitive part is accumulated; when the sensitivity value is less than the minimum value in the set sensitivity interval, a low-sensitive part is accumulated; count the number of highly sensitive parts, moderately sensitive parts and low-sensitive parts in the assembly equipment respectively, and record them as M1, M2, and M3 respectively; sum the sensitivity values of highly sensitive parts, moderately sensitive parts and low-sensitive parts respectively to obtain highly sensitive values, moderately sensitive values and low-sensitive values, and record them as M4, M5, and M6 respectively; through the set formula The temperature interference value Ma is obtained by calculation, where a1, a2, and a3 are respectively the set proportional coefficients, a1>a2>0 and a3>a4>0; it can be seen from the formula that when the sensitivity value of each part in the assembly equipment is greater, it means that it is more affected by temperature, and the temperature interference value is greater; thus, the temperature interference value at each acquisition moment is recorded as Mai;
[0055] The noise intensity value Si, actual power Wi, vibration amplitude Di, temperature interference value Mai, standard power HW and standard amplitude HD are calculated through the set formula The state value STi is obtained by calculation, where h1, h2, h3, and h4 are respectively the set proportional coefficients; it can be seen from the formula that when the noise intensity value is larger, it indicates that there is a risk of friction, looseness or other mechanical problems inside the equipment, and the state value is smaller; when the actual power is closer to the standard power, the state value is larger; when the vibration amplitude is larger than the standard amplitude, it indicates that there is a risk of mechanical parts wear or imbalance in the equipment, and the state value is smaller; when the temperature interference value is larger, the state value is smaller;
[0056] A two-dimensional rectangular coordinate system is constructed with time as the horizontal coordinate and the state value as the vertical coordinate, and the state value is placed at the second point of the coordinate axis according to its corresponding collection time to obtain the state point; a smooth curve is used to connect each state point in sequence to obtain a relationship diagram of the state value of the assembly equipment of each parking space changing with time; a tangent line is drawn at each state point, and the slope of each tangent line is calculated to obtain the state slope, which is recorded as Li; it should be noted that when the state slope is greater than zero, it means that the state value at the state point shows an increasing trend, indicating that the operating state of the equipment is in an improving state; when the state slope is less than zero, it means that the state value at the state point shows a decreasing trend, indicating that the operating state of the equipment is in a deteriorating state;
[0057] The state slopes greater than zero are summed to obtain the state increase value, recorded as A1, and the state slopes less than zero are summed and then the absolute value is calculated to obtain the state decrease value, recorded as A2; using the set formula The operation evaluation index Ag is calculated, where g1, g2, and g3 are the set proportional coefficients. is the mean value of the state slope of each state point; from the formula, it can be seen that when the state increase value is larger and the state decrease value is smaller, the operation evaluation index is larger; when the state slope is more volatile, it means that the operation state of the assembly equipment is more unstable, and the operation evaluation value is smaller;
[0058] By acquiring the operating status of the assembly equipment in each parking space (such as noise intensity, actual power, vibration amplitude, temperature, etc.) and analyzing it to obtain the operation evaluation index, the operating status of the assembly equipment in each parking space can be grasped in real time, so that the staff can discover potential fault problems in time; using the operation evaluation index, a quantitative evaluation of the operating status of the equipment can be achieved, which can provide data support for the intelligent parking of AGV.
[0059] The intelligent parking module dynamically adjusts the heavy-load frame assembly based on the operation evaluation index of each workstation to achieve intelligent parking of AGV and improve assembly efficiency; specifically:
[0060] Step 1: Retrieve the delivery time and delivery volume of the order to which each AGV's assembly task belongs; the tighter the delivery time and the larger the delivery volume, the more urgent the corresponding order and the more urgent the assembly task corresponding to the order; calculate the time difference between the delivery time and the current system time to obtain the delivery time, and record the delivery time and delivery volume as F1 and F2 respectively, and use the set formula Calculate and obtain the delivery value Fb, where b1 and b2 are the set proportional coefficients respectively; sort the AGVs according to their delivery values, and select the AGV with the largest delivery value as the calibration AGV;
[0061] Step 2: Retrieve the assembly history records of each parking station, where the assembly history records include the number of assemblies and the corresponding three-dimensional assembly images for each assembly (the three-dimensional assembly image here refers to the three-dimensional image after the assembly is completed), the assembly start time and the assembly end time; calculate the time difference between the assembly start time and the assembly end time to obtain the assembly time; It should be noted that the specific steps for obtaining the assembly three-dimensional image are: capture the assembled image through a visual sensor to identify the relative position and angle of each component; provide accurate distance measurement through a laser rangefinder to detect the gap and relative position between components; scan the entire assembly structure through a 3D scanner to generate detailed three-dimensional point cloud data, including complete shape, size, position and angle information, and process the acquired data information to generate an assembly three-dimensional image;
[0062] It is assumed that there is a standard 3D image. It should be noted that the standard 3D image refers to a reference model or image used to verify and evaluate the actual assembly quality during the assembly process. It is a benchmark that defines the precise position, size and angle of each component in the assembly, and serves as a standard for evaluating the actual image after the assembly is completed; a number of feature points are set in the standard 3D image, where the feature point values are representatively recorded as positions, which usually play a key role in the quality and accuracy of the entire assembly, such as connection points, edge points, center points, plane intersections, etc.; the 3D coordinates of each feature point in the standard 3D image can be obtained and recorded as P 标准 (X 标准 j, Y 标准 j, Z 标准 j), where j = 1, 2, 3...J, j is the serial number of any feature point, J represents the total number of feature points, and J is a positive integer;
[0063] Identify the feature points corresponding to those in the standard 3D image in the assembled 3D image, thereby forming feature point pairs, and calculate the position deviation of each pair of feature point pairs; specifically:
[0064] R1: Feature point recognition: In the assembled 3D image, use algorithms (such as SIFT, SURF, ORB, etc.) to identify representative feature points. These points are usually points where planes intersect, have high curvature, and have prominent edges, similar to the feature points in standard 3D images;
[0065] R2: Geometric information extraction: Extract the geometric information of feature points (such as position, normal vector, angle, etc.), use the nearest neighbor algorithm (such as KD tree, FLANN, etc.), preliminarily match the feature points in the assembled 3D image corresponding to the feature points in the standard 3D image, and calculate the Euclidean distance of each pair of matched feature points, preliminarily screen out several feature points whose positions are close to the feature points in the standard 3D image, thereby forming preliminary feature point pairs;
[0066] R3: Use the RANSAC (Random Sampling Consensus) algorithm to remove incorrectly matched point pairs and determine the optimal point pair matching through random sampling and fitting;
[0067] R4: Use the ICP algorithm for fine alignment, repeatedly iteratively minimize the distance between the corresponding feature points in the assembled 3D image and the standard 3D image until convergence; the feature points in the assembled 3D image corresponding to the feature points in the standard 3D image are recorded as P (Xj, Yj, Zj);
[0068] R5: Through the Euclidean space distance formula The position deviation ΔPj of each feature point is obtained; when the position deviation of the feature point is greater than zero, the feature point pair is recorded as a deviation feature point pair, thereby obtaining several pairs of deviation feature point pairs;
[0069] Compare and analyze the position deviation with the set deviation interval. When the position deviation is greater than the maximum value in the set deviation interval, a high deviation point pair is accumulated; when the position deviation is in the set deviation interval, a medium deviation point pair is accumulated; when the position deviation is less than the minimum value in the set deviation interval, a low deviation point pair is accumulated; the cumulative number of high deviation point pairs, medium deviation point pairs and low deviation point pairs are counted respectively, and they are recorded as C1, C2 and C3 respectively; the position deviations of high deviation point pairs, medium deviation point pairs and low deviation point pairs are summed up to obtain high deviation values, medium deviation values and low deviation values, and they are recorded as C4, C5 and C6 respectively; using the set formula The assembly deviation value Cf is calculated, where f1, f2, and f3 are the set proportional coefficients, and f1>f2>f3>0; thus, the assembly deviation value corresponding to each assembly in the assembly history record can be obtained, and the mean deviation is calculated and recorded as ; It should be noted that the mean deviation represents its assembly accuracy. The larger the mean deviation, the lower the assembly accuracy.
[0070] By accurately evaluating the assembly quality of the assembly 3D image and the standard 3D image to obtain the assembly deviation value, the assembly accuracy of each parking space is judged based on it, and the assembly quality of each parking space is monitored;
[0071] The average assembly time corresponding to each assembly is calculated and recorded as B, and the average assembly time is calculated and compared with the deviation mean. Substitute the set formula The assembly performance value CB of each parking space is calculated, where f4 and f5 are respectively the set proportional coefficients; it can be seen from the formula that when the assembly average length is larger, the deviation mean is larger, indicating that the efficiency of the assembly equipment of the parking space is lower, and the accuracy is lower, then the assembly performance value is smaller;
[0072] Step 3: Get the position of the AGV, and calculate the distance difference between it and the position of each parking space to get the transport distance, which is recorded as N1; get the number of assembly AGVs to be processed in each parking space, and record it as N2; retrieve the assembly performance value CB and operation evaluation index Ag of each parking space closest to the current time of the system, and use the set formula Calculate and obtain the assembly value Nk, where k1, k2, k3, and k4 are respectively the set proportional coefficients; select the parking space with the largest assembly value as the calibration parking space, and assign the assembly task of the calibration AGV to the corresponding calibration parking space, then the number of pending assembly AGVs in the corresponding calibration parking space increases by one;
[0073] Step 4: Repeat steps 1 to 3 until all AGV assembly tasks are assigned;
[0074] Step 5: Each AGV generates an assembly path according to its assigned parking space, and goes to the corresponding parking space for assembly according to the assembly path; every time an AGV assembly task is completed in a parking space, the number of AGVs to be assembled in the parking space is reduced by one, and the assembly start time, assembly end time and assembly 3D image of this assembly are recorded; the assembly start time, assembly end time and assembly 3D image of this assembly are integrated into an assembly history record, and the assembly history record is updated to step 2;
[0075] By dynamically adjusting the AGV assembly task according to the real-time assembly situation (such as the urgency of the assembly task, the equipment status, the current load of the parking space, etc.), the intelligent parking and assembly path selection of the AGV can be optimized, avoiding the limitations of fixed routes; the dynamic optimal configuration of assembly resources is achieved, avoiding idle or excessive use of resources, and improving the resource utilization of the production system; at the same time, the system continuously updates the assembly history records. Through the long-term accumulation and analysis of the assembly history records, it not only provides data support for the current assembly tasks, but also provides a basis for future assembly optimization, promoting continuous optimization.
[0076] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0077] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.
Claims
1. A heavy-duty frame assembly AGV intelligent parking system, characterized in that: include: Assemble monitoring module and intelligent parking module; The assembly monitoring module performs real-time monitoring and analysis on the assembly equipment of each parking space based on the operation information of each parking space to obtain an operation evaluation index, and sends it to the intelligent parking module; The intelligent parking module dynamically adjusts the heavy-load frame assembly based on the operation evaluation index of each workstation, and controls the AGV to perform intelligent parking accordingly; specifically: Step 1: retrieve the delivery time and delivery quantity of the order to which the assembly task of each AGV belongs, calculate the time difference between the delivery time and the current time of the system to obtain the delivery time, normalize the delivery time and the delivery quantity and take their values, and analyze the values to obtain the delivery value; sort the AGVs according to their delivery values, and select the AGV with the largest delivery value as the calibration AGV; the more urgent the delivery time and the larger the delivery quantity, the larger the delivery value, and the more urgent the corresponding delivery order; Step 2: Retrieve the assembly history records of each parking station, where the assembly history records include the number of assemblies, the three-dimensional assembly image corresponding to each assembly, the assembly start time and the assembly end time; calculate the time difference between the assembly start time and the assembly end time to obtain the assembly duration; Assuming that there is a standard three-dimensional image, a feature matching analysis is performed between the standard three-dimensional image and the assembly three-dimensional image to obtain an assembly deviation value between the assembly three-dimensional image and the standard three-dimensional image; Thus, the assembly deviation value corresponding to each assembly in the assembly history record can be obtained, and the mean value of the deviation can be obtained by averaging it; The average assembly time corresponding to each assembly is calculated to obtain the average assembly length, and the average assembly length is calculated and analyzed with the deviation mean to obtain the assembly performance value of each parking space; Step 3: Get the position of the AGV, and calculate the distance difference between it and the position of each parking space to get the transport distance; get the number of assembly AGVs to be processed in each parking space; retrieve the assembly performance value and operation evaluation index of each parking space closest to the current time of the system, and normalize them with the transport distance and the number of assembly AGVs to be processed and take their values, and analyze the values to get the assembly value; The parking space with the largest assembly value is selected as the calibration parking space, and the assembly task of the calibration AGV is assigned to the corresponding calibration parking space, and the number of pending assembly AGVs in the corresponding calibration parking space increases by one; Step 4: Repeat steps 1 to 3 until all AGV assembly tasks are assigned; Step 5: Each AGV generates an assembly path according to its assigned parking space, and goes to the corresponding parking space for assembly according to the assembly path; every time an AGV assembly task is completed in a parking space, the number of AGVs to be assembled in the parking space is reduced by one, and the assembly start time, assembly end time and assembly 3D image of this assembly are recorded; the assembly start time, assembly end time and assembly 3D image of this assembly are integrated into an assembly history record, and the assembly history record is updated to step 2; The specific process of real-time monitoring and analysis of the assembly equipment in each parking space based on the operation information of each parking space is as follows: 401: Retrieve the operation information corresponding to each acquisition moment, wherein the operation information includes the noise intensity value, actual power value, vibration amplitude and temperature of each part of the assembly equipment; set a standard operation parameter corresponding to each assembly equipment, wherein the standard operation parameter includes standard power and standard amplitude; 402: Comprehensively quantify the interference degree of each part of the assembly equipment according to the temperature thereof to obtain the temperature interference value of the assembly equipment at each collection time; 403: normalizing the noise intensity value, the actual power, the vibration amplitude, the temperature interference value, the standard power and the standard amplitude and taking their values, and analyzing the values to obtain the state value; A two-dimensional rectangular coordinate system is constructed with time as the horizontal coordinate and the state value as the vertical coordinate. The state value is placed at the second point of the coordinate axis according to its corresponding collection time to obtain the state point; a smooth curve is used to connect each state point in sequence to obtain a relationship diagram of the assembly equipment state value of each parking space over time, and based on this, a deeper analysis of the assembly equipment state change of each parking space is carried out to obtain the operation evaluation index, which is specifically: Draw a tangent line at each state point, and calculate the slope of each tangent line to obtain the state slope; The state slopes greater than zero are summed up to obtain the state increase value, and the state slopes less than zero are summed up and then the absolute value is taken to obtain the state decrease value; Substitute the state value STi, state slope Li, state increase value A1 and state decrease value A2 at each acquisition time into the set formula The operation evaluation index Ag is calculated, where g1, g2, and g3 are the set proportional coefficients. is the mean value of the state slope of each state point, i=1,2,3…I, I is a positive integer, I represents the total number of acquisition moments, and i represents the sequence number of any acquisition moment.
2. According to claim 1, a heavy-duty frame assembly AGV intelligent parking system is characterized in that: The specific process of feature matching analysis is as follows: 201: Set a number of feature points in the standard three-dimensional image, thereby obtaining the three-dimensional coordinates of each feature point in the standard three-dimensional image, and record it as P 标准 (X 标准 j, Y 标准 j, Z 标准 j), where j = 1, 2, 3...J, j is the serial number of any feature point, J represents the total number of feature points, and J is a positive integer; 202: identifying feature points corresponding to those in the standard three-dimensional image in the assembled three-dimensional image, thereby forming feature point pairs, and calculating the position deviation of each pair of feature point pairs; when the position deviation of the feature point pair is greater than zero, the feature point pair is recorded as a deviation feature point pair, thereby obtaining a plurality of deviation feature point pairs; 203: Compare and analyze the position deviation with the set deviation interval. When the position deviation is greater than the maximum value in the set deviation interval, a high deviation point pair is accumulated; when the position deviation is in the set deviation interval, a medium deviation point pair is accumulated; when the position deviation is less than the minimum value in the set deviation interval, a low deviation point pair is accumulated; the accumulated numbers of high deviation point pairs, medium deviation point pairs and low deviation point pairs are counted respectively, and the position deviations of the high deviation point pairs, medium deviation point pairs and low deviation point pairs are summed up to obtain the high deviation value, medium deviation value and low deviation value respectively; The cumulative number of high deviation point pairs, the cumulative number of medium deviation point pairs, the cumulative number of low deviation point pairs, the high deviation value, the medium deviation value and the low deviation value are formulated and analyzed to obtain the assembly deviation value.
3. The heavy-duty frame assembly AGV intelligent parking system according to claim 2 is characterized in that: The specific process of identifying feature points and calculating position deviations in the assembly 3D image is as follows: 301: Feature point recognition: In the assembly 3D image, use the algorithm to identify representative feature points; 302: Extracting geometric information: extracting geometric information of feature points, using the nearest neighbor algorithm, preliminarily matching feature points in the assembled 3D image corresponding to feature points in the standard 3D image, and calculating the Euclidean distance of each pair of matched feature points, preliminarily screening out a number of feature points whose position distances from feature points in the standard 3D image are within a set range, thereby forming preliminary feature point pairs; 303: Use the RANSAC algorithm to remove incorrectly matched point pairs, and determine the optimal point pair matching through random sampling and fitting; 304: Use the ICP algorithm to perform fine alignment, repeatedly iteratively minimize the distance between the corresponding feature points in the assembled 3D image and the standard 3D image until convergence; thus, the feature points in the assembled 3D image corresponding to the feature points in the standard 3D image are recorded as P (Xj, Yj, Zj); 305: Through the Euclidean space distance formula The position deviation ΔPj of each feature point is obtained.
4. The heavy-duty frame assembly AGV intelligent parking system according to claim 1 is characterized in that: The specific process of comprehensively quantifying the interference degree of each part of the assembly equipment according to the temperature is as follows: Set a temperature sensitivity coefficient corresponding to each part of the assembly equipment, and multiply the temperature of each part by the corresponding temperature sensitivity coefficient to obtain the sensitivity value of each part; Compare and analyze the sensitivity value of each part with the set sensitivity range to divide the parts corresponding to the sensitivity value into highly sensitive parts, moderately sensitive parts and low sensitive parts, count the number of highly sensitive parts, moderately sensitive parts and low sensitive parts in the assembly equipment respectively, and sum and calculate the sensitivity values of the highly sensitive parts, moderately sensitive parts and low sensitive parts respectively to obtain the highly sensitive value, moderately sensitive value and low sensitive value; The number of highly sensitive parts, the number of moderately sensitive parts, the number of low sensitive parts and the high sensitivity value, the moderate sensitivity value and the low sensitivity value are calculated and analyzed by formulating to obtain the temperature interference value, thereby obtaining the temperature interference value at each collection time.
Citation Information
Patent Citations
Handling robot and mechanical three-dimensional garage using the handling robot
CN103089046A
Multi-AGV scheduling method and system based on artificial intelligence and visual perception
CN113342004A
Equipment assembly method and system, equipment and medium
CN117215372A
Path planning method for substation inspection robot
EP3940494A1
Method, depot device and system for transferring a plurality of packages to a plurality of recipients and / or for receiving a plurality of packages from a plurality of deliverers
WO2022106598A1
