A forklift identification area method and system based on ground coding

Abstract

This invention belongs to the field of warehousing and logistics automation technology. It provides a method and system for forklift identification area based on ground coding. When a tightly coupled relationship is identified, the optimal empty forklift identification coordinates are obtained. Based on these coordinates, matching tests are performed on unmanned forklifts under different load conditions to evaluate the matching degree of empty coordinates. This helps determine the applicable range of the optimal empty forklift identification coordinates and whether personalized customization is needed for unmanned forklifts under different load conditions. In cases of low matching of empty coordinates, the optimal empty forklift identification coordinates are used as a benchmark to obtain the identification adjustment amount for unmanned forklifts under different load conditions. The optimal identification coordinates for unmanned forklifts under different load conditions are then determined, and a forklift identification coordinate sequence under different load conditions is constructed. This improves the dynamic matching capability and identification accuracy of the ground coding for unmanned forklift identification under different load conditions.

Description

Technical Field

[0001] This invention belongs to the field of warehouse logistics automation technology, specifically a method and system for forklift identification area based on ground coding. Background Technology

[0002] In automated warehousing and logistics scenarios, unmanned forklifts have become a key technology for area positioning and path planning by recognizing ground codes. However, if the spacing between ground codes at adjacent intersections is too close or their clarity differs (e.g., worn codes, uneven lighting), it can easily lead to a "jump" phenomenon during camera recognition—the forklift may misjudge the target intersection, causing a steering error. Traditional solutions rely solely on single-dimensional parameters (such as spacing thresholds or contrast thresholds) to assess risk, lacking a comprehensive quantitative model that considers both spacing and clarity factors, making it impossible to accurately predict the probability of such jumps.

[0003] Changes in forklift load alter the camera's tilt angle, causing a shift in the effective recognition distance (diagonal length of the camera's field of view). Current technology lacks a dynamic correlation mechanism between load status and recognition parameters. The optimal recognition coordinates determined under empty conditions may become invalid under full load due to tilt angle changes, leading to recognition errors. A single coordinate system is insufficient to adapt to the entire load range from empty to full load. When the optimal empty coordinates are applied to a full load state, a "low-match" phenomenon often occurs—the recognition jump risk value exceeds a threshold, requiring manual intervention or path adjustment, impacting operational efficiency. Current technology lacks a systematic analysis of the coupling relationship between "recognition jump risk, driving position, and effective recognition distance," resulting in an imbalance between operational efficiency and safety.

[0004] Therefore, the present invention provides a method and system for forklift identification area based on ground coding. Summary of the Invention

[0005] In order to overcome the shortcomings of the prior art, at least one technical problem raised in the background art is solved.

[0006] The technical solution adopted by this invention to solve its technical problem is: A forklift identification region method based on ground coding, comprising: When an unmanned forklift in an unloaded state passes through a T-junction by recognizing ground codes, the ground codes in each intersection are extracted and analyzed from the dimensions of the spacing between adjacent ground codes and the clarity between adjacent ground codes to determine the degree of recognition jump risk of the unloaded forklift during the turning process. When assessing the risk of high recognition jump, the adjacent coding distance data is extracted as a reference, a simulation test period is set, the recognition jump risk level of the forklift in different driving positions under no-load conditions is extracted, and the coupling correlation between the recognition jump risk level at each driving position and the effective recognition distance is analyzed. When the situation is determined to be tightly coupled, the optimal empty forklift identification coordinates are obtained, and based on the optimal empty forklift identification coordinates, matching tests are conducted on unmanned forklifts under different load conditions to evaluate the empty coordinate matching degree. In the case of low matching of empty coordinates, the recognition adjustment amount of unmanned forklifts under different load conditions is obtained based on the best empty forklift recognition coordinates. The best recognition coordinates of unmanned forklifts under different load conditions are obtained, and a forklift recognition coordinate sequence under different load conditions is constructed.

[0007] As a further aspect of the present invention, the process for determining the risk level of a sudden change during a turn of an unloaded forklift is as follows: Construct a spatial coordinate system for code recognition, and extract the center point of the ground code within each intersection as the code coordinate; Extract the coding coordinates on both sides of the horizontal X-axis in the coding recognition spatial coordinate system, and obtain the distance between the coding coordinates as the coding interval; If the coding interval is less than or equal to the effective recognition interval, the difference between the effective recognition interval and the coding interval is calculated, and then the ratio is calculated with the effective recognition interval to obtain the coding interval interference value. Extract the ground code on each side of the coded coordinates, and obtain the black and white contrast value of each side of the ground code. Subtract the values ​​to obtain the black and white contrast difference value. The black-and-white contrast difference ratio is calculated by comparing the black-and-white contrast difference with the black-and-white contrast threshold. The interference value of the coding interval is summed with the black-and-white contrast difference ratio to obtain the identification jump risk value. If the identification jump risk value is greater than the identification jump risk threshold, it is displayed as a high identification jump risk signal.

[0008] As a further aspect of the present invention, the coupling correlation analysis process between the degree of recognition jump risk and the effective recognition distance at each driving position is as follows: The set simulation test cycle is equally divided into several simulation test monitoring points. The identification jump risk value corresponding to the monitoring no-load coordinate of each unit is extracted. The identification jump risk values ​​corresponding to the monitoring no-load coordinate of adjacent units are combined into a risk analysis group to obtain multiple risk analysis groups. By combining the unloaded coordinates of adjacent units into a coordinate analysis group, multiple coordinate analysis groups are obtained. Within the coordinate analysis group, the distance between the monitoring empty coordinates of adjacent units is obtained as the spacing between adjacent empty coordinates, and the ratio is calculated with the length of the corresponding prescribed travel path of the unmanned forklift to obtain the spacing ratio between adjacent empty coordinates. Within the risk analysis group, the difference between the identified jump risk values ​​corresponding to the unloaded coordinates of each unit is calculated to obtain the jump risk difference value. The correlation analysis value is obtained by calculating the ratio of the jump risk difference to the distance between adjacent spatial coordinates.

[0009] As a further aspect of the present invention, the evaluation process for coupling correlation is as follows: The mean of the association analysis is obtained by summing and averaging all the association analysis values. Calculate the standard deviation of all correlation analysis values ​​to obtain the correlation analysis standard deviation; The standard deviation and mean of the correlation analysis were calculated using the coefficient of variation formula to obtain the coupling correlation evaluation value. If the coupling correlation evaluation value is less than or equal to the coupling correlation evaluation threshold, it is displayed as a tightly correlated signal.

[0010] As a further aspect of the present invention, the process for obtaining the optimal coordinates for identifying an empty forklift is as follows: Extract the unit monitoring no-load coordinates at each simulated test monitoring point, obtain the identification jump risk value corresponding to each unit monitoring no-load coordinate, compare the magnitudes, and select the unit monitoring no-load coordinates corresponding to the minimum identification jump risk value as the optimal no-load forklift identification coordinates.

[0011] As a further aspect of the present invention, the process of performing matching tests on unmanned forklifts under different load conditions is as follows: When a fully loaded unmanned forklift travels to the optimal identification coordinates for an unloaded forklift, the corresponding identification jump risk value is obtained and compared with the identification jump risk threshold. If the identification jump risk value is greater than the identification jump risk threshold, it indicates that the fully loaded unmanned forklift has a high probability of identification jump when turning at a T-junction due to the coding spacing being too close or the clarity difference. This is displayed as a low-degree matching signal of the unit. The identification jump risk value is subtracted from the identification jump risk threshold and then the ratio is calculated to obtain the identification jump risk level value. Following the analysis method of unmanned forklifts under full load traveling to the optimal empty forklift identification coordinates, the identification of unmanned forklifts under different load conditions traveling to the optimal empty forklift identification coordinates is analyzed, and the proportion of the total number of low-degree matching signals of the unit to the total number of matching signals of all units is extracted as the low-degree matching number ratio.

[0012] As a further aspect of the present invention, the evaluation process for the unloaded coordinate matching degree is as follows: The low-match degree value is obtained by summing and averaging the identification jump risk level values ​​corresponding to the low-match signal of each unit. The sum of the ratio of low-score matches and the low-score match degree value is used to obtain the match test analysis value. If the match test analysis value is greater than the match test analysis threshold, it is displayed as a highly matched signal.

[0013] As a further aspect of the present invention, the process of obtaining the recognition adjustment amount of the unmanned forklift under different load conditions is as follows: Extract the mean of the association analysis as the association coefficient; When the unmanned forklift is fully loaded, extract the risk level of the recognition jump when the unmanned forklift is at the optimal recognition coordinates of the unmanned forklift under the full load state, and calculate the ratio with the correlation coefficient to obtain the full load recognition adjustment amount. Based on the method of obtaining the full-load recognition adjustment amount, the recognition adjustment amount corresponding to the unmanned forklift under different load conditions is obtained.

[0014] As a further aspect of the present invention, the process of constructing the forklift identification coordinate sequence under different load conditions is as follows: Based on the optimal empty forklift recognition coordinates, the spatial distance of the full-load recognition adjustment is added to obtain the optimal full-load forklift recognition coordinates. According to the method of obtaining the optimal full-load and empty forklift recognition coordinates, the optimal recognition coordinates corresponding to unmanned forklifts under different load conditions are obtained. And according to the load level, the optimal recognition coordinates corresponding to unmanned forklifts under different load conditions are sorted to construct a forklift recognition coordinate sequence under different load conditions.

[0015] A forklift identification area system based on ground coding, comprising: Jump Risk Assessment Module: When an unmanned forklift in an unloaded state passes through a T-junction by recognizing ground codes, the module extracts the ground codes in each intersection and analyzes them from the dimensions of the spacing between adjacent ground codes and the clarity between adjacent ground codes to determine the degree of recognition jump risk of the unloaded forklift during the turning process. Coupling and Correlation Analysis Module: When assessing the risk of high recognition jump, the adjacent coding distance data is extracted as a reference, a simulation test cycle is set, the recognition jump risk level of the forklift in different driving positions under no-load conditions is extracted, and the coupling correlation between the recognition jump risk level at each driving position and the effective recognition distance is analyzed. Coordinate matching evaluation module: When the situation is determined to be tightly coupled, the optimal empty forklift identification coordinates are obtained, and based on the optimal empty forklift identification coordinates, matching tests are conducted on unmanned forklifts under different load conditions to evaluate the empty coordinate matching degree. Recognition sequence construction module: Under the condition of low matching of empty coordinates, the recognition adjustment amount of unmanned forklifts under different load conditions is obtained based on the best empty forklift recognition coordinates. The optimal recognition coordinates of unmanned forklifts under different load conditions are obtained, and the recognition coordinate sequence of forklifts under different load conditions is constructed.

[0016] The beneficial effects of this invention are as follows: This invention, when an unmanned forklift in an unloaded state is recognizing ground codes at a T-junction, extracts the ground codes within each intersection and analyzes them from the dimensions of spacing and clarity between adjacent ground codes to determine the recognition jump risk level of the unloaded forklift during turning. It quantifies the probability of recognition jumps caused by excessively close code spacing or clarity differences when the unmanned forklift turns at a T-junction. When assessing a high recognition jump risk, the distance data between adjacent codes is extracted as a reference, and a simulation test cycle is set to extract the recognition jump risk level of the forklift in an unloaded state at different driving positions. The coupling correlation between the recognition jump risk level and the effective recognition distance at each driving position is analyzed to obtain the correlation pattern between the recognition jump risk level and the effective recognition distance of the unmanned forklift at different driving positions, improving the accuracy of the unmanned forklift's automatic driving. Furthermore, it allows for a high-speed straight-through strategy in low-risk areas and a low-speed turning or avoidance strategy in high-risk areas, achieving dynamic optimization of path planning.

[0017] When a tightly coupled relationship is identified, this invention obtains the optimal empty forklift identification coordinates and performs matching tests on unmanned forklifts under different load conditions based on these coordinates. The matching degree of the empty coordinates is evaluated, which helps determine the applicable range of the optimal empty forklift identification coordinates and whether personalized customization is needed for unmanned forklifts under different load conditions. In cases of low matching of empty coordinates, the optimal empty forklift identification coordinates are used as a benchmark to obtain the identification adjustment amount for unmanned forklifts under different load conditions. The optimal identification coordinates for unmanned forklifts under different load conditions are then determined, and a forklift identification coordinate sequence under different load conditions is constructed. This improves the dynamic matching capability and recognition accuracy of the ground coding for unmanned forklift identification under different load conditions. Attached Figure Description

[0018] The invention will now be further described with reference to the accompanying drawings.

[0019] Figure 1 This is a flowchart of the steps of a forklift identification area method based on ground coding according to the present invention; Figure 2 This is a flowchart of a forklift identification area system based on ground coding according to the present invention. Detailed Implementation

[0020] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments. Example

[0021] When an automated forklift needs to turn at a T-junction after recognizing ground codes, the ground codes at each intersection are located near the center. During the forklift's camera recognition of these codes, the proximity and inconsistent clarity of codes at adjacent intersections can cause jumps in the recognition process, preventing the forklift from turning at the designated intersection. Furthermore, the forklift's load status can affect the camera's angle of view during recognition, leading to inaccurate identification. Therefore, please refer to [further details needed]. Figure 1 As shown in the embodiment of the present invention, a forklift identification area method based on ground coding includes the following steps: Step 1: When an unmanned forklift is in an unloaded state and is recognizing ground codes at a T-junction, extract the ground codes in each intersection and analyze them from the dimensions of the spacing between adjacent ground codes and the clarity between adjacent ground codes to determine the degree of recognition jump risk of the unloaded forklift during the turning process. In some embodiments, the center point of the intersection center area within the T-junction is extracted as the origin of the two-dimensional spatial coordinate system, with the horizontal adjacent intersection channel as the spatial X-axis and the vertical adjacent intersection channel as the spatial Y-axis, to construct the coding recognition spatial coordinate system; Extract the center point of the ground code within each intersection as the code coordinate; Extract the coding coordinates on both sides of the horizontal X-axis in the coding recognition spatial coordinate system, and obtain the distance between the coding coordinates as the coding interval; If the coding interval is greater than the effective recognition area interval, it means that the unmanned forklift is less affected by the recognition jump interference when it passes through the camera at the T-junction, and it is not considered. If the coding interval is less than or equal to the effective recognition interval, it means that the unmanned forklift is greatly affected by the recognition jump interference when it is recognized by the camera at the T-junction. Therefore, the difference between the effective recognition interval and the coding interval is calculated and then the ratio is calculated with the effective recognition interval to obtain the coding interval interference value. It should be noted that the effective recognition interval refers to the diagonal field of view of the camera recognition coding area on the unmanned forklift. Since the camera recognition area on the unmanned forklift is fixed, the effective recognition interval is also fixed. Extract the ground code on each side of the coded coordinates, and obtain the black and white contrast value of each side of the ground code. Subtract the values ​​to obtain the black and white contrast difference value. It should be noted that the black-and-white contrast difference is obtained by subtracting the larger black-and-white contrast value on one side from the smaller black-and-white contrast value on the other side, and it is a positive number. The black-and-white contrast difference ratio is calculated by comparing the black-and-white contrast difference with the black-and-white contrast threshold. It should be noted that the black-and-white contrast threshold refers to the black-and-white contrast of the symbol camera during encoding and recognition. The identification jump risk value is obtained by summing the coding gap interference value and the black-and-white contrast difference ratio. It is understandable that the identification jump risk value represents the following meaning: it is used to quantify the potential risk of identification jump caused by the close spacing of adjacent codes and inconsistent clarity when an unmanned forklift identifies ground codes through a camera at a T-junction. On the one hand, the code spacing interference value reflects the degree of physical overlap risk caused by the close distance of adjacent ground codes at the intersection. On the other hand, the black and white contrast difference ratio reflects the degree of identification reliability risk caused by the difference in clarity of adjacent codes. Specifically, if the identification jump risk value is larger, it means that the probability of identification jump caused by the close spacing of codes or the difference in clarity when the unmanned forklift turns at a T-junction is higher. If the identification jump risk value is smaller, it means that the probability of identification jump caused by the close spacing of codes or the difference in clarity when the unmanned forklift turns at a T-junction is lower. If the identification jump risk value is greater than the identification jump risk threshold, it means that the unmanned forklift in the unloaded state has a high probability of identification jump when turning at a T-junction due to the coding spacing being too close or the clarity difference, and is displayed as a high identification jump risk signal. If the identified jump risk value is less than or equal to the identified jump risk threshold, it indicates that the probability of the unmanned forklift making a jump when turning at a T-junction due to the coding spacing being too close or the clarity difference is low, and it is displayed as a low identified jump risk signal.

[0022] Step 2: When assessing the risk of high recognition jump, extract the adjacent coding distance data as a reference, set a simulation test cycle, extract the recognition jump risk level of the forklift in different driving positions under no-load conditions, and analyze the coupling correlation between the recognition jump risk level at each driving position and the effective recognition distance. It should be noted that the adjacent code spacing data refers to the spacing between adjacent ground codes; In some embodiments, the set simulation test period is equally divided into several simulation test monitoring points, wherein the interval between adjacent simulation test monitoring points is equal. The unloaded coordinates of the forklift in the unloaded state at each simulated test monitoring point are obtained in the coding recognition space coordinate system and used as the unit monitoring unloaded coordinates. It should be noted that each simulated test monitoring point corresponds to one unit monitoring no-load coordinate; Extract the identification jump risk value corresponding to the unloaded coordinate of each unit monitoring, and combine the identification jump risk values ​​corresponding to the unloaded coordinate of adjacent units monitoring into a risk analysis group to obtain multiple risk analysis groups; By combining the unloaded coordinates of adjacent units into a coordinate analysis group, multiple coordinate analysis groups are obtained. Within the coordinate analysis group, the distance between the monitoring empty coordinates of adjacent units is obtained as the spacing between adjacent empty coordinates, and the ratio is calculated with the length of the corresponding prescribed travel path of the unmanned forklift to obtain the spacing ratio between adjacent empty coordinates. Within the risk analysis group, the difference between the identified jump risk values ​​corresponding to the unloaded coordinates of each unit is calculated to obtain the jump risk difference value. The correlation analysis value is obtained by calculating the ratio of the jump risk difference to the distance between adjacent spatial coordinates; It should be noted that the risk analysis group corresponding to the jump risk difference and the coordinate analysis group corresponding to the adjacent spatial coordinate spacing ratio both correspond to the same adjacent simulated test monitoring points. For example, if the risk analysis group corresponding to the jump risk difference is obtained from simulated test monitoring points A and B, then the coordinate analysis group corresponding to the adjacent spatial coordinate spacing ratio is also obtained from simulated test monitoring points A and B. The mean of the association analysis is obtained by summing and averaging all the association analysis values. Calculate the standard deviation of all correlation analysis values ​​to obtain the correlation analysis standard deviation; The standard deviation and mean of the correlation analysis were calculated using the coefficient of variation formula to obtain the coupling correlation evaluation value. It is understandable that the coupling correlation evaluation value represents the core indicator for measuring the stability of the correlation between the identified jump risk value and the effective identification distance at different driving positions. In essence, it quantifies the dispersion of the correlation analysis value through the coefficient of variation. Specifically, if the coupling correlation evaluation value is larger, it means that the correlation between the identified jump risk value and the effective identification distance is less consistent at different driving positions. If the coupling correlation evaluation value is smaller, it means that the correlation between the identified jump risk value and the effective identification distance is more consistent at different driving positions. If the coupling correlation evaluation value is greater than the coupling correlation evaluation threshold, it indicates that the correlation between the identified jump risk value and the effective identification distance is not consistent across different driving positions, and is displayed as a non-closely correlated signal. If the coupling correlation evaluation value is less than or equal to the coupling correlation evaluation threshold, it indicates that the correlation between the identified jump risk value and the effective identification distance is highly consistent across different driving positions, showing a closely correlated signal.

[0023] The specific solution in this embodiment is as follows: When the unmanned forklift is in an unloaded state and passes through a T-junction by recognizing ground codes, the ground codes in each intersection are extracted, and the distance between adjacent ground codes and the clarity between adjacent ground codes are analyzed to determine the recognition jump risk level of the unloaded forklift during the turning process. This can quantify the recognition jump probability caused by the unmanned forklift turning at the T-junction due to the code spacing being too close or the clarity difference. When assessing a high recognition jump risk, the distance data between adjacent codes is extracted as a reference, a simulation test period is set, and the recognition jump risk level of the forklift in the unloaded state at different driving positions is extracted. The coupling correlation between the recognition jump risk level at each driving position and the effective recognition distance is analyzed to obtain the correlation law between the recognition jump risk level and the effective recognition distance of the unmanned forklift at different driving positions, thereby improving the accuracy of the unmanned forklift's automatic driving. Moreover, it can adopt a high-speed straight-through strategy in low-risk areas and a low-speed turning or avoidance strategy in high-risk areas, realizing dynamic optimization of path planning. Example

[0024] Please see Figure 1 As shown in the embodiment of the present invention, a forklift identification area method based on ground coding further includes the following steps: Step 3: When the situation is determined to be tightly coupled, obtain the optimal empty forklift identification coordinates, and based on the optimal empty forklift identification coordinates, conduct matching tests on unmanned forklifts under different load conditions to evaluate the empty coordinate matching degree. In some embodiments, the unit monitoring no-load coordinates at each simulated test monitoring point are extracted, and the identification jump risk value corresponding to each unit monitoring no-load coordinate is obtained. The values ​​are compared, and the unit monitoring no-load coordinates corresponding to the minimum identification jump risk value are selected as the optimal no-load forklift identification coordinates. For example, when a fully loaded unmanned forklift travels to the optimal empty forklift identification coordinates, the corresponding identification jump risk value is obtained and compared with the identification jump risk threshold. If the identification jump risk value is greater than the identification jump risk threshold, it means that the fully loaded unmanned forklift has a high probability of identification jump when turning at a T-junction due to the coding spacing being too close or the clarity difference. This is displayed as a low-degree matching signal of the unit. The identification jump risk value is subtracted from the identification jump risk threshold and then the ratio is calculated with the identification jump risk threshold to obtain the identification jump risk level value. If the identified jump risk value is less than or equal to the identified jump risk threshold, it means that the probability of the unmanned forklift under full load turning at a T-junction due to the coding spacing being too close or the clarity difference is low, and it is displayed as a unit height matching signal. Following the analysis method of unmanned forklifts under full load traveling to the optimal empty forklift identification coordinates, the identification of unmanned forklifts under different load conditions traveling to the optimal empty forklift identification coordinates is analyzed, and the proportion of the total number of low-degree matching signals of the unit to the total number of matching signals of all units is extracted as the low-degree matching number ratio. It should be noted that the total number of unit matching signals refers to the sum of the number of unit low-degree matching signals and unit high-degree matching signals displayed. The low-match degree value is obtained by summing and averaging the identification jump risk level values ​​corresponding to the low-match signal of each unit. The sum of the ratio of low-score matches and the low-score match degree value is used to obtain the match test analysis value. It is understandable that the meaning of the matching test analysis value is: the core indicator for measuring the comprehensive matching effect of the optimal empty forklift identification coordinates under different load conditions. By quantifying the proportion of low matching cases and the severity of risk, it reflects the applicability and reliability of the coordinates in multi-load scenarios. Specifically, if the matching test analysis value is larger, it means that the frequency of low matching is higher and the risk of identification jump is higher in the unmanned forklift simulation test of the optimal empty forklift identification coordinates under different load conditions. If the matching test analysis value is smaller, it means that the frequency of low matching is lower and the risk of identification jump is lower in the unmanned forklift simulation test of the optimal empty forklift identification coordinates under different load conditions. If the matching test analysis value is greater than the matching test analysis threshold, it indicates that during the simulation test of unmanned forklifts under different load conditions, the frequency of low matching is relatively high and the risk of identification jump is relatively high, which is displayed as a low matching signal. If the matching test analysis value is less than or equal to the matching test analysis threshold, it indicates that the optimal empty forklift identification coordinates are tested in the unmanned forklift simulation test under different load conditions. The low matching frequency is low and the risk of identification jump is low, which shows a high matching signal.

[0025] Step 4: Under the condition of low matching of empty coordinates, based on the best empty forklift recognition coordinates, obtain the recognition adjustment amount of unmanned forklifts under different load conditions, obtain the best recognition coordinates of unmanned forklifts under different load conditions, and construct the forklift recognition coordinate sequence under different load conditions. In some embodiments, the mean of the association analysis is extracted as the association coefficient; For example, when the unmanned forklift is fully loaded, the identification jump risk level value of the unmanned forklift under the full load state at the optimal empty forklift identification coordinate is extracted, and the ratio is calculated with the correlation coefficient to obtain the full load identification adjustment amount. The optimal coordinates for identifying a fully loaded forklift are obtained by adding the spatial distance for adjusting the fully loaded forklift recognition based on the optimal coordinates for identifying an empty forklift. Based on the method of obtaining the optimal recognition coordinates of fully loaded and empty forklifts with full load recognition adjustment, the recognition adjustment amount corresponding to unmanned forklifts under different load conditions is obtained. The optimal recognition coordinates corresponding to unmanned forklifts under different load conditions are obtained. According to the load level, the optimal recognition coordinates corresponding to unmanned forklifts under different load conditions are sorted to construct a forklift recognition coordinate sequence under different load conditions. The specific solution in this embodiment is as follows: When the situation is determined to be tightly coupled, the optimal empty forklift identification coordinates are obtained. Based on the optimal empty forklift identification coordinates, matching tests are performed on unmanned forklifts under different load conditions to evaluate the empty coordinate matching degree. This helps to determine the applicable range of the optimal empty forklift identification coordinates and to determine whether it is necessary to make personalized adjustments for unmanned forklifts under different load conditions. In the case of low matching of empty coordinates, the optimal empty forklift identification coordinates are used as the benchmark to obtain the identification adjustment amount of unmanned forklifts under different load conditions. The optimal identification coordinates of unmanned forklifts under different load conditions are obtained, and a forklift identification coordinate sequence under different load conditions is constructed. This improves the dynamic matching capability and identification accuracy of the ground code for unmanned forklift identification under different load conditions. Example

[0026] Based on the same inventive concept as the forklift identification area method based on ground coding in the foregoing embodiments, such as Figure 2 As shown, this application provides a forklift identification area system based on ground coding, wherein the system specifically includes: Jump Risk Assessment Module: When an unmanned forklift in an unloaded state passes through a T-junction by recognizing ground codes, the module extracts the ground codes in each intersection and analyzes them from the dimensions of the spacing between adjacent ground codes and the clarity between adjacent ground codes to determine the degree of recognition jump risk of the unloaded forklift during the turning process. Coupling and Correlation Analysis Module: When assessing the risk of high recognition jump, the adjacent coding distance data is extracted as a reference, a simulation test cycle is set, the recognition jump risk level of the forklift in different driving positions under no-load conditions is extracted, and the coupling correlation between the recognition jump risk level at each driving position and the effective recognition distance is analyzed. Coordinate matching evaluation module: When the situation is determined to be tightly coupled, the optimal empty forklift identification coordinates are obtained, and based on the optimal empty forklift identification coordinates, matching tests are conducted on unmanned forklifts under different load conditions to evaluate the empty coordinate matching degree. Recognition sequence construction module: Under the condition of low matching of empty coordinates, the recognition adjustment amount of unmanned forklifts under different load conditions is obtained based on the best empty forklift recognition coordinates. The optimal recognition coordinates of unmanned forklifts under different load conditions are obtained, and the recognition coordinate sequence of forklifts under different load conditions is constructed.

[0027] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for identifying a region of a forklift based on ground coding, the method comprising: include: ​ When an unmanned forklift in an unloaded state passes through a T-junction by recognizing ground codes, the ground codes in each intersection are extracted and analyzed from the dimensions of the spacing between adjacent ground codes and the clarity between adjacent ground codes to determine the degree of recognition jump risk of the unloaded forklift during the turning process. When assessing the risk of high recognition jump, the adjacent coding distance data is extracted as a reference, a simulation test period is set, the recognition jump risk level of the forklift in different driving positions under no-load conditions is extracted, and the coupling correlation between the recognition jump risk level at each driving position and the effective recognition distance is analyzed. When the situation is determined to be tightly coupled, the optimal empty forklift identification coordinates are obtained, and based on the optimal empty forklift identification coordinates, matching tests are conducted on unmanned forklifts under different load conditions to evaluate the empty coordinate matching degree. In the case of low matching of empty coordinates, the recognition adjustment amount of unmanned forklifts under different load conditions is obtained based on the best empty forklift recognition coordinates. The best recognition coordinates of unmanned forklifts under different load conditions are obtained, and a forklift recognition coordinate sequence under different load conditions is constructed.

2. The method of claim 1, wherein: The process for determining the risk level of a sudden change during a turn of an unloaded forklift is as follows: Construct a spatial coordinate system for code recognition, and extract the center point of the ground code within each intersection as the code coordinate; Extract the coding coordinates on both sides of the horizontal X-axis in the coding recognition spatial coordinate system, and obtain the distance between the coding coordinates as the coding interval; If the coding interval is less than or equal to the effective recognition interval, the difference between the effective recognition interval and the coding interval is calculated, and then the ratio is calculated with the effective recognition interval to obtain the coding interval interference value. Extract the ground code on each side of the coded coordinates, and obtain the black and white contrast value of each side of the ground code. Subtract the values ​​to obtain the black and white contrast difference value. The black-and-white contrast difference ratio is calculated by comparing the black-and-white contrast difference with the black-and-white contrast threshold. The interference value of the coding interval is summed with the black-and-white contrast difference ratio to obtain the identification jump risk value. If the identification jump risk value is greater than the identification jump risk threshold, it is displayed as a high identification jump risk signal.

3. The method of claim 1, wherein: The coupling correlation analysis process between the recognition jump risk level and the effective recognition distance at each driving position is as follows: The set simulation test cycle is equally divided into several simulation test monitoring points. The identification jump risk value corresponding to the monitoring no-load coordinate of each unit is extracted. The identification jump risk values ​​corresponding to the monitoring no-load coordinate of adjacent units are combined into a risk analysis group to obtain multiple risk analysis groups. By combining the unloaded coordinates of adjacent units into a coordinate analysis group, multiple coordinate analysis groups are obtained. Within the coordinate analysis group, the distance between the monitoring empty coordinates of adjacent units is obtained as the spacing between adjacent empty coordinates, and the ratio is calculated with the length of the corresponding prescribed travel path of the unmanned forklift to obtain the spacing ratio between adjacent empty coordinates. Within the risk analysis group, the difference between the identified jump risk values ​​corresponding to the unloaded coordinates of each unit is calculated to obtain the jump risk difference value. The correlation analysis value is obtained by calculating the ratio of the jump risk difference to the distance between adjacent spatial coordinates.

4. The method of claim 3, wherein: The evaluation process for coupling correlation is as follows: The mean of the association analysis is obtained by summing and averaging all the association analysis values. Calculate the standard deviation of all correlation analysis values ​​to obtain the correlation analysis standard deviation; The standard deviation and mean of the correlation analysis were calculated using the coefficient of variation formula to obtain the coupling correlation evaluation value. If the coupling correlation evaluation value is less than or equal to the coupling correlation evaluation threshold, it is displayed as a tightly correlated signal.

5. The method of claim 1, wherein: The process of obtaining the optimal coordinates for identifying an empty forklift is as follows: Extract the unit monitoring no-load coordinates at each simulated test monitoring point, obtain the identification jump risk value corresponding to each unit monitoring no-load coordinate, compare the magnitudes, and select the unit monitoring no-load coordinates corresponding to the minimum identification jump risk value as the optimal no-load forklift identification coordinates.

6. The method of claim 5, wherein: The process of matching tests for unmanned forklifts under different load conditions is as follows: When a fully loaded unmanned forklift travels to the optimal identification coordinates for an unloaded forklift, the corresponding identification jump risk value is obtained and compared with the identification jump risk threshold. If the identification jump risk value is greater than the identification jump risk threshold, it indicates that the fully loaded unmanned forklift has a high probability of identification jump when turning at a T-junction due to the coding spacing being too close or the clarity difference. This is displayed as a low-degree matching signal of the unit. The difference between the identification jump risk value and the identification jump risk threshold is calculated, and then the ratio is calculated with the identification jump risk threshold to obtain the identification jump risk level value. Following the analysis method of unmanned forklifts under full load traveling to the optimal empty forklift identification coordinates, the identification of unmanned forklifts under different load conditions traveling to the optimal empty forklift identification coordinates is analyzed, and the proportion of the total number of low-degree matching signals of the unit to the total number of matching signals of all units is extracted as the low-degree matching number ratio.

7. The method of claim 6, wherein: The evaluation process for unloaded coordinate matching is as follows: The low-match degree value is obtained by summing and averaging the identification jump risk level values ​​corresponding to the low-match signal of each unit. The sum of the ratio of low-score matches and the low-score match degree value is used to obtain the match test analysis value. If the match test analysis value is greater than the match test analysis threshold, it is displayed as a highly matched signal.

8. The method of claim 1, wherein: The process of obtaining the recognition and adjustment values ​​of the unmanned forklift under different load conditions is as follows: Extract the mean of the association analysis as the association coefficient; When the unmanned forklift is fully loaded, extract the risk level of the recognition jump when the unmanned forklift is at the optimal recognition coordinates of the unmanned forklift under the full load state, and calculate the ratio with the correlation coefficient to obtain the full load recognition adjustment amount. Based on the method of obtaining the full-load recognition adjustment amount, the recognition adjustment amount corresponding to the unmanned forklift under different load conditions is obtained.

9. The ground code-based forklift identification zone method of claim 8, wherein: The process of constructing the forklift identification coordinate sequence under different load conditions is as follows: Based on the optimal empty forklift recognition coordinates, the spatial distance of the full-load recognition adjustment is added to obtain the optimal full-load forklift recognition coordinates. According to the method of obtaining the optimal full-load and empty forklift recognition coordinates, the optimal recognition coordinates corresponding to unmanned forklifts under different load conditions are obtained. And according to the load level, the optimal recognition coordinates corresponding to unmanned forklifts under different load conditions are sorted to construct a forklift recognition coordinate sequence under different load conditions.

10. A ground code based forklift identification zone system, characterized by: Includes the following modules: Jump Risk Assessment Module: When an unmanned forklift in an unloaded state passes through a T-junction by recognizing ground codes, the module extracts the ground codes in each intersection and analyzes them from the dimensions of the spacing between adjacent ground codes and the clarity between adjacent ground codes to determine the degree of recognition jump risk of the unloaded forklift during the turning process. Coupling and Correlation Analysis Module: When assessing the risk of high recognition jump, the adjacent coding distance data is extracted as a reference, a simulation test cycle is set, the recognition jump risk level of the forklift in different driving positions under no-load conditions is extracted, and the coupling correlation between the recognition jump risk level at each driving position and the effective recognition distance is analyzed. Coordinate matching evaluation module: When the situation is determined to be tightly coupled, the optimal empty forklift identification coordinates are obtained, and based on the optimal empty forklift identification coordinates, matching tests are conducted on unmanned forklifts under different load conditions to evaluate the empty coordinate matching degree. Recognition sequence construction module: Under the condition of low matching of empty coordinates, the recognition adjustment amount of unmanned forklifts under different load conditions is obtained based on the best empty forklift recognition coordinates. The optimal recognition coordinates of unmanned forklifts under different load conditions are obtained, and the recognition coordinate sequence of forklifts under different load conditions is constructed.

Images