An intelligent logistics park unmanned logistics distribution vehicle route navigation method and system

Abstract

The application relates to the technical field of route navigation, in particular to an intelligent logistics park unmanned logistics distribution vehicle route navigation method and system, which comprises the following steps: acquiring the current actual position, historical position and planned distribution route of an unmanned vehicle and signal labels; obtaining each signal monitoring group according to the distance distribution condition between the historical position of the unmanned vehicle and the signal labels and the straight-line distance between the signal labels and the planned distribution route; solving the suspected vehicle position according to the overlapping condition between the signal strength ranges of the signal labels in the signal monitoring group; analyzing the error distribution range according to the obstacle blocking degree of the signal labels in the signal monitoring group, and obtaining the current corrected position by combining the information credibility of the overlapping area between the error distribution ranges; and performing route navigation on the unmanned vehicle according to the current corrected position and the current actual position. The application makes the identification result of the current position of the unmanned vehicle more accurate, and thus more accurate navigation results can be obtained.

Description

Technical Field

[0001] This invention relates to the field of route navigation technology, specifically to a route navigation method and system for unmanned logistics delivery vehicles in intelligent logistics parks. Background Technology

[0002] With the development of the online economy, the logistics industry has also become increasingly developed, and logistics parks, as transit points for goods, shoulder an important mission. Due to the large area and frequent flow of goods in logistics parks, manual transportation costs have always been high. With the gradual popularization of autonomous vehicles, replacing manual labor with unmanned logistics vehicles can significantly improve the delivery efficiency of logistics parks. However, because logistics parks often include indoor areas, metal shelves, and densely packed goods, GPS signals are easily affected by obstruction and multipath effects, leading to positioning errors. Traditional navigation solutions do not consider the impact of signal obstruction, resulting in a decrease in the navigation accuracy of delivery vehicles. Summary of the Invention

[0003] To address the technical problem of reduced navigation accuracy for delivery vehicles due to the failure of traditional navigation schemes to consider signal obstruction, the present invention aims to provide a route navigation method and system for unmanned logistics delivery vehicles in intelligent logistics parks. The specific technical solution adopted is as follows: In a first aspect, the present invention provides a method for navigating routes of unmanned delivery vehicles in intelligent logistics parks, comprising: Obtain the current actual location, historical location, and planned delivery route of the unmanned vehicle, as well as the signal tag corresponding to each RFID tag identified by the unmanned vehicle; Based on the distance distribution between the historical location of the unmanned vehicle and each signal tag, as well as the straight-line distance between each signal tag and the planned delivery route, the degree of obstacle occlusion of each signal tag is analyzed, and the signal tags are grouped to obtain each signal monitoring group; Based on the overlap between the signal strength ranges of each signal tag within each signal monitoring group, the suspected vehicle locations for each signal monitoring group can be determined. Based on the degree of obstruction of signal tags in each signal monitoring group, the error distribution range of the corresponding suspected vehicle position is analyzed. Combined with the credibility of the information of the overlapping area between the error distribution ranges of suspected vehicle positions in each signal monitoring group, the current corrected position of the unmanned vehicle is obtained. The unmanned vehicle is guided to navigate the planned delivery route based on its current corrected location and its current actual location.

[0004] Preferably, the step of analyzing the error distribution range of the suspected vehicle position based on the degree of obstacle obstruction of the signal tags in each signal monitoring group, and combining the reliability of the information of the overlapping areas between the error distribution ranges of the suspected vehicle positions in each signal monitoring group to obtain the current corrected position of the unmanned vehicle specifically includes: Based on the distribution of the degree of obstruction of each signal tag within each signal monitoring group, the error range of the suspected vehicle location for each signal monitoring group is obtained; All grids covered by the error areas corresponding to all signal monitoring groups are recorded as locations to be screened. Based on the obstacle occlusion of each signal tag in each signal monitoring group corresponding to the error area of ​​each location to be screened, the information credibility of all error areas is superimposed to obtain the credibility weight of each location to be screened. The suspected vehicle location corresponding to the maximum confidence weight of each location to be screened is taken as the current corrected location of the unmanned vehicle.

[0005] Preferably, the step of obtaining the error region of the suspected vehicle position for each signal monitoring group based on the distribution of obstacle obstruction degree of each signal tag within each signal monitoring group specifically includes: The average degree of obstruction of all signal tags in each signal monitoring group is used as the obstruction feature value of each signal monitoring group; the adjustment coefficient of each signal monitoring group is determined based on the ratio between the obstruction feature value of each signal monitoring group and the minimum value of all obstruction feature values. The error length of each signal monitoring group is obtained by multiplying the adjustment coefficient and the preset size for each signal monitoring group; a circular area with the suspected vehicle position of each signal monitoring group as the center and the error length as the radius is the error area of ​​the suspected vehicle position of each signal monitoring group.

[0006] Preferably, the step of summing up the information credibility of all error regions based on the obstacle occlusion status of each signal tag in each signal monitoring group corresponding to the error region of each location to be screened, to obtain the credibility weight of each location to be screened, specifically includes: The negative correlation coefficient of the occlusion feature value of each signal monitoring group is used as the feature confidence of the error region corresponding to each signal monitoring group; all signal monitoring groups that contain each location to be screened within the error region are obtained as reference monitoring groups for each location to be screened, and the sum of the feature confidence of the error region corresponding to all reference monitoring groups for each location to be screened is used as the confidence weight of the location to be screened.

[0007] Preferably, the step of analyzing the degree of obstacle obstruction of each signal tag based on the distance distribution between the historical location of the unmanned vehicle and each signal tag, and the straight-line distance between each signal tag and the planned delivery route, and grouping the signal tags to obtain each signal monitoring group, specifically includes: Based on the shortest straight-line distance between the location of each signal tag and the planned delivery route, determine the mapping position corresponding to each signal tag on the planned delivery route; Based on the straight-line distance between the historical location of the unmanned vehicle and the mapped location of each signal tag, and the route distance on the planned delivery route, combined with the shortest straight-line distance, the degree of obstacle occlusion of each signal tag is obtained; Based on the distribution of the degree of obstruction for each signal tag, all signal tags are grouped to obtain each signal monitoring group.

[0008] Preferably, the step of determining the obstacle occlusion degree of each signal tag based on the straight-line distance between the historical location of the unmanned vehicle and the mapped location of each signal tag, and the route distance on the planned delivery route, combined with the shortest straight-line distance, specifically includes: The route distance on the planned delivery route between the historical location of the unmanned vehicle and the mapped location of each signal tag is used as the first feature distance; The straight-line distance between the historical location of the unmanned vehicle and the mapped location of each signal tag is used as the second feature distance; the shortest straight-line distance between the location of each signal tag and the planned delivery route is used as the third feature distance. The degree of obstacle occlusion for each signal tag is obtained based on the first feature distance, the second feature distance, and the third feature distance. The first feature distance and the third feature distance are both positively correlated with the degree of obstacle occlusion, while the second feature distance is negatively correlated with the degree of obstacle occlusion.

[0009] Preferably, the step of grouping all signal tags into each signal monitoring group based on the distribution of the degree of obstruction for each signal tag specifically includes: All signal tags are arranged in ascending order of obstruction level to form a signal tag sequence. Each pair of adjacent signal tags in the signal tag sequence constitutes a signal monitoring group.

[0010] Preferably, the step of determining the suspected vehicle location for each signal monitoring group based on the overlap between the signal strength ranges of each signal tag within each signal monitoring group specifically includes: For any signal monitoring group, the straight-line distance between the current actual position of the unmanned vehicle and the position of each signal tag is taken as the signal strength distribution length of each signal tag; Using the length of the signal strength distribution of each signal tag as the radius and the location of each signal tag as the center, obtain the position of each intersection point between the circular areas of all signal tags in the signal monitoring group; The intersection point closest to the historical location of the unmanned vehicle is taken as the suspected vehicle location for any of the signal monitoring groups.

[0011] Preferably, the method for obtaining the planned delivery route of the unmanned vehicle specifically includes: Based on the start and end positions of the current delivery task of the unmanned vehicle, the shortest planned path of the unmanned vehicle is obtained using a path planning algorithm, which serves as the planned delivery route for the unmanned vehicle.

[0012] Secondly, the present invention provides a route navigation system for unmanned logistics delivery vehicles in an intelligent logistics park, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the computer program is executed by the processor, it implements the steps of a route navigation method for unmanned logistics delivery vehicles in an intelligent logistics park.

[0013] The embodiments of the present invention have at least the following beneficial effects: This invention first acquires data to provide a foundation for subsequent data analysis. By analyzing the distance distribution between signal tags identified by the unmanned vehicle and their historical locations, the signal occlusion between each tag and the vehicle is assessed. Then, based on the degree of obstacle occlusion affecting different tag types, signal monitoring groups with varying location recognition effectiveness are formed. Next, by analyzing the overlap of signal strength distribution ranges, the possible location of the vehicle monitored by each signal monitoring group can be dynamically determined. Furthermore, by fusing the positioning information from multiple signal monitoring groups and considering the impact of obstacle occlusion on data reliability, the most probable vehicle coordinates are determined using probability density principles, thus establishing the current corrected position. Finally, by combining the current corrected position with the actual position of the unmanned vehicle, an adaptive navigation method can be determined. This invention fully considers the potential occlusion effects on each radio frequency signal, resulting in more accurate identification of the unmanned vehicle's current location and consequently, more accurate navigation results. Attached Figure Description

[0014] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1This is a flowchart illustrating the steps of a method for navigating unmanned delivery vehicles in an intelligent logistics park, provided by the present invention. Figure 2 This is a flowchart of the steps of the method for acquiring each signal monitoring group provided by the present invention; Figure 3 This is a schematic diagram of the mapping position corresponding to the signal tag provided by the present invention; Figure 4 This is a flowchart of the steps involved in solving the suspected vehicle location for each signal monitoring group provided by the present invention; Figure 5 This is a schematic diagram of the suspected vehicle location of a signal monitoring group provided by the present invention; Figure 6 This is a flowchart of the steps for obtaining the current corrected position of an unmanned vehicle provided by the present invention; Figure 7 This is a schematic diagram of the error range of the suspected vehicle location corresponding to the signal monitoring group provided by the present invention. Detailed Implementation

[0016] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a route navigation method and system for unmanned logistics delivery vehicles in an intelligent logistics park proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0017] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0018] The following description, in conjunction with the accompanying drawings, details a specific scheme for a route navigation method and system for unmanned logistics delivery vehicles in an intelligent logistics park provided by the present invention.

[0019] Please see Figure 1 The diagram illustrates a flowchart of a method for navigating unmanned delivery vehicles in an intelligent logistics park, according to an embodiment of the present invention. The method includes the following steps: Step S100: Obtain the current actual location, historical location, and planned delivery route of the unmanned vehicle, as well as the signal tag corresponding to each radio frequency tag identified by the unmanned vehicle.

[0020] First, a planar coordinate system is established for the logistics park. The current actual position of the unmanned vehicle refers to its coordinates at the current moment, while its historical position refers to its coordinates at the previous historical moment adjacent to the current moment. Radio Frequency Identification (RFID) tags are an automatic identification technology that enables contactless data interaction via radio waves. Each RFID tag corresponds to a signal tag that includes the tag's location coordinates and its ID.

[0021] It should be noted that the distribution of RFID tags can be set by the implementer according to the specific scenario within the logistics park. As an example, RFID tags are placed on both sides of the road in the high-precision operation area of ​​the logistics park for route guidance during the unmanned vehicle's operation. The spacing between RFID tags should not exceed 50m. A tag reader / writer is installed at the front of the unmanned vehicle, supporting long-distance identification. The reader / writer only reads from each RFID tag, without writing to it. The identification radius is no less than 50m, and the reading rate supported by a single reader / writer is no less than 20 tags / second.

[0022] The planned delivery route of unmanned vehicles can be obtained by path planning based on the start and end positions of the delivery task. Specifically, based on the start and end points of the current delivery task of the unmanned vehicle, the shortest planned path of the unmanned vehicle is obtained by using a path planning algorithm, which serves as the planned delivery route of the unmanned vehicle.

[0023] More specifically, the path planning method is a well-known technique, and will only be briefly introduced here. The planar map of the logistics park is rasterized, and obstacles and the start and end positions of the current delivery task of the unmanned vehicle are marked on the rasterized map. Then, using A... The algorithm plans the delivery route for unmanned vehicles to obtain the shortest planned path. Each grid cell has an actual side length of 1 meter. When marking obstacles, the edges of the obstacles can be enlarged to avoid the planned path being too close to the obstacle's location; for example, the radius of the enlarged obstacle edge is at least one grid cell.

[0024] It should be noted that in this embodiment, the time interval for route deviation guidance is set to 50ms, that is, the time interval between adjacent moments is 50ms. Implementers can set this according to their specific implementation scenarios. It should be understood that the signal tags corresponding to the RFID tags involved in subsequent steps refer to the signal tags corresponding to all RFID tags identified by the unmanned vehicle at the current moment.

[0025] Typically, route navigation involves several steps: first, obtaining the planned route; second, acquiring the vehicle's real-time location information to determine its position and deviation within the planned route; and third, combining this with historical route data to determine the direction of deviation. In this process, if the distance between the vehicle and the RFID tag can be accurately obtained, only the coordinates of the two known RFID tags are needed, and the vehicle's coordinates can be obtained using the triangulation method. However, in most cases, the RFID tag and the vehicle are not completely unobstructed. Different obstacles may exist between different RFID tags and the vehicle, causing variations in signal attenuation between the tags, making it difficult to accurately determine the distance between the vehicle and the tag.

[0026] Since the globally planned routes are typically obstacle-free, when a vehicle detects an RFID tag, it can project the tag's location onto the nearest planned route. The longer the distance from the projection point to the vehicle's current location, the more obstacles exist between the RFID tag and the vehicle's current location, potentially leading to higher signal attenuation. The vehicle's current location is the precise location ultimately desired. Considering the limited distance traveled by the vehicle during continuous positioning monitoring, the vehicle's location from the previous positioning moment can be used for calculation.

[0027] Therefore, firstly, based on the obtained global planning path and the signal tags currently identified by the vehicle, the obstacle occlusion between each signal tag and the vehicle's position is evaluated; secondly, the vehicle's suspected coordinate position is obtained by locating the monitoring combination of different signal tags obtained under different obstacle occlusion conditions, and the vehicle's accurate positioning information is obtained by fusion; finally, route guidance is achieved through accurate positioning information and planning path, and the specific method is implemented from steps S200 to S500.

[0028] Step S200: Based on the distance distribution between the historical location of the unmanned vehicle and each signal tag, and the straight-line distance between each signal tag and the planned delivery route, analyze the degree of obstacle occlusion of each signal tag, and group the signal tags to obtain each signal monitoring group.

[0029] Within logistics parks, dynamic obstacles such as goods and vehicles frequently change, making it impossible to predict obstacle distribution by directly analyzing tag coordinates. However, planned paths are statically pre-planned; by mapping locations, the impact of dynamic obstacles can be transformed into a dynamic difference between the planned distance and the straight-line distance, facilitating real-time calculation. The planned delivery route for unmanned vehicles is the optimal route to avoid obstacles, theoretically without any obstacles. When the location of a signal tag identified by the vehicle is mapped onto the planned delivery route, the closer the planned distance between the mapped location of the signal tag and the unmanned vehicle is to the straight-line distance, the fewer obstacles exist along the path between the unmanned vehicle and the mapped location of the signal tag. This allows for the grouping and arrangement of all signal tags identified by the unmanned vehicle.

[0030] As a concrete example, such as Figure 2 As shown, the acquisition method for each signal monitoring group can be implemented by steps S201 to S203.

[0031] Step S201: Based on the shortest straight-line distance between the location of each signal tag and the planned delivery route, determine the mapping position corresponding to each signal tag on the planned delivery route.

[0032] It should be understood that the distance between the location of each signal tag identified by the autonomous vehicle and its mapped location is equal to the shortest straight-line distance between the location of each signal tag and the planned delivery route. More specifically, a perpendicular line is drawn from the coordinate point corresponding to the location of each signal tag to the planned delivery route; the foot of this perpendicular line is the mapped location of the corresponding signal tag on the planned delivery route. Figure 3 As shown, the mapping location reflects the location of each signal tag to the nearest location on the accessible route, providing a data foundation for subsequent analysis of obstacle occlusion.

[0033] Step S202: Based on the straight-line distance between the historical location of the unmanned vehicle and the mapped location of each signal tag, and the route distance on the planned delivery route, combined with the shortest straight-line distance, the degree of obstacle occlusion of each signal tag is obtained.

[0034] Considering that the displacement distance of the unmanned vehicle at adjacent detection times is relatively small compared to the distance between the unmanned vehicle and the signal tag, this embodiment uses the location of the unmanned vehicle at the previous time adjacent to the current time for feature analysis during the obstacle occlusion analysis process.

[0035] Specifically, the route distance between the historical location of the unmanned vehicle and the mapped location of each signal tag on the planned delivery route is used as the first feature distance; the straight-line distance between the historical location of the unmanned vehicle and the mapped location of each signal tag is used as the second feature distance; the shortest straight-line distance between the location of each signal tag and the planned delivery route is used as the third feature distance; the degree of obstacle occlusion of each signal tag is obtained based on the first feature distance, the second feature distance, and the third feature distance. The first feature distance and the third feature distance are both positively correlated with the degree of obstacle occlusion, and the second feature distance is negatively correlated with the degree of obstacle occlusion.

[0036] As a concrete example, taking any signal tag identified by an autonomous vehicle as an example, the calculation process of the obstacle occlusion degree of the m-th signal tag can be expressed as follows: ,in This indicates the degree of obstruction for the m-th signal tag. This represents the route distance on the planned delivery route between the historical location of the unmanned vehicle and the mapped location of the m-th signal tag. This represents the straight-line distance between the historical location of the unmanned vehicle and the mapped location of the m-th signal tag. This represents the shortest straight-line distance between the location of the m-th signal tag and the planned delivery route. Norm is a normalization function. For example, the maximum and minimum value normalization method can be used. The maximum and minimum values ​​in the normalization process can be obtained by statistically analyzing the ratios calculated between the historical locations of the unmanned vehicle and each signal tag. The value is a preset minimum positive number. In order to avoid division by zero error when the denominator is 0, the value can be 0.1 in this embodiment.

[0037] like Figure 3 As shown, the first feature distance That is, the solid line between the unmanned vehicle and the mapped location ( Figure 3 The length of the planned route (in the middle) reflects the actual unobstructed route distribution between the historical location of the unmanned vehicle and the signal tag, and the second feature distance... That is, the dashed line between the unmanned vehicle and the mapped location ( Figure 3 The length of the mapped route (in the middle) reflects the ideal, unobstructed route distribution between the historical location of the unmanned vehicle and the signal tag.

[0038] The closer the actual route distribution between the historical location of the autonomous vehicle and the signal tag is to the ideal route distribution, the fewer obstacles obstruct the signal tag and the location of the autonomous vehicle, and the lower the impact of obstacle obstruction on the signal strength of the signal tag. When evaluating the obstacle obstruction between each signal tag and the autonomous vehicle, the distance distribution between the signal tag's location and the mapped location is also considered, that is, the shortest straight-line distance between the signal tag's location and the planned delivery route. The larger this distance, the farther the signal tag deviates from the barrier-free planned route, and the more potential obstacle areas (such as stacks of goods, metal shelves, etc.) the signal needs to cross on its propagation path, thus increasing the probability and degree of obstruction. The shorter this distance distribution, the better the mapping effect, and the more accurate the analysis results of obstacle obstruction.

[0039] It should be noted that all straight-line distances in this step can be calculated using the Euclidean distance between the coordinates of two different locations, and this calculation method is a well-known technique, so it will not be described in detail here.

[0040] Step S203: Based on the distribution of the degree of obstruction of each signal tag, group all signal tags to obtain each signal monitoring group.

[0041] The greater the obstacle occlusion degree of each signal tag, the more severe the obstacle occlusion between the corresponding signal tag and the historical position of the unmanned vehicle. Therefore, all signal tags identified by the unmanned vehicle can be sorted by obstacle occlusion degree, and the current position of the unmanned vehicle can be located using the signal tags with less obstacle occlusion. At the same time, considering that it is impossible to completely avoid the influence of obstacle occlusion on signal tags in the actual environment, in order to improve the accuracy of the unmanned vehicle's positioning results, this embodiment considers the combined identification of multiple signal tags.

[0042] Specifically, all signal tags are arranged in ascending order of obstacle occlusion to form a signal tag sequence. Each pair of adjacent signal tags in the sequence forms a signal monitoring group. For example, if the unmanned vehicle identifies N signal tags at the current moment, the first and second signal tags in the sequence form the first signal monitoring group, the second and third signal tags form the second signal monitoring group, and so on, resulting in a total of N-1 signal monitoring groups.

[0043] Step S300: Based on the overlap between the signal strength ranges of each signal tag within each signal monitoring group, determine the suspected vehicle location for each signal monitoring group.

[0044] The signal strength emitted by RFID tags attenuates with increasing propagation distance, and the degree of attenuation is related to obstacle occlusion. By analyzing the intersection of the signal strength distribution ranges of two tags within the same signal monitoring group, the location and identification capabilities of each group's RFID signal for unmanned vehicles can be determined. In other words, a single tag can only determine if a vehicle is on a circle, while dual tags can narrow down the range to two points through the intersection of the circles. Combining this with historical location data, the location of a suspected vehicle can be uniquely determined. Simultaneous monitoring of paired tags allows for mutual verification of signal reliability, reducing false positives caused by single tag occlusion.

[0045] As a concrete example, such as Figure 4 As shown, the process of solving the suspected vehicle location for each signal monitoring group can be implemented by steps S301 to S303.

[0046] Step S301: For any signal monitoring group, the straight-line distance between the current actual position of the unmanned vehicle and the position of each signal tag is taken as the signal strength distribution length of each signal tag.

[0047] The length of the signal strength distribution of each signal tag represents the relative positional distribution between the location of each signal tag and the unmanned vehicle.

[0048] In other embodiments, considering that the strength of the radio frequency signal attenuates with increasing distance during propagation, the tag signal strength can be converted into distance by establishing a distance-signal strength curve. More specifically, the correspondence between radio frequency signal strength and distance is fitted using the radio frequency signal strength corresponding to different distances in a historical database to obtain a distance-signal strength curve. Then, based on the strength of the tag signal received by the unmanned vehicle at the current moment, the distance corresponding to that strength is found on the curve. This represents the distance distribution between the current location of the unmanned vehicle and the location of each signal tag, which is also the signal strength distribution length of each signal tag. The data fitting method can employ least squares, etc., and the implementer can choose according to the specific implementation scenario.

[0049] Step S302: Using the signal strength distribution length of each signal tag as the radius and the location of each signal tag as the center, obtain the position of each intersection point between the circular areas of all signal tags in the signal monitoring group.

[0050] The signal strength is equal on the circle corresponding to each signal tag. According to the three-point positioning method, the intersection of the two circles is the distribution location of the unmanned vehicle. However, there are two intersection points between the two circles, so it is necessary to filter the location of the two intersection points by combining the historical location of the unmanned vehicle.

[0051] It should be noted that the method of finding the intersection of two circles based on the circular areas corresponding to two signal tags within the same signal monitoring group is a well-known technique and will not be described in detail here.

[0052] Step S303: The intersection point closest to the historical location of the unmanned vehicle is taken as the suspected vehicle location of any one of the signal monitoring groups.

[0053] like Figure 5 As shown, S1 and S2 represent the locations of two different signal tags within the same signal monitoring group, and L1 and L2 represent the signal strength distribution lengths corresponding to signal tag S1 and signal tag S2, respectively.

[0054] It should be understood that the distance between different locations can be obtained by calculating the Euclidean distance using the coordinates of the corresponding locations.

[0055] Step S400: Analyze the error distribution range of the suspected vehicle position based on the degree of obstacle obstruction of the signal tags in each signal monitoring group, and combine the information credibility of the overlapping areas between the error distribution ranges of the suspected vehicle positions in each signal monitoring group to obtain the current corrected position of the unmanned vehicle.

[0056] The location of a suspected vehicle obtained by a single signal monitoring group may be affected by obstacles, leading to errors in the localization and identification results. Therefore, by superimposing the error ranges of multiple signal monitoring groups, the true location of the vehicle is more likely to appear in the common coverage area of ​​multiple reliable suspected vehicle locations. Based on this consideration, this embodiment, when determining the true location of an unmanned vehicle at the current moment, firstly, considers the obstacle occlusion between the signal tag and the unmanned vehicle within a single signal monitoring group to determine the possible error range distribution of the suspected vehicle location identified by that group. Secondly, it considers the superposition of the common coverage of the error ranges of all suspected vehicle locations to determine the true location with the highest confidence.

[0057] As a concrete example, such as Figure 6 As shown, the method for obtaining the current corrected position of the unmanned vehicle can be implemented by steps S401 to S403.

[0058] Step S401: Based on the distribution of the degree of obstruction of each signal tag in each signal monitoring group, the error area of ​​the suspected vehicle position of each signal monitoring group is obtained.

[0059] The degree of obstacle obstruction corresponding to each signal tag characterizes the extent to which the signal strength between the signal tag and the unmanned vehicle is affected by obstacle obstruction. The larger the value of the obstacle obstruction degree corresponding to the signal tag, the greater the error level that the signal tag may correspond to, and consequently, the larger the error range of the suspected vehicle location identified by the signal monitoring group where the tag's signal is located.

[0060] Specifically, the method for obtaining the error region of the suspected vehicle location for each signal monitoring group is as follows: the average obstruction degree of all signal tags in each signal monitoring group is used as the obstruction feature value of each signal monitoring group; the adjustment coefficient of each signal monitoring group is determined based on the ratio between the obstruction feature value of each signal monitoring group and the minimum value of all obstruction feature values; the error length of each signal monitoring group is obtained by multiplying the adjustment coefficient of each signal monitoring group by a preset size; and a circular area with the suspected vehicle location of each signal monitoring group as the center and the error length as the radius is the error region of the suspected vehicle location of each signal monitoring group.

[0061] As a concrete example, taking any signal monitoring group as an example, the method for obtaining the adjustment coefficient of the g-th signal monitoring group can be expressed by the formula: ,in, This represents the adjustment coefficient for the g-th signal monitoring group. This represents the occlusion characteristic value of the g-th signal monitoring group. This represents the minimum value of the occlusion characteristic value corresponding to all signal monitoring groups. This is a preset minimum positive number. To avoid division by zero errors when the minimum value is 0, the value can be 0.1 in this embodiment.

[0062] As a specific example, this embodiment sets the preset size to the side length of one grid, which is 1 meter. The error region of the suspected vehicle position for each signal monitoring group indicates that the probability of the vehicle's actual position falling within that region decreases with increasing distance. A lower occlusion feature value corresponds to an error region of the signal monitoring group, meaning that the location near the suspected vehicle position of that signal monitoring group is more likely to be the actual vehicle position. Therefore, the smaller the occlusion feature value of the signal monitoring group, the smaller the range of error in the corresponding signal monitoring group, and the higher the accuracy and reliability of the suspected vehicle position corresponding to that signal monitoring group. It should be noted that a value of 0 for the adjustment coefficient of the signal monitoring group indicates that there are no obstacles between all signal tags in the signal monitoring group and the unmanned vehicle, the signal propagation path is a straight line, and there are no multipath effects or abnormal signal attenuation. At this time, the signal strength ranging accuracy of the RFID tag reaches the theoretical maximum value, and the suspected vehicle position solved by the dual-tag intersection method has almost no error.

[0063] Step S402: Record several grids covered by the error area corresponding to all signal monitoring groups as locations to be screened; based on the obstacle occlusion of each signal tag in each signal monitoring group corresponding to the error area of ​​each location to be screened, sum up the information credibility of all error areas to obtain the credibility weight of each location to be screened.

[0064] Specifically, the negative correlation coefficient of the occlusion feature value of each signal monitoring group is used as the feature confidence of the error region corresponding to each signal monitoring group; all signal monitoring groups that contain each location to be screened within the error region are obtained as reference monitoring groups for each location to be screened, and the sum of the feature confidence of the error regions corresponding to all reference monitoring groups for each location to be screened is used as the confidence weight of the location to be screened. Here, a reference monitoring group for each location to be screened means that, for the grid cell containing the location to be screened, the error regions of all reference monitoring groups corresponding to that location cover that grid cell.

[0065] It should be noted that, as described in step S100, when planning the path for unmanned vehicles, the floor plan of the logistics park is divided into several grids. When screening vehicle locations, all grids covered by the error range of all suspected vehicle locations are taken as analysis objects. Within the local area of ​​each grid, the greater the superposition result of reliable information that the error area of ​​the suspected vehicle location falls into the local area, the greater the probability that the location corresponding to the grid is the real vehicle location.

[0066] As a concrete example, taking any signal monitoring group as an example, the negative correlation coefficient of the occlusion feature value of the g-th signal monitoring group can be expressed as: , This represents the occlusion characteristic value of the g-th signal monitoring group. This represents the normalization function. For example, the maximum and minimum values ​​can be normalized using the maximum and minimum value normalization method, where the maximum and minimum values ​​of the normalization process are obtained by statistically analyzing the occlusion feature values ​​of all signal monitoring groups.

[0067] The occlusion feature value of each signal monitoring group represents the average occlusion situation within the signal monitoring group. The larger the value of the occlusion feature value, the more severe the occlusion of the signal tag of the corresponding signal monitoring group is by obstacles. Consequently, the reliability of the identification result of the suspected vehicle location corresponding to that signal monitoring group is lower, that is, the lower the value of the corresponding feature reliability.

[0068] like Figure 7As shown in the figure, a, b, c, d and e are the suspected vehicle locations corresponding to different signal monitoring groups, and they are obtained according to the order of the signal label sequence. The obstacle occlusion degree of the signal monitoring group corresponding to suspected vehicle location a is small, the data accuracy is high, and the distribution of the vehicle's true coordinate range is also small.

[0069] Step S403: The suspected vehicle location corresponding to the maximum value of the confidence weight of each location to be screened is taken as the current corrected location of the unmanned vehicle.

[0070] By fusing positioning information from multiple signal monitoring groups and considering the impact of obstacle occlusion on data reliability, the most probable vehicle location coordinates are determined using probability density principles. Directly targeting the location corresponding to a grid cell, the reliability information from multiple signal monitoring groups is integrated to filter out high-density areas, thus determining the high-probability location corresponding to the actual vehicle position—which is the current corrected position.

[0071] This step considers the varying degrees of obstruction affecting different signal monitoring groups. Signal monitoring groups with less obstruction experience lower occlusion levels, resulting in higher accuracy in calculating suspected vehicle locations. Therefore, a higher density weight, or feature confidence level, is assigned to their error regions. The feature confidence level is reflected by the negative correlation coefficient of the mean obstruction level; lower obstruction levels result in higher feature confidence. Next, the feature confidence levels of all error regions on each grid are summed to obtain the total density weight for each grid, which is also the confidence weight. Because the probability of the vehicle's true location being in the overlapping area of ​​multiple suspected intersection error ranges is higher, selecting the grid point with the highest density weight determines the vehicle's current location. This method transforms multiple uncertain single-point measurements into a probabilistically optimal solution, effectively improving the accuracy and reliability of positioning.

[0072] Step S500: Navigate the unmanned vehicle by determining the distance and direction to the planned delivery route based on its current corrected location and actual location.

[0073] Specifically, the grid cell with the shortest straight-line distance between the current corrected position of the unmanned vehicle and the planned delivery route is obtained on the planned delivery route and used as the target position for the unmanned vehicle. The direction in which the unmanned vehicle travels from its current actual position to the target position is the current direction of travel for the unmanned vehicle. It should be understood that the target position is located on the planned delivery route.

[0074] Furthermore, to avoid frequent sharp turns caused by real-time deviations, it is also necessary to evaluate the vehicle's historical travel direction. Specifically, the vehicle's travel direction at the previous moment adjacent to the current moment is obtained. The direction of the bisector of the angle between the current travel direction of the unmanned vehicle and the travel direction at the previous moment is used as the estimated guidance direction for the unmanned vehicle. The unmanned vehicle travels according to the estimated guidance direction until it enters the planned delivery route, and then performs the delivery task according to the planned delivery route.

[0075] Based on the predicted guidance direction, the unmanned vehicle operates through its control and driving module. Simultaneously, the unmanned vehicle is equipped with an active braking system, primarily consisting of ultrasonic radar, used to monitor for obstacles around the vehicle. When the ultrasonic signal reflection is strong, the vehicle actively decelerates and stops, waiting for the ultrasonic signal reflection to return to normal. After obtaining a local guidance path through the aforementioned steps, the vehicle is controlled to continue driving until it reaches its final destination.

[0076] This invention also provides a route navigation system for unmanned delivery vehicles in intelligent logistics parks, including a memory, a processor, and a computer program stored in the memory and running on the processor. When executed by the processor, the computer program implements the steps of a route navigation method for unmanned delivery vehicles in intelligent logistics parks. Since an embodiment of a route navigation method for unmanned delivery vehicles in intelligent logistics parks has already been described in detail, it will not be elaborated further here.

[0077] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for navigating unmanned delivery vehicles in an intelligent logistics park, characterized in that, The method includes the following steps: Obtain the current actual location, historical location, and planned delivery route of the unmanned vehicle, as well as the signal tag corresponding to each RFID tag identified by the unmanned vehicle; Based on the distance distribution between the historical location of the unmanned vehicle and each signal tag, and the straight-line distance between each signal tag and the planned delivery route, the degree of obstacle occlusion of each signal tag is analyzed. The signal tags are then grouped to obtain each signal monitoring group. This includes: determining the mapping position of each signal tag on the planned delivery route based on the shortest straight-line distance between the location of each signal tag and the planned delivery route; obtaining the degree of obstacle occlusion of each signal tag based on the straight-line distance between the historical location of the unmanned vehicle and the mapping position of each signal tag, as well as the route distance on the planned delivery route, combined with the shortest straight-line distance; and grouping all signal tags according to the distribution of the degree of obstacle occlusion of each signal tag to obtain each signal monitoring group. Based on the overlap between the signal strength ranges of each signal tag within each signal monitoring group, the suspected vehicle locations for each signal monitoring group can be determined. The error distribution range of suspected vehicle locations is analyzed based on the degree of obstruction of signal tags within each signal monitoring group. Combining this with the information reliability of overlapping areas between the error distribution ranges of suspected vehicle locations within each signal monitoring group, the current corrected location of the unmanned vehicle is obtained. This process includes: determining the error region of suspected vehicle locations for each signal monitoring group based on the distribution of obstruction of each signal tag within each group; recording several grid cells covered by the error regions corresponding to all signal monitoring groups as locations to be screened; summing the information reliability of all error regions based on the obstruction of each signal tag within each signal monitoring group corresponding to the error region of each location to be screened to obtain a reliability weight for each location to be screened; and using the suspected vehicle location corresponding to the maximum reliability weight of each location to be screened as the current corrected location of the unmanned vehicle. The unmanned vehicle is guided to navigate the planned delivery route based on its current corrected location and its current actual location.

2. The method for navigating unmanned delivery vehicles in an intelligent logistics park according to claim 1, characterized in that, The method of obtaining the error range of suspected vehicle positions for each signal monitoring group based on the distribution of obstacle obstruction levels for each signal tag within each signal monitoring group specifically includes: The average degree of obstruction of all signal tags in each signal monitoring group is used as the obstruction feature value of each signal monitoring group; the adjustment coefficient of each signal monitoring group is determined based on the ratio between the obstruction feature value of each signal monitoring group and the minimum value of all obstruction feature values. The error length of each signal monitoring group is obtained by multiplying the adjustment coefficient and the preset size for each signal monitoring group; a circular area with the suspected vehicle position of each signal monitoring group as the center and the error length as the radius is the error area of ​​the suspected vehicle position of each signal monitoring group.

3. The method for navigating unmanned delivery vehicles in an intelligent logistics park according to claim 2, characterized in that, The step involves summing the information credibility of all error regions based on the obstacle occlusion status of each signal tag within each signal monitoring group corresponding to the error region of each location to be screened, to obtain the credibility weight of each location to be screened. Specifically, this includes: The negative correlation coefficient of the occlusion feature value of each signal monitoring group is used as the feature confidence of the error region corresponding to each signal monitoring group; all signal monitoring groups that contain each location to be screened within the error region are obtained as reference monitoring groups for each location to be screened, and the sum of the feature confidence of the error region corresponding to all reference monitoring groups for each location to be screened is used as the confidence weight of the location to be screened.

4. The method for navigating unmanned delivery vehicles in an intelligent logistics park according to claim 1, characterized in that, The method of determining the obstacle occlusion level of each signal tag based on the straight-line distance between the historical location of the unmanned vehicle and the mapped location of each signal tag, the route distance on the planned delivery route, and the shortest straight-line distance, specifically includes: The route distance on the planned delivery route between the historical location of the unmanned vehicle and the mapped location of each signal tag is used as the first feature distance; The straight-line distance between the historical location of the unmanned vehicle and the mapped location of each signal tag is used as the second feature distance; the shortest straight-line distance between the location of each signal tag and the planned delivery route is used as the third feature distance. The degree of obstacle occlusion for each signal tag is obtained based on the first feature distance, the second feature distance, and the third feature distance. The first feature distance and the third feature distance are both positively correlated with the degree of obstacle occlusion, while the second feature distance is negatively correlated with the degree of obstacle occlusion.

5. The method for route navigation of unmanned logistics delivery vehicles in an intelligent logistics park according to claim 1, characterized in that, The process of grouping all signal tags into signal monitoring groups based on the distribution of obstacle obstruction levels for each tag specifically includes: All signal tags are arranged in ascending order of obstruction level to form a signal tag sequence. Each pair of adjacent signal tags in the signal tag sequence constitutes a signal monitoring group.

6. The method for navigating unmanned delivery vehicles in an intelligent logistics park according to claim 5, characterized in that, The step of determining the suspected vehicle location for each signal monitoring group based on the overlap between the signal strength ranges of each signal tag within each group specifically includes: For any signal monitoring group, the straight-line distance between the current actual position of the unmanned vehicle and the position of each signal tag is taken as the signal strength distribution length of each signal tag; Using the length of the signal strength distribution of each signal tag as the radius and the location of each signal tag as the center, obtain the position of each intersection point between the circular areas of all signal tags in the signal monitoring group; The intersection point closest to the historical location of the unmanned vehicle is taken as the suspected vehicle location for any of the signal monitoring groups.

7. The method for navigating unmanned delivery vehicles in an intelligent logistics park according to claim 1, characterized in that, The method for obtaining the planned delivery routes of the unmanned vehicles specifically includes: Based on the start and end positions of the current delivery task of the unmanned vehicle, the shortest planned path of the unmanned vehicle is obtained using a path planning algorithm, which serves as the planned delivery route for the unmanned vehicle.

8. A route navigation system for unmanned delivery vehicles in an intelligent logistics park, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the computer program is executed by the processor, it implements the steps of the route navigation method for unmanned logistics delivery vehicles in an intelligent logistics park as described in any one of claims 1-7.

Images