Method, device, electronic device and medium for aligning route data and path monitoring

Abstract

The present disclosure relates to a method, device, electronic device and medium for aligning route data and path monitoring. The method for aligning route data includes: determining candidate road network roads associated with the planned route information from a road network road library according to the planned route information of a traveling object; calculating the probability of each route between adjacent sections according to the distance between each candidate road network road of adjacent sections, and obtaining a probability matrix containing the probability of routes between all adjacent sections; performing beam search processing on the above probability matrix to obtain target route information in the road network. The above method can be applied to the transportation route supervision of hazardous chemical vehicles or the scenario where artificial intelligence equipment generates the planned route of the actual road network based on voice / text input, so as to realize the alignment of planned route information that may have errors or deviations to the existing road network information, and can overcome the problem of not being able to match the actual road due to input errors or irregularities.

Description

Technical Field

[0001] The present disclosure relates to the field of urban governance and vehicle supervision technology, and in particular to a method, device, electronic device and medium for aligning route data and path monitoring. Background Art

[0002] In the process of urban governance, it is necessary to supervise the driving routes of some vehicles transporting hazardous chemicals. For example, in the scenario of hazardous chemical vehicle supervision in XX City, before the departure of vehicles transporting highly toxic hazardous chemicals, the transport personnel or relevant personnel need to report the planned route information to the regulatory department for filing by entering and filling in the supervision application or manually filling out the request form.

[0003] However, currently, for the reported manually input route information, if there are errors in the manually input route information or if the route information is input according to non-standard human names, the supervision application or its background or manual work cannot confirm which road network in the actual road network constitutes the route that the route information reported by the vehicle corresponds to, which makes it impossible to effectively supervise the above-mentioned vehicles. Summary of the invention

[0004] In order to solve the above technical problems or at least partially solve the above technical problems, embodiments of the present disclosure provide a method, device, electronic device and medium for aligning route data and path monitoring.

[0005] In a first aspect, an embodiment of the present disclosure provides a method for aligning route data. The method for aligning route data includes: determining candidate road networks associated with the planned route information from a road network road library according to the planned route information of a driving object; calculating the probability of each route between adjacent sections according to the distance between each candidate road network road of adjacent sections, and obtaining a probability matrix containing the probability of routes between all adjacent sections; and performing a beam search process on the probability matrix to obtain target route information in the road network.

[0006] According to an embodiment of the present disclosure, the above-mentioned method calculates the probabilities of each route between adjacent road sections based on the distances between each candidate road network road of adjacent road sections to obtain a probability matrix containing the probabilities of routes between all adjacent road sections, including: for any road section among all road sections, calculates the distances between the candidate road network roads of the current road section and the candidate road network roads of the next road section, and obtains a distance vector composed of distance elements corresponding to the current road section; obtains a distance matrix based on the distance vectors of all road sections; performs numerical transformation on the distance vector of each row in the above-mentioned distance matrix to obtain a transformed target distance vector, wherein the sum of the corresponding distance elements before and after the transformation is a fixed value; inputs each target distance vector into a soft maximization function for processing, and outputs the probability distribution value corresponding to each distance element in the above-mentioned target distance vector, wherein the probability distribution value corresponding to the current distance element is used to characterize the probability of the route corresponding to the above-mentioned current distance element; obtains a probability matrix containing the probabilities of routes between all adjacent road sections based on the probability distribution values ​​corresponding to all target distance vectors.

[0007] According to an embodiment of the present disclosure, the distance matrix is ​​obtained based on the distance vectors of all road sections, including: taking the distance vectors of each road section in all road sections as each row of the distance matrix; taking the maximum value of the number of columns of the distance vectors in all road sections as the column scale of the above distance matrix, and supplementing default values ​​in specific distance vectors whose number of columns is less than the above column scale; wherein, when performing numerical transformation on the distance vector of each row, the above default values ​​are not involved in the calculation, and the maximum value of the distance element in the current distance vector is used to make a difference with each distance element to obtain the transformed distance element; when the above target distance vector is input into the soft maximization function for processing, the output probability corresponding to the above default value is a preset value.

[0008] According to an embodiment of the present disclosure, the above-mentioned probability matrix is ​​subjected to beam search processing to obtain target route information, including: for the first row of the above-mentioned probability matrix, the routes corresponding to the first K with the highest probability scores of the routes between each group of adjacent road sections in the current row are determined as candidate routes, K being the set search beam width; for each row in the subsequent rows after the first row of the above-mentioned probability matrix, the conditional probability scores of the routes corresponding to the current row based on the K candidate routes corresponding to the previous row are determined, and the routes corresponding to the first K conditional probability scores are determined as target candidate routes; the route with the highest score among the above-mentioned target candidate routes is determined as the target route information.

[0009] According to an embodiment of the present disclosure, the above-mentioned planned route information includes: a sequence of various road sections passed in sequence. Among them, the above-mentioned determining the candidate road network roads associated with the above-mentioned planned route information from the road network road library according to the planned route information of the driving object includes: performing text extraction on each road section in the above-mentioned planned route information to obtain the road section key information for each road section; and performing screening and matching in the road network road library according to the above-mentioned road section key information to obtain the candidate road network roads for each road section.

[0010] According to an embodiment of the present disclosure, the above-mentioned determining, based on the planned route information of the traveling object, candidate road network roads associated with the above-mentioned planned route information from a road network road library also includes: in the absence of a candidate road network road that matches the road segment key information of a specific road segment, based on the candidate road network roads of an adjacent road segment adjacent to the above-mentioned specific road segment, determining in the above-mentioned road network road library a specific road network road that intersects with the candidate road network roads of the above-mentioned adjacent road segment as a candidate road network road for the above-mentioned specific road segment.

[0011] According to an embodiment of the present disclosure, based on the above-mentioned road section key information, screening and matching are performed in the road network road library to obtain candidate road network roads for each road section, including:

[0012] According to the road segment keyword information, screening and matching are performed in the road network road database to obtain candidate road networks for each road segment; if there is no candidate road network road matching the road segment keyword information of a specific road segment, it is deemed that there is no candidate road network road matching the road segment keyword information of the specific road segment; or,

[0013] Based on the road section keyword information, screening and matching are performed in the road network road library to obtain candidate road network roads for each road section; in the case that there are no candidate road network roads that match the road section keyword information of a specific road section, the road section keyword information of the above-mentioned specific road section is split into more fine-grained road section keyword information; based on the above-mentioned road section keyword information, screening and matching are performed in the road network road library to obtain candidate road network roads for the above-mentioned specific road section; in the case that there are no candidate road network roads that match the road section keyword information of the above-mentioned specific road section, it is deemed that there are no candidate road network roads that match the road section keyword information of the above-mentioned specific road section.

[0014] In a second aspect, an embodiment of the present disclosure provides a method for path monitoring. The path monitoring method includes: using the method of aligning route data as described above to determine the target route information of a driving object; obtaining the actual route information of the driving object; matching the actual route information with the target route information; and generating route deviation warning information when the actual route information does not match the target route information.

[0015] In a third aspect, an embodiment of the present disclosure provides an apparatus for aligning route data. The apparatus for aligning route data includes: a road network matching module, a probability calculation module, and a processing module. The road network matching module is configured to determine candidate road network roads associated with the planned route information from a road network road library according to the planned route information of a driving object. The probability calculation module is configured to calculate the probabilities of each route between adjacent road sections according to the distances between the candidate road network roads of each adjacent road section, and obtain a probability matrix including the probabilities of all routes between adjacent road sections. The processing module is configured to perform beam search processing on the probability matrix to obtain target route information in the road network roads.

[0016] In a fourth aspect, an embodiment of the present disclosure provides an apparatus for path monitoring. The apparatus includes: a road network matching module, a probability calculation module, a processing module, an actual route acquisition module, a route information matching module, and an early warning information generation module. The road network matching module is configured to determine candidate road network roads associated with the planned route information from a road network road library according to the planned route information of a driving object. The probability calculation module is configured to calculate the probabilities of each route between adjacent road sections according to the distances between the candidate road network roads of each adjacent road section, and obtain a probability matrix including the probabilities of all routes between adjacent road sections. The processing module is configured to perform beam search processing on the probability matrix to obtain target route information in the road network roads. The actual route acquisition module is configured to acquire the actual route information of the driving object. The route information matching module is configured to match the actual route information with the target route information. The early warning information generation module is configured to generate an early warning information of route deviation when the actual route information does not match the target route information.

[0017] In a fifth aspect, an embodiment of the present disclosure provides an electronic device. The electronic device includes a processor, a communication interface, a memory, and a communication bus. Among them, the processor, the communication interface, and the memory complete communication with each other through the communication bus; the memory is used for storing a computer program; when the processor executes the program stored in the memory, it implements the method for aligning route data or the method for path monitoring as described above.

[0018] In a sixth aspect, an embodiment of the present disclosure provides a computer-readable storage medium. A computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, it implements the method for aligning route data or the method for path monitoring as described above.

[0019] At least some of the above technical solutions provided by the embodiments of the present disclosure have the following advantages:

[0020] By determining the candidate road network roads associated with the above-mentioned planned route information from the road network road library according to the planned route information of the driving object; in the case that there are errors or deviations in the planned route information, the above-mentioned associated candidate road network roads can be used as one or more candidate associated options, so that when the target route information is subsequently determined, the probability of each route between adjacent sections is further calculated according to the distance between each candidate road network road of the adjacent sections, and a probability matrix containing the probabilities of the routes between all adjacent sections is obtained; by performing beam search processing on the above-mentioned probability matrix, the planned route that is closest to the correct state can be selected from one or more candidate road network roads in each section. The target route information of the information realizes the alignment of the planned route information that may have errors (for example, errors or deviations in the information input by manual voice or text, errors or deviations in the planned route information issued by electronic devices, etc.) with the actual road network information, and can overcome the problem of not being able to match the actual road due to incorrect or irregular filling or input to a certain extent; in the application scenario, the user's actual route can be supervised and warned based on the above-mentioned target route information, which can be applied to the scenario of transportation route supervision of hazardous chemicals vehicles in urban governance or the scenario where artificial intelligence equipment generates a planned route for the actual road network based on voice / text input. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0022] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or related technical descriptions are briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0023] Figure 1 The system architecture of the method for aligning route data and the method for path monitoring applicable to the embodiments of the present disclosure is schematically shown;

[0024] Figure 2 A flowchart of a method for aligning route data according to an embodiment of the present disclosure is schematically shown;

[0025] Figure 3A The detailed implementation flow chart of operation S201 according to an embodiment of the present disclosure is schematically shown;

[0026] Figure 3B Schematically shows a detailed implementation flow chart of operation S201 according to another embodiment of the present disclosure;

[0027] Figure 3C Schematically shows a detailed implementation flow chart of operation S201 according to another embodiment of the present disclosure;

[0028] Figure 3D The detailed implementation flow chart of operation S201 according to another embodiment of the present disclosure is schematically shown;

[0029] Figure 4 The detailed implementation flow chart of operation S202 according to the embodiment of the present disclosure is schematically shown;

[0030] Figure 5 The detailed implementation flow chart of operation S203 according to the embodiment of the present disclosure is schematically shown;

[0031] Figure 6 The flowchart of the path monitoring method according to the embodiment of the present disclosure is schematically shown;

[0032] Figure 7 The structure block diagram of the device for aligning route data according to an embodiment of the present disclosure is schematically shown;

[0033] Figure 8 A structural block diagram schematically shows a path monitoring device according to an embodiment of the present disclosure; and

[0034] Fig. 9 The structural block diagram of the electronic device provided by the embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0035] Embodiments of the present disclosure provide a method, device, electronic device and medium for aligning route data and path monitoring. The method for aligning route data includes: determining, from a road network road library, candidate road networks associated with the planned route information according to the planned route information of a traveling object; calculating, based on the distances between each candidate road network road of adjacent road sections, the probability of each route between adjacent road sections, and obtaining a probability matrix containing the probabilities of routes between all adjacent road sections; and performing beam search processing on the probability matrix to obtain target route information in the road network.

[0036] The above-mentioned path monitoring method includes: using the method of aligning route data as mentioned above to determine the target route information of the driving object; obtaining the actual route information of the above-mentioned driving object; matching the above-mentioned actual route information with the above-mentioned target route information; and generating route deviation warning information when the above-mentioned actual route information does not match the above-mentioned target route information.

[0037] The above method can align planned route information that may contain errors (for example, errors or deviations in information input by manual voice or text, errors or deviations in planned route information issued by electronic devices, etc.) with the actual road network information, and can overcome the problem of being unable to match actual roads due to incorrect or irregular filling or input to a certain extent; in the application scenario, the user's actual route can be supervised and warned based on the above target route information, and can be applied to the scenario of transportation route supervision of hazardous chemicals vehicles in urban management.

[0038] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present disclosure.

[0039] Figure 1 The system architecture of the method for aligning route data and the method for path monitoring applicable to the embodiments of the present disclosure is schematically shown.

[0040] Reference Figure 1 As shown, the system architecture 100 applicable to the method for aligning route data and the method for path monitoring of the embodiment of the present disclosure includes: terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is used to provide a medium for a communication link between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0041] The user can use the terminal devices 101, 102, 103 to interact with the server 105 through the network 104 to receive or send messages, etc. The terminal devices 101, 102, 103 may be installed with route supervision applications.

[0042] Furthermore, other communication client applications may be installed on the above-mentioned terminal devices 101, 102, and 103, such as shopping applications, web browser applications, search applications, video playback applications, music playback applications, instant messaging tools, email clients, social platform software, etc. (for example only).

[0043] Terminal devices 101, 102, and 103 can be various electronic devices with display screens, support voice or text input, and have positioning functions, such as electronic devices including but not limited to vehicles (such as vehicles for transporting hazardous chemicals), smart phones, tablet computers, laptops, desktop computers, smart watches, smart bracelets, etc.

[0044] The server 105 may be a server that provides various services, such as a background management server (only as an example) that provides service support for the route reporting and route alignment functions of the route supervision applications on the terminal devices 101, 102, and 103. The background management server may analyze and process the received data such as the planned route information, and feed back the processing results (such as the target route information in the road network) to the terminal device.

[0045] It should be noted that the method for aligning route data and the method for path monitoring provided in the embodiments of the present disclosure can generally be executed by a server 105 (which can be a traditional application server or a cloud server) or a terminal device 101, 102, 103 with certain computing capabilities.

[0046] In an exemplary implementation scenario, for transport personnel or related personnel on vehicles transporting hazardous chemicals, it is necessary to fill in the planned route information (corresponding account permissions such as upload permissions and permissions to view one's own data) in advance through a route supervision application (which can be filled in by text input or voice input). After the electronic device or the corresponding server that has installed the route supervision application executes the above-mentioned method of aligning route data, the target route information in the road network corresponding to the planned route information can be obtained.

[0047] The administrator of this route supervision application can log in to the corresponding administrator account to obtain the target route information of all vehicles in the supervised area.

[0048] In one embodiment, the route monitoring application can be installed on a vehicle transporting hazardous chemicals. In this case, the route monitoring application can also have the function of actual route monitoring, or can access the data of other map navigation applications or route monitoring applications to obtain actual route information. The server or the vehicle performs the above-mentioned route monitoring method, and generates route deviation warning information when the actual route information does not match the target route information.

[0049] In another embodiment, the above-mentioned route supervision application can be installed on a smart device carried by the transportation personnel or related personnel. During transportation, the transportation personnel or related personnel can connect to the vehicle navigation system via Bluetooth, and the smart device or server processes the reported planned route information and obtains the target route information as an option for the navigation route.

[0050] Although the path supervision of hazardous chemicals is used as an example of an implementation scenario, it should be noted that the method for aligning route data provided by the embodiments of the present disclosure is not limited to the field of path supervision, but can also be applied to scenarios where artificial intelligence devices generate planned routes for actual road networks based on voice / text input, or can also be applied to route alignment in scenarios where there are errors or deviations in the planned route information sent by electronic devices.

[0051] For example, in the field of warehousing and logistics, by inputting planned route information into the intelligent transport robot in the form of text or voice, the target route information is output, which can be used as a basis for the transportation path of the intelligent transport robot. The user can use the above-mentioned transportation path as a whole as the actual transportation path of the intelligent transport robot, or modify individual sections on the above-mentioned transportation path to obtain the modified path as the actual transportation path of the intelligent transport robot.

[0052] The embodiments of the present disclosure are described in detail below with reference to the accompanying drawings.

[0053] A first exemplary embodiment of the present disclosure provides a method for aligning route data.

[0054] Figure 2 The flowchart of the method for aligning route data according to an embodiment of the present disclosure is schematically shown.

[0055] Reference Figure 2 As shown, the method for aligning route data provided by the embodiment of the present disclosure includes the following operations: S201, S202 and S203. The operations S201 to S203 can be performed by Figure 1 The example server 105 or the terminal devices 101, 102, 103 are used for execution.

[0056] In operation S201, according to the planned route information of the traveling object, a candidate road network road associated with the planned route information is determined from a road network road library.

[0057] The moving objects are, for example, vehicles, intelligent transport robots, etc.

[0058] The method of obtaining the above-mentioned planned route information includes but is not limited to the following scenarios: the user may obtain the planned route information of the driving object by inputting it in the form of text or voice, or the planned route information of the driving object may be obtained by filling out a paper form and scanning it into an image or PDF format and then using text recognition to obtain the planned route information of the driving object, or the electronic device may send it to the driving object to obtain the planned route information of the driving object. The planned route information may include multiple road sections, each road section has its own planned road information, and there may be one or more candidate road network roads associated with the planned road information in each road section in the above-mentioned planned route information. The above-mentioned planned route information may deviate from the actual road network information or the description of some road sections may be incorrect.

[0059] For example, the input planning route information is a sequence R, where R = [r 0 ,r 1 ,r 2 ,r 3 ,…,r n-1 ],r 0 ,r 1 ,r 2 ,r 3 ,…,r n-1 They correspond to the planned road information of the first section, the planned road information of the second section, the planned road information of the third section, the planned road information of the fourth section, and so on, and the planned road information of the nth section, where n is the total number of sections and n is a positive integer. For the convenience of description, the i-th (i=0, 1, 2, 3, ..., n-1) section containing planned road information is directly expressed as section r. i .

[0060] For each road segment r i , i = 0, 1, 2, 3, ..., n-1, and the planned road information r of the section i The associated candidate road network is m i The candidate road network roads are in vector form and are expressed as

[0061] In operation S202, the probability of each route between adjacent road segments is calculated according to the distances between each candidate road network road of the adjacent road segments, and a probability matrix containing the probabilities of the routes between all adjacent road segments is obtained.

[0062] The probabilities of routes corresponding to different combinations are calculated based on the distances obtained by combining candidate road network roads of the two adjacent sections, thereby generating a probability matrix.

[0063] Since the adjacent described road segments are sequential and coherent, the probability of the route corresponding to the candidate road network road that is far away from two adjacent road segments obtained by calculation is relatively small, and the probability of the route corresponding to the candidate road network road that is close to each other is relatively large.

[0064] In operation S203, a beam search process is performed on the above probability matrix to obtain target route information in the road network.

[0065] Beam Search (also known as cluster search) is a heuristic graph search algorithm that can use reasonable memory space and quickly locate the global optimal route by performing beam search on the probability matrix, thereby reducing the time and space cost of locating the optimal route (target route information).

[0066] In special cases, when the planned route information contains only one road segment, the target route information is determined based on the similarity between the candidate road network roads of the road segment and the planned route information.

[0067] Based on the above operations S201 to S203, by determining the candidate road network roads associated with the above planned route information from the road network road library according to the planned route information of the driving object; in the case where there is an error or deviation in the planned route information, the above associated candidate road network roads can be used as one or more candidate associated options, so that when the target route information is subsequently determined, the probability of each route between adjacent sections is further calculated according to the distance between each candidate road network road of the adjacent sections, and a probability matrix containing the probability of routes between all adjacent sections is obtained; by performing beam search processing on the above probability matrix, it is possible to obtain a probability matrix containing the probability of routes between all adjacent sections in one or more of each section. The target route information that is closest to the correct planned route information is selected from multiple candidate road networks, so that the planned route information that may have errors (for example, errors or deviations in information input by manual voice or text, errors or deviations in planned route information sent by electronic devices, etc.) is aligned with the actual road network information, and can overcome the problem of not being able to match the actual road due to incorrect or irregular filling / input to a certain extent; in the application scenario, the user's actual route can be supervised and warned based on the above-mentioned target route information, which can be applied to the scenario of transportation route supervision of hazardous chemicals vehicles in urban governance.

[0068] According to an embodiment of the present disclosure, in the above operation S201, based on the planned route information of the traveling object, candidate road network roads associated with the above planned route information are determined from the road network road library, including: performing text extraction on each road section in the above planned route information to obtain road section key information for each road section; based on the above road section key information, screening and matching are performed in the road network road library to obtain candidate road network roads for each road section.

[0069] In some embodiments, the above-mentioned road section key information includes: road section keyword information; in other embodiments, the above-mentioned road section key information includes: road section keyword information and road section keyword information.

[0070] According to an embodiment of the present disclosure, based on the above-mentioned road section key information, screening and matching are performed in the road network road library to obtain candidate road network roads for each road section, including:

[0071] According to the road segment keyword information, screening and matching are performed in the road network road database to obtain candidate road networks for each road segment; if there is no candidate road network road matching the road segment keyword information of a specific road segment, it is deemed that there is no candidate road network road matching the road segment keyword information of the specific road segment; or,

[0072] Based on the road section keyword information, screening and matching are performed in the road network road library to obtain candidate road network roads for each road section; in the case that there are no candidate road network roads that match the road section keyword information of a specific road section, the road section keyword information of the above-mentioned specific road section is split into more fine-grained road section keyword information; based on the above-mentioned road section keyword information, screening and matching are performed in the road network road library to obtain candidate road network roads for the above-mentioned specific road section; in the case that there are no candidate road network roads that match the road section keyword information of the above-mentioned specific road section, it is deemed that there are no candidate road network roads that match the road section keyword information of the above-mentioned specific road section.

[0073] According to an embodiment of the present disclosure, the above-mentioned determining, based on the planned route information of the traveling object, candidate road network roads associated with the above-mentioned planned route information from a road network road library also includes: in the absence of a candidate road network road that matches the road segment key information of a specific road segment, based on the candidate road network roads of an adjacent road segment adjacent to the above-mentioned specific road segment, determining in the above-mentioned road network road library a specific road network road that intersects with the candidate road network roads of the above-mentioned adjacent road segment as a candidate road network road for the above-mentioned specific road segment.

[0074] Combine the following Figure 3A to Figure 3D Each embodiment is described in detail.

[0075] Figure 3A The detailed implementation flow chart of operation S201 according to an embodiment of the present disclosure is schematically shown.

[0076] According to an embodiment of the present disclosure, the above-mentioned planned route information includes: a sequence of road sections passed in sequence, including but not limited to: a road section sequence converted into text form by a user based on text input or voice input, or a road section sequence in text form issued by an electronic device, etc.

[0077] For example, in one scenario, the planned route information of the driving object input by the user is the following sequence Ra: {"Tongsheng Avenue", "Jiangshan Road", "Dongfang Avenue", "Tonghu Avenue Ground Road", "Shenhai Expressway (Xiaohai Exit)", "S15 Yangtong Expressway (Rudong Exit)", "Jugang Town East Second Ring Road", "Su 223 Line 19 km", "Ninghai Line (National Highway 328)"}. The content in the above brackets is also the remarks input by the user. The total number of road sections n corresponding here is 9.

[0078] In one embodiment, in the above operation S201, according to the planned route information of the traveling object, determining the candidate road network associated with the planned route information from the road network road library includes the following operations: S311 and S312.

[0079] In operation S311, text extraction is performed on each road section in the above-mentioned planned route information to obtain road section keyword information for each road section.

[0080] First, remove non-keyword parts using dictionaries, regular expressions, and other methods. For example, remove non-keyword information such as province, city, district, and county information, "expressway," "avenue," "highway," "highway," "street," "section," "intersection," and "entrance." Then, segment the text after removing the keyword information, and use the segmentation results as the road segment r. i Keywords k i Represents road segment r i The total number of keywords.

[0081] Text extraction is performed on each section of the above sequence Ra, and the keywords corresponding to the first section, the second section, ... the ninth section are: [{'Tongsheng'}, {'Jiangshan'}, {'Dongfang'}, {'Tonghu', 'Ground'}, {'Shenhai', 'Xiaohai'}, {'Yangtong', '15'}, {'East Second Ring Road', 'Jugang'}, {'223', '19'}, {'328', 'Ninghai'}].

[0082] In operation S312, based on the above-mentioned road segment keyword information, screening and matching are performed in the road network library to obtain candidate road network roads for each road segment.

[0083] For example, by screening and matching based on the above-mentioned road section keyword information in the road network road database, the result of the candidate road network roads corresponding to each road section in sequence is Rta: {["Tongsheng Avenue", "Tongsheng Avenue Elevated", "Tongsheng South Road"], ["Jiangshan Road"], ["Dongfang Road", "Dongfang Avenue Elevated", "Dongfanghong Bridge", "Dongfang Avenue", "Xingyi Village Dongfanghe Road", "Dongfanghe Road", "Dongfangsi Road"], ["Tonghu Avenue", "Tonghu Avenue Elevated", "Tonghu Road"], ["Shenhai Expressway", "Xiaohai Hub", "Xiaohai Road"], ["Yangtong Road", "Yangtong Expressway", "Y157", "Group 15 Center Road", "C215", "Y159", "Y152", "C159", "Shidi Village 15, Lushigang Town "Group", "Y155", "Y615", "Group 15 South Road", "G15", "S15", "C151", "Y315", "X215", "C153", "X315", "Y153", "Y151"], ["East Second Ring Road"], ["S223", "S223 (Old)", "Group 19 Nanzhuang Line", "S19", "Y019", "Y194", "Y195"], ["G328", "G328 (Old)", "C328", "Ninghai Road", "Ninghai East Road", "Ninghai South Road", "Ninghai West Road"]}; In the result, the information in the square brackets is one or more candidate road network roads corresponding to the current section, and the curly brackets represent the candidate road network roads of all sections of the planned route information as a whole.

[0084] In most scenarios, there will be candidate road networks that match the above-mentioned road segment keyword information. In some individual scenarios, there will be the following situation: when screening and matching the road network road database according to the above-mentioned road segment keyword information, no matching candidate road network roads are obtained in one or several road segments, that is, there is no candidate road network road that matches the road segment keyword information of a specific road segment. In this case, you can use Figure 3B or Figure 3C The example scenario.

[0085] Figure 3B The detailed implementation flowchart of operation S201 according to another embodiment of the present disclosure is schematically shown.

[0086] Reference Figure 3B As shown, according to an embodiment of the present disclosure, in the above operation S201, based on the planned route information of the traveling object, a candidate road network road associated with the above planned route information is determined from a road network road library, including the following operations: S321, S322 and S323.

[0087] In operation S321, text extraction is performed on each road section in the above-mentioned planned route information to obtain road section keyword information for each road section.

[0088] In operation S322, based on the above-mentioned road segment keyword information, screening and matching are performed in the road network road library to obtain candidate road network roads for each road segment.

[0089] When there are candidate road networks for each road segment, the corresponding situation can refer to the above combination Figure 3A As shown in the described embodiment, the detailed description of operation S321 can refer to the description of operation S311 in the aforementioned embodiment, and the detailed description of operation S322 can refer to the description of operation S312 in the aforementioned embodiment, which will not be repeated here.

[0090] In the case that there is no candidate road network road matching the road segment keyword information of the specific road segment, operation S323 is performed.

[0091] In operation S323, based on the candidate road network roads of the adjacent road segment adjacent to the specific road segment, a specific road network road intersecting with the candidate road network roads of the adjacent road segment is determined in the road network road library as a candidate road network road for the specific road segment.

[0092] In this embodiment, based on operations S321 to S323, when the candidate road network roads of one or more specific road sections cannot be obtained by matching keyword information, the candidate road network roads of the specific road section are determined by adopting the route relationship between the candidate road network roads of adjacent road sections and the specific road section. This can solve the problem of path identification and alignment caused by the missing candidate road network data when there are errors or deviations in the planned route information. For example, it can solve the following scenario: the problem of missing candidate road network data in the overall path due to user input errors or non-standard words input by the user.

[0093] Figure 3C The detailed implementation flowchart of operation S201 according to another embodiment of the present disclosure is schematically shown.

[0094] Reference Figure 3C As shown, according to another embodiment of the present disclosure, in the above operation S201, based on the planned route information of the traveling object, a candidate road network road associated with the above planned route information is determined from the road network road library, including the following operations: S331, S332, S333 and S334.

[0095] In operation S331, text extraction is performed on each road section in the above-mentioned planned route information to obtain road section keyword information for each road section.

[0096] In operation S332, based on the above-mentioned road segment keyword information, screening and matching are performed in the road network road library to obtain candidate road network roads for each road segment.

[0097] When there are candidate road networks for each road segment, the corresponding situation can refer to the above combination Figure 3A As shown in the described embodiment, the detailed description of operation S331 can refer to the description of operation S311 in the aforementioned embodiment, and the detailed description of operation S332 can refer to the description of operation S312 in the aforementioned embodiment, which will not be repeated here.

[0098] In the case that there is no candidate road network road that matches the section keyword information of the specific section, operations S333 and S334 are performed.

[0099] In operation S333, the road segment keyword information of the specific road segment is split into more fine-grained road segment keyword information.

[0100] For example, the keyword information of the road section is: 'East Second Ring Road', 'Jue Port', and the keyword information of the road section obtained by splitting is: 'East', 'Second', 'Ring', 'Jue', 'Port'.

[0101] In operation S334, based on the above-mentioned road segment keyword information, screening and matching are performed in the road network road library to obtain candidate road network roads for the above-mentioned specific road segment.

[0102] In this embodiment, based on operations S331 to S334, when candidate road network roads for one or more specific road sections cannot be obtained by matching keyword information, screening and matching are performed based on finer-grained road section keyword information to obtain candidate road network roads for the specific road section. This can, to a certain extent, solve the problem of path identification and alignment caused by missing candidate road network data when there are errors or deviations in the planned route information. For example, it can solve the following scenario: the problem of missing candidate road network data in the overall path due to user input errors or non-standard words input by the user.

[0103] Figure 3D The detailed implementation flow chart of operation S201 according to another embodiment of the present disclosure is schematically shown.

[0104] In the above Figure 3C On the basis of the exemplary embodiment, the implementation scenario corresponding to this embodiment is: when screening and matching is performed based on the road segment keyword information with a finer granularity, there is no candidate road network road that matches the road segment keyword information of a specific road segment.

[0105] Reference Figure 3DAs shown, in the above operation S201, according to the planned route information of the traveling object, a candidate road network road associated with the above planned route information is determined from the road network road library, including the following operations: S341, S342, S343, S344 and S345.

[0106] In operation S341, text extraction is performed on each road section in the above-mentioned planned route information to obtain road section keyword information for each road section.

[0107] In operation S342, based on the above-mentioned road segment keyword information, screening and matching are performed in the road network road library to obtain candidate road network roads for each road segment.

[0108] When there are candidate road networks for each road segment, the corresponding situation can refer to the above combination Figure 3A As shown in the described embodiment, the detailed description of operation S341 can refer to the description of operation S311 in the aforementioned embodiment, and the detailed description of operation S342 can refer to the description of operation S312 in the aforementioned embodiment, which will not be repeated here.

[0109] In the case that there is no candidate road network road that matches the section keyword information of the specific section, operations S343 and S344 are performed.

[0110] In operation S343, the road segment keyword information of the specific road segment is split into more fine-grained road segment keyword information.

[0111] In operation S344, based on the above-mentioned road segment keyword information, screening and matching are performed in the road network road library to obtain candidate road network roads for the above-mentioned specific road segment.

[0112] In the case that there is no candidate road network road matching the road segment keyword information of the specific road segment, operation S345 is performed.

[0113] In operation S345, based on the candidate road network roads of the adjacent road segment adjacent to the specific road segment, a specific road network road intersecting with the candidate road network road of the adjacent road segment is determined in the road network road database as a candidate road network road for the specific road segment.

[0114] Based on the above operations S341 to S345, screening and matching are first performed based on the segment keyword information. When a specific segment cannot obtain a candidate road network road based on the segment keyword information, screening and matching are further performed based on the segment keyword information with a finer granularity. For a specific segment that still cannot obtain a candidate road network road match based on the segment keyword information, the candidate road network road for the specific segment is determined by adopting the route relationship between the candidate road network roads of adjacent segments and the specific segment. Based on layer-by-layer screening and supplementary data based on adjacent segments, the problem of path identification and alignment caused by missing candidate road network data when there are errors or deviations in the planned route information can be solved. For example, the following scenario can be solved: the problem of missing candidate road network data in the overall path due to user input errors or non-standard words input by the user.

[0115] Figure 4 The detailed implementation flow chart of operation S202 according to the embodiment of the present disclosure is schematically shown.

[0116] According to the embodiments of the present disclosure, referring to Figure 4 As shown, in the above operation S202, the probabilities of each route between adjacent sections are calculated according to the distances between each candidate road network road of the adjacent sections, and a probability matrix containing the probabilities of the routes between all adjacent sections is obtained, including the following operations: S401, S402, S403, S404 and S405.

[0117] In operation S401, for any one of all road segments, the distances between candidate road network roads of the current road segment and candidate road network roads of the next road segment are calculated to obtain a distance vector consisting of distance elements corresponding to the current road segment.

[0118] In one embodiment, the candidate road network roads C of the previous road segment are respectively i , the candidate road network road C of the rear section i+1 Combine them in order and calculate the shortest straight-line distance d between the coordinate points of the two roads according to the road network data. i ={dis(c i,j ,c i+1,j′ ),j=0,1,2,…,m i -1,j′=0,1,2,…,m i+1 -1},dis(c i,j ,c i+1,j′ ) represents the candidate road network road C of the previous section i The jth candidate road network road in the next section and the candidate road network road C i+1 The shortest straight-line distance between the coordinate points of the j′th candidate road network.

[0119] In operation S402, a distance matrix is ​​obtained according to the distance vectors of all road segments.

[0120] According to an embodiment of the present disclosure, in the above operation S402, a distance matrix is ​​obtained according to the distance vectors of all road sections, including the following sub-operations: S402a and S402b.

[0121] In sub-operation S402a, the distance vectors of all the road segments are used as rows of the distance matrix.

[0122] In sub-operation S402b, the maximum number of columns of the distance vectors in all road segments is used as the column scale of the distance matrix, and a default value is added to a specific distance vector whose number of columns is smaller than the column scale to make up the number of columns.

[0123] Among them, in the subsequent operation S403, when the distance vector of each row is numerically transformed, the above default value does not participate in the calculation, and the maximum value of the distance element in the current distance vector is subtracted from each distance element to obtain the transformed distance element; in the subsequent operation S404, when the above target distance vector is input into the soft maximization function for processing, the output probability corresponding to the above default value is the preset value.

[0124] In this embodiment, l = max({m i *m i+1 ,i=0,1,2,…,n-2})(wherein, since the distance of the route is taken by combining two adjacent sections, the maximum sequence number of the current section corresponding to i is n-2, and the second to last section is the maximum value of the sequence number of the current section),d i When there are less than l elements, the default value default_d is used to fill the gap. i As the i-th row of the distance matrix D, the scale of D is (n-1,l), which means that the distance matrix D has a total of n-1 rows and l columns, the serial numbers of the n rows are 0, 1, 2, ...n-2, and the serial numbers of the l columns are 0, 1, 2, ...l-1.

[0125] For example, in actual operation, the default value may be set to infinity, and accordingly, the preset value of the corresponding output probability is directly set to the minimum value 1e-08 for the convenience of calculation.

[0126] In a specific embodiment, for the result Rta of the above-mentioned candidate road network roads, the number of distance elements contained in the distance vector in the 0th row can be obtained as: 3 ("Tongsheng Avenue", "Tongsheng Avenue Elevated Road", "Tongsheng South Road", a total of 3) * 1 ("Jiangshan Road", a total of one) = 3, similarly, the number of distance elements contained in the distance vector in the 1st row is obtained as: 1*7=7, the number of distance elements contained in the distance vector in the 2nd row is obtained as: 7*3=21, the number of distance elements contained in the distance vector in the 3rd row is obtained as: 3*3=9, the number of distance elements contained in the distance vector in the 4th row is obtained as: 3*21=63, the number of distance elements contained in the distance vector in the 5th row is obtained as: 21*1=21, the number of distance elements contained in the distance vector in the 6th row is obtained as: 1*7=7, and the number of distance elements contained in the distance vector in the 7th row is obtained as: 7*7=49.

[0127] According to all the distance vectors, the candidate road network result Rta can be obtained. The distance matrix corresponding to the result Rta contains 8 rows and 63 columns of distance elements.

[0128] In operation S403, a numerical transformation is performed on the distance vector of each row in the distance matrix to obtain a transformed target distance vector, and the sum of the corresponding distance elements before and after the transformation is a fixed value.

[0129] For example, in one embodiment, for each row in the distance matrix D, each calculated distance element in each row of the distance vector is replaced by the maximum distance element value of the row (excluding the default value) minus the value of the element, to obtain a distance matrix D' containing the transformed target distance vector. In other embodiments, other numerical transformation methods can also be used, as long as the transformation method can transform the larger value in the original distance matrix into a smaller value, in order to ensure that the probability of the corresponding calculation of the relatively close distance between the candidate road network roads of adjacent sections is relatively large.

[0130] In operation S404, each target distance vector is input into a soft maximization function for processing, and the probability distribution value corresponding to each distance element in the target distance vector is output, wherein the probability distribution value corresponding to the current distance element is used to characterize the probability of the route corresponding to the current distance element.

[0131] In operation S405, a probability matrix including the probabilities of routes between all adjacent road segments is obtained according to the probability distribution values ​​corresponding to all target distance vectors.

[0132] For the distance matrix D′, the rows of the distance matrix D′ are input into the softmax function for calculation, and the output of each row is summarized to obtain a probability matrix P. That is, P i,: =softmax(D′ i,:), i = 0, 1, ..., n-2; a superscript ":" in D' represents the column number of each element in the same row, P i,: It represents the probability score distribution of the combinations of candidate road networks between the given description segment i and segment i+1. The size of the probability matrix P is (n-1, l), which means that the probability matrix P has n-1 rows and l columns, the n rows are numbered 0, 1, 2, ..., n-2, and the l columns are numbered 0, 1, 2, ..., l-1.

[0133] Assume that s is used to represent any column number / label in the probability matrix, and the value of s is 0, 1, 2, ..., l-1, then P i,s Represents the candidate road network road C i Candidate roads c x To candidate road network road C i+1 Candidate roads c y The probability of scoring; where x = |C i+1 ||s(The value of x is equal to the column number s divided by C i+1 Get the quotient), y = s% | C i+1 |(The value of y is equal to the column number s divided by C i+1 Get the remainder).

[0134] For example, the road sections are numbered starting from 0. Assume that the 0th road section (candidate road network road C in vector form) 0 )Total 3(m i An example of) candidate road network roads, for example, represented as (the superscript indicates the segment number, the subscript indicates the candidate network road number, the following expressions are the same), the first segment (candidate network road C in vector form 1 )Total 2(m i+1 An example of) candidate road network roads, for example, represented as Then, after the target distance vector corresponding to the 0th section to the 1st section is input into the softmax function, a total of 6 probability distributions are output: P 0,0 , P 0,1 , P 0,2 , P 0,3 , P 0,4 , P 0,5 Among them, |C i+1 |The corresponding value is 2, then when the sequence number s=0, x=0, y=0; when the sequence number s=1, x=0, y=1; when the sequence number s=2, x=1, y=0; when the sequence number s=3, x=1, y=1; when the sequence number s=4, x=2, y=0; when the sequence number s=5, x=2, y=1; that is, P 0,0 , P 0,1 , P 0,2, P 0,3 , P 0,4 , P 0,5 Corresponding to the route The respective probabilities, p 0,0 +P 0,1 +P 0,2 +P 0,3 +P 0,4 +P 0,5 =1.

[0135] In other embodiments, if each road section or candidate road network road is not numbered from 0, they are processed to obtain numbers starting from 0, and the correspondence between the serial numbers of each column in the probability matrix and the routes can be obtained similarly.

[0136] In a specific embodiment, the planned route information for the above example is the following sequence Ra: {"Tongsheng Avenue", "Jiangshan Road", "Dongfang Avenue", "Tonghu Avenue Ground Road", "Shenhai Expressway (Xiaohai Exit)", "S15 Yangtong Expressway (Rudong Exit)", "Jugang Town East Second Ring Road", "Su 223 Line 19 km", "Ninghai Line (National Highway 328)"}, and the element P in the probability matrix 7,48 It is 0.02329721202244289, which corresponds to the probability elements in the 7th row (the probability corresponding to the route from the 7th to the 8th road segment, and the road segment number starts from 0) and the 48th column (the maximum value of the product of the total number of candidate road network roads in the previous road segment and the total number of candidate road network roads in the next road segment in the adjacent road segment is 63).

[0137] The probability element P 7,48 It indicates the 48th candidate road network road among the 63 candidate road network roads after the completion between the road section "Su 223 Line 19 km" and the road section "Ninghai Line (328 National Highway)" in the planned route information. According to the example, |C 8 |=7,7|48=6(quotient is 6), 48%7=6(remainder is 6), P 7,48 This means that the probability of the route from "Y195" (the candidate road network road numbered 6 in the 7th section, with the serial numbers starting from 0) to "Ninghai West Road" (the candidate road network road numbered 6 in the 8th section, with the serial numbers starting from 0) is 0.02329721202244289.

[0138] Figure 5 The detailed implementation flow chart of operation S203 according to the embodiment of the present disclosure is schematically shown.

[0139] According to the embodiments of the present disclosure, referring to Figure 5As shown, in the above operation S203, beam search processing is performed on the above probability matrix to obtain target route information, including the following operations: S501, S502 and S503.

[0140] In operation S501 , for the first row of the probability matrix, the first K routes corresponding to the highest probability scores of the routes between each group of adjacent road segments in the current row are determined as candidate routes, where K is a set search beam width.

[0141] For example, K=2, the first two routes with the highest scores corresponding to the first row (row 0) are the candidate routes obtained in step 1: "Tongsheng South Road"→"Jiangshan Road"; "Tongsheng Avenue"→"Jiangshan Road".

[0142] In operation S502, for each row in the subsequent rows after the first row of the above probability matrix, the conditional probability score of the route corresponding to the current row based on the K candidate routes corresponding to the previous row is determined, and the route corresponding to the first K conditional probability scores is determined as the target candidate route.

[0143] For the subsequent rows of the probability matrix, such as the first row after the first row, by calculating the conditional probability scores of the route in the current row under the two candidate routes corresponding to the 0th row, the first two routes with the highest scores are determined as the candidate routes obtained in the second step. For example, the conditional probability scores corresponding to each route in the first row of the probability matrix under the premise of "Tongsheng South Road" → "Jiangshan Road" are calculated; the conditional probability scores corresponding to each route in the first row of the probability matrix under the premise of "Tongsheng Avenue" → "Jiangshan Road" are calculated, and the first two routes with the highest scores are determined as the target candidate routes in the second step. For example, the results obtained are: "Tongsheng South Road" → "Jiangshan Road" → "Dongfang Avenue Elevated", "Tongsheng Avenue" → "Jiangshan Road" → "Dongfang Avenue". And so on, the two target candidate routes output in step 8 are finally obtained.

[0144] In operation S503 , a route with the highest score among the target candidate routes is determined as target route information.

[0145] For example, the route with the largest score among the two target candidate routes output in step 8 is determined as the target route information.

[0146] Based on the above operations S501 to S503, efficient calculation can be achieved within a limited calculation space, and the target route information can be quickly located, and the target route information is likely to conform to the planned route information input by the user.

[0147] In one embodiment, based on a set search beam width K (eg, K=2-3), a beam search is performed row by row on the probability matrix P.

[0148] For time step t, calculate the score Among them SC 0,: =1.0. t,: Sort by large to small, and take the scores SC corresponding to the first K records t,s Enter the calculation and sorting of the next time step, and record the value sequence of the corresponding sequence number / index s.

[0149] Here, log(P t+i-1,s ) function uses base 10 for calculation, so the negative sign before log makes the value positive.

[0150] For example, in the search of step 1 (time step t = 1), first the probability vector P of the 0th row in the probability matrix P is 0,s A search is performed and the first two (in the example of K=2) routes with the highest scores are calculated as candidate routes.

[0151] Finally, take max(s n-1,: ) as the bundle search result, according to the corresponding index s value sequence S and the previous road segment candidate road network road C i , the candidate road network road C of the rear section i+1 Decode from front to back. i =S i-1 %|C i |, i = 1, 2, ..., n-1, special x 0 =S 0 / |C 1 |,x i The final choice is C i The candidate road index is obtained to obtain the final converted road network road sequence.

[0152] For example, for the above embodiment in which the planned route information is a sequence Ra, the obtained value sequence S of the sequence number / index s is [2, 1, 3, 0, 1, 1, 0, 0].

[0153] x 0 =J 0 / |C 1 |=2 / 1=2,x 1 =J 0 %|C 1 |=2%1=0, then "Tongsheng South Road" and "Jiangshan Road" are selected as the first and second road section candidate roads for the final selection. 2 =J 1 %|C 2|=1%7=1, corresponding to selecting "Dongfang Avenue Elevated Road" as the third section candidate road to be finally selected. The final result obtained by decoding in the above manner is ["Tongsheng South Road", "Jiangshan Road", "Dongfang Avenue Elevated Road", "Tonghu Avenue", "Shenhai Expressway", "Yangtong Expressway", "East Second Ring Road", "S223", "G328"].

[0154] A second exemplary embodiment of the present disclosure provides a path monitoring method.

[0155] Figure 6 The flowchart of the path monitoring method according to the embodiment of the present disclosure is schematically shown.

[0156] Reference Figure 6 As shown, the path monitoring method provided by the embodiment of the present disclosure includes the following operations: S601, S602, S603 and S604. The operations S601 to S604 can be performed by a terminal device or a server.

[0157] In operation S601, a method of aligning route data is used to determine target route information of a traveling object.

[0158] The operation S601 can be implemented by using the method including operations S201 to S203 described in the first embodiment.

[0159] In operation S602, actual route information of the traveling object is obtained.

[0160] The above-mentioned actual route information can be obtained by the terminal device or server through the actual driving path monitoring function of the path supervision application itself, or obtained through the path supervision application to the map navigation system or other driving path monitoring system that has opened data access rights.

[0161] In operation S603, the actual route information is matched with the target route information.

[0162] For example, the actual route information and the target route information may be matched by comparing them section by section.

[0163] In operation S604, when the actual route information does not match the target route information, route deviation warning information is generated.

[0164] The generated warning information can be fed back to the administrator, allowing the administrator (such as relevant personnel from the route supervision department) to decide whether to adopt corresponding supervision measures; the generated warning information can also be fed back to relevant personnel of the transport vehicles using the route supervision application, so that the relevant personnel of the transport vehicles can check whether their own driving routes deviate from the planned route information based on the warning information.

[0165] Based on the above operations S601 to S604, it is possible to align the planned route information input in text with the existing road network information, and to a certain extent overcome the problem of being unable to match the actual road due to incorrect or irregular filling / input; in addition, by obtaining the actual route information and comparing and matching the actual route information with the target route information, it is possible to supervise and give early warning prompts for the user's actual route based on the above target route information, which can be applied to the scenario of transport route supervision of hazardous chemicals vehicles in urban management.

[0166] A third exemplary embodiment of the present disclosure provides an apparatus for aligning route data.

[0167] Figure 7 The structure block diagram of the device for aligning route data according to an embodiment of the present disclosure is schematically shown.

[0168] Reference Figure 7 As shown, the device 700 for aligning route data provided by the embodiment of the present disclosure includes: a road network matching module 701, a probability calculation module 702 and a processing module 703.

[0169] The road network matching module 701 is used to determine the candidate road network roads associated with the planned route information from the road network road library according to the planned route information of the driving object. The road network matching module 701 includes functional modules or functional submodules corresponding to the operations S311-S312, S321-S323, S331-S334, S341-S345.

[0170] The probability calculation module 702 is used to calculate the probability of each route between adjacent sections according to the distance between each candidate road network road of the adjacent sections, and obtain a probability matrix containing the probabilities of the routes between all adjacent sections. The probability calculation module 702 includes the functional modules or functional submodules corresponding to the operations S401 to S405.

[0171] The processing module 703 is used to perform beam search processing on the probability matrix to obtain target route information in the road network. The processing module 703 includes functional modules or functional submodules corresponding to the operations S501 to S503.

[0172] A fourth exemplary embodiment of the present disclosure provides a path monitoring device.

[0173] Figure 8 The structure block diagram of the path monitoring device according to the embodiment of the present disclosure is schematically shown.

[0174] Reference Figure 8As shown, the path monitoring device 800 provided in the embodiment of the present disclosure includes: a road network matching module 801, a probability calculation module 802, a processing module 803, an actual route acquisition module 804, a route information matching module 805 and a warning information generation module 806.

[0175] The road network matching module 801 is used to determine, based on the planned route information of the driving object, a candidate road network road associated with the planned route information from a road network road library.

[0176] The probability calculation module 802 is used to calculate the probability of each route between adjacent road sections according to the distances between each candidate road network road of the adjacent road sections, and obtain a probability matrix containing the probabilities of the routes between all adjacent road sections.

[0177] The processing module 803 is used to perform beam search processing on the probability matrix to obtain target route information in the road network.

[0178] The actual route acquisition module 804 is used to acquire the actual route information of the driving object.

[0179] The route information matching module 805 is used to match the actual route information with the target route information.

[0180] The warning information generating module 806 is used to generate warning information of route deviation when the actual route information does not match the target route information.

[0181] Any multiple of the modules in the above-mentioned device 700 for aligning route data or the modules in the above-mentioned device 800 for path monitoring can be combined in one module for implementation, or any one of the modules can be split into multiple modules. Alternatively, at least part of the functions of one or more of these modules can be combined with at least part of the functions of other modules and implemented in one module. At least one of the modules in the device 700 for aligning route data or the above-mentioned device 800 for path monitoring can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application-specific integrated circuit (ASIC), or can be implemented by hardware or firmware such as any other reasonable way of integrating or encapsulating the circuit, or implemented in any one of the three implementation methods of software, hardware and firmware or in any appropriate combination of any of them. Alternatively, at least one of the modules in the device 700 for aligning route data or the above-mentioned device 800 for path monitoring can be at least partially implemented as a computer program module, and when the computer program module is run, the corresponding function can be executed.

[0182] A fifth exemplary embodiment of the present disclosure provides an electronic device.

[0183] Fig. 9 The structural block diagram of the electronic device provided by the embodiment of the present disclosure is schematically shown.

[0184] Reference Fig. 9 As shown, the electronic device 900 provided by the embodiment of the present disclosure includes a processor 901, a communication interface 902, a memory 903 and a communication bus 904, wherein the processor 901, the communication interface 902 and the memory 903 communicate with each other through the communication bus 904; the memory 903 is used to store computer programs; the processor 901 is used to implement the method for aligning route data or the method for path monitoring as described above when executing the program stored in the memory.

[0185] The sixth exemplary embodiment of the present disclosure further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for aligning route data or the method for path monitoring as described above is implemented.

[0186] The computer-readable storage medium may be included in the device / apparatus described in the above embodiment; or it may exist independently without being assembled into the device / apparatus. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed, the method according to the embodiment of the present disclosure is implemented.

[0187] According to an embodiment of the present disclosure, a computer-readable storage medium may be a non-volatile computer-readable storage medium, for example, may include but is not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, apparatus, or device.

[0188] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0189] The foregoing is merely a specific embodiment of the present disclosure, which enables those skilled in the art to understand or implement the present disclosure. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present disclosure. Therefore, the present disclosure will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features applied herein.

Claims

1. A method for aligning route data, It is characterized in that include: According to the planned route information of the driving object, determining a candidate road network road associated with the planned route information from a road network road library; According to the distances between the candidate road network roads of the adjacent sections, the probabilities of the routes between the adjacent sections are calculated, and a probability matrix containing the probabilities of the routes between all the adjacent sections is obtained; wherein, the probabilities of the routes corresponding to the different combinations are calculated by combining the candidate road network roads of the two adjacent sections in each pair, and a probability matrix is ​​generated; the probability of the route corresponding to the candidate road network road that is far away from the two adjacent sections is relatively small, and the probability of the route corresponding to the candidate road network road that is close to the two adjacent sections is relatively large; A beam search process is performed on the probability matrix to obtain target route information in the road network.

2. The method according to claim 1, It is characterized in that According to the distances between each candidate road network road of adjacent sections, the probability of each route between adjacent sections is calculated to obtain a probability matrix containing the probabilities of routes between all adjacent sections, including: For any one of all the road sections, calculate the distances between candidate road network roads of the current road section and candidate road network roads of the next road section, and obtain a distance vector composed of distance elements corresponding to the current road section; According to the distance vectors of all road sections, a distance matrix is ​​obtained; Performing numerical transformation on the distance vector of each row in the distance matrix to obtain a transformed target distance vector, wherein the sum of the corresponding distance elements before and after the transformation is a fixed value; Input each target distance vector into the soft maximization function for processing, and output the probability distribution value corresponding to each distance element in the target distance vector, wherein the probability distribution value corresponding to the current distance element is used to characterize the probability of the route corresponding to the current distance element; According to the probability distribution values ​​corresponding to all target distance vectors, a probability matrix containing the probabilities of routes between all adjacent road sections is obtained.

3. The method according to claim 2, It is characterized in that The distance matrix is ​​obtained according to the distance vectors of all road sections, including: The distance vectors of all road segments are used as rows of the distance matrix; The maximum number of columns of the distance vectors in all road segments is used as the column scale of the distance matrix, and a default value is added to a specific distance vector whose number of columns is smaller than the column scale; Among them, when the distance vector of each row is numerically transformed, the default value does not participate in the calculation, and the maximum value of the distance element in the current distance vector is subtracted from each distance element to obtain the transformed distance element; when the target distance vector is input into the soft maximization function for processing, the output probability corresponding to the default value is the preset value.

4. The method according to claim 1, It is characterized in that The performing a beam search process on the probability matrix to obtain target route information includes: For the first row of the probability matrix, the first K routes corresponding to the highest probability scores of the routes between each group of adjacent road segments in the current row are determined as candidate routes, where K is the set search beam width; For each row in the subsequent rows after the first row of the probability matrix, determine the conditional probability score of the route corresponding to the current row under the premise of the K candidate routes corresponding to the previous row, and determine the routes corresponding to the first K with the highest conditional probability scores as the target candidate routes; The route with the largest score among the target candidate routes is determined as the target route information.

5. The method according to claim 1, It is characterized in that The planned route information includes: a sequence of road sections to be passed in sequence; Wherein, determining, from a road network road library, a candidate road network associated with the planned route information according to the planned route information of the traveling object comprises: Performing text extraction on each road section in the planned route information to obtain key road section information for each road section; According to the key information of the road section, screening and matching are performed in the road network database to obtain candidate road network roads for each road section.

6. The method according to claim 5, It is characterized in that The step of determining, based on the planned route information of the traveling object, a candidate road network associated with the planned route information from a road network road library further includes: In the absence of a candidate road network road that matches the road segment key information of a specific road segment, based on the candidate road network roads of an adjacent road segment adjacent to the specific road segment, a specific road network road that intersects with the candidate road network roads of the adjacent road segment is determined in the road network road library as a candidate road network road for the specific road segment.

7. The method according to claim 5 or 6, It is characterized in that According to the key information of the road section, screening and matching are performed in the road network database to obtain candidate road networks for each road section, including: According to the road segment keyword information, screening and matching are performed in the road network road database to obtain candidate road networks for each road segment; if there is no candidate road network road matching the road segment keyword information of a specific road segment, it is deemed that there is no candidate road network road matching the road segment keyword information of the specific road segment; or, According to the road section keyword information, screening and matching are performed in the road network road library to obtain candidate road network roads for each road section; in the case that there are no candidate road network roads that match the road section keyword information of a specific road section, the road section keyword information of the specific road section is split into more fine-grained road section keyword information; according to the road section keyword information, screening and matching are performed in the road network road library to obtain candidate road network roads for the specific road section; in the case that there are no candidate road network roads that match the road section keyword information of the specific road section, it is deemed that there are no candidate road network roads that match the road section keyword information of the specific road section.

8. A method for path monitoring, It is characterized in that include: Determine the target route information of the traveling object by using the method for aligning route data as described in any one of claims 1 to 7; Acquiring actual route information of the traveling object; matching the actual route information with the target route information; In the case where the actual route information does not match the target route information, a warning message of route deviation is generated.

9. A device for aligning route data, It is characterized in that include: A road network matching module, used to determine, from a road network road library, a candidate road network road associated with the planned route information of the traveling object according to the planned route information; The probability calculation module is used to calculate the probability of each route between adjacent sections according to the distance between each candidate road network road of the adjacent sections, and obtain a probability matrix containing the probability of routes between all adjacent sections; wherein, the probabilities of routes corresponding to different combinations are calculated by combining the distances obtained by each candidate road network road of the two adjacent sections in the adjacent sections, and a probability matrix is ​​generated; the probability of the route corresponding to the candidate road network road that is far away from the two adjacent sections is relatively small, and the probability of the route corresponding to the candidate road network road that is close to the two adjacent sections is relatively large; The processing module is used to perform beam search processing on the probability matrix to obtain target route information in the road network.

10. A path monitoring device, It is characterized in that include: A road network matching module, used to determine, from a road network road library, a candidate road network road associated with the planned route information of the traveling object according to the planned route information; The probability calculation module is used to calculate the probability of each route between adjacent sections according to the distance between each candidate road network road of the adjacent sections, and obtain a probability matrix containing the probability of routes between all adjacent sections; wherein, the probabilities of routes corresponding to different combinations are calculated by combining the distances obtained by each candidate road network road of the two adjacent sections in the adjacent sections, and a probability matrix is ​​generated; the probability of the route corresponding to the candidate road network road that is far away from the two adjacent sections is relatively small, and the probability of the route corresponding to the candidate road network road that is close to the two adjacent sections is relatively large; A processing module, used for performing a beam search process on the probability matrix to obtain target route information in the road network; An actual route acquisition module, used to acquire actual route information of the traveling object; A route information matching module, used for matching the actual route information with the target route information; The warning information generating module is used to generate warning information of route deviation when the actual route information does not match the target route information.

11. An electronic device, It is characterized in that It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; Memory, used to store computer programs; A processor, for implementing the method according to any one of claims 1 to 8 when executing a program stored in a memory.

12. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images