Positioning map construction method and device for unmanned bus and electronic equipment

Abstract

The application discloses a positioning map construction method and device for an unmanned bus and electronic equipment. The method comprises the following steps: acquiring road image data in a preset area range corresponding to each station position of the unmanned bus based on the station position; constructing a local map subgraph corresponding to each station position according to the road image data in the preset area range corresponding to each station position, wherein the local map subgraph comprises a local point cloud map and a local semantic map; and establishing a geographic fence corresponding to each station position according to the local map subgraph corresponding to each station position to complete the construction of the positioning map for the unmanned bus. The positioning map construction method for the unmanned bus in the embodiment of the application uses visual technology to construct a corresponding local map subgraph for each station position, which comprises a local point cloud map and a local semantic map. Compared with the method of constructing a global map, the sparsity of the map and the positioning efficiency and positioning accuracy are ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic driving, in particular to a positioning map construction method and device for a Robobus and an electronic device. BACKGROUND

[0002] With the increasing maturity of automatic driving vehicle technology and the increasing perfection of automatic driving implementation policy, the Robobus is increasingly widely applied in urban traffic. Since the operation route of the Robobus is relatively fixed, the Robobus has a lower demand for scene adaptability in positioning compared to other types of automatic driving vehicles. For example, in the fixed operation route in the urban area, complex positioning scenes such as tunnels and bridges are not involved.

[0003] Based on the above situation, the positioning technology of the Robobus is mainly realized based on GNSS (Global Navigation Satellite System) / RTK (Real Time Kinematic) + IMU (Inertial Measurement Unit) + camera semantic recognition (including lane lines, road arrows, etc.) + high-precision map and other technologies.

[0004] However, due to the high-rise buildings around the station position, the quality of the GNSS positioning signal is poor, which causes deviation in positioning. Moreover, the visual positioning result obtained by simply matching the existing semantic recognition algorithm cannot give an accurate positioning result, which specifically shows that:

[0005] 1) Due to the special situation of entering and leaving the station, for example, in some time periods with heavy traffic such as commuting time, the camera cannot well recognize the ground semantic elements;

[0006] 2) Due to the special situation of entering and leaving the station, for example, in some special sections such as auxiliary roads, the camera cannot well recognize the ground semantic elements or there is no ground semantic element;

[0007] 3) There is no corresponding road semantic element in the high-precision map, which causes the matching to be unable to be performed. SUMMARY

[0008] The embodiments of the present application provide a positioning map construction method and device for a Robobus to provide accurate and reliable positioning basis for the Robobus.

[0009] The embodiments of the present application adopt the following technical solutions:

[0010] In a first aspect, the embodiments of the present application provide a positioning map construction method for a Robobus, wherein the method comprises:

[0011] acquire road image data in a preset area range corresponding to each station position based on the respective station positions of the unmanned bus;

[0012] construct a local map subgraph corresponding to each station position according to the road image data in the preset area range corresponding to each station position;

[0013] establish a geographic fence corresponding to each station position according to the local map subgraph corresponding to each station position, to complete the construction of the positioning map for the unmanned bus.

[0014] Optionally, the constructing a local map subgraph corresponding to each station position according to the road image data in the preset area range corresponding to each station position comprises:

[0015] constructing a local point cloud map corresponding to each station position according to the road image data in the preset area range corresponding to each station position by using a preset point cloud map construction algorithm;

[0016] constructing a local semantic map corresponding to each station position according to the road image data in the preset area range corresponding to each station position by using a preset semantic map construction algorithm.

[0017] Optionally, the constructing a local semantic map corresponding to each station position according to the road image data in the preset area range corresponding to each station position by using a preset semantic map construction algorithm comprises:

[0018] identifying road elements in the preset area range corresponding to each station position according to the road image data in the preset area range corresponding to each station position by using a preset road element recognition model, to obtain a road element recognition result;

[0019] constructing a local semantic map corresponding to each station position according to the road element recognition result.

[0020] Optionally, the road image data is acquired by a collection vehicle, and the constructing a local map subgraph corresponding to each station position according to the road image data in the preset area range corresponding to each station position comprises:

[0021] determining absolute positions of each map element in the local map subgraph based on an absolute position of the collection vehicle;

[0022] assigning absolute positions to each map element according to the absolute positions of each map element in the local map subgraph.

[0023] Optionally, the establishing the geographic fence corresponding to each site position based on the local map subgraph corresponding to each site position comprises:

[0024] According to the relative order of each site position, the local map subgraph corresponding to each site position is assigned a map subgraph identifier;

[0025] According to a preset geographic fence construction strategy, a geographic fence area of the local map subgraph corresponding to each site position is determined;

[0026] According to the geographic fence area of the local map subgraph corresponding to each site position, the geographic fence corresponding to each site position is established.

[0027] In a second aspect, the embodiments of the present application also provide a positioning method of an unmanned bus, wherein the method comprises:

[0028] Obtaining current positioning information of the unmanned bus, and determining a site position corresponding to the current positioning information of the unmanned bus;

[0029] According to the site position corresponding to the current positioning information of the unmanned bus, determining a geographic fence corresponding to the site position;

[0030] In a case where the current positioning information of the unmanned bus enters the geographic fence corresponding to the site position, obtaining current road image data collected by the unmanned bus;

[0031] According to the current road image data and a local map subgraph corresponding to the geographic fence, determining observation positioning information of the unmanned bus, and determining final positioning information of the unmanned bus according to the observation positioning information;

[0032] The geographic fence and the local map subgraph are constructed based on any one of the foregoing positioning map construction methods of the unmanned bus.

[0033] Optionally, the local map subgraph comprises a local point cloud map and a local semantic map, and the determining the observation positioning information of the unmanned bus according to the current road image data and the local map subgraph corresponding to the geographic fence comprises:

[0034] According to the current road image data and the local point cloud map corresponding to the geographic fence, determining first observation positioning information of the unmanned bus by using a first preset matching algorithm;

[0035] According to the current road image data and the local semantic map corresponding to the geographic fence, determining second observation positioning information of the unmanned bus by using a second preset matching algorithm;

[0036] determine the final observation positioning information according to the first observation positioning information and the second observation positioning information.

[0037] Optionally, the determining the final observation positioning information according to the first observation positioning information and the second observation positioning information comprises:

[0038] determining weights corresponding to the first observation positioning information and the second observation positioning information;

[0039] performing weighted processing on the first observation positioning information and the second observation positioning information according to the weights corresponding to the first observation positioning information and the second observation positioning information by using a preset weighting algorithm, to obtain the final observation positioning information.

[0040] In a third aspect, an embodiment of the present application further provides a positioning map construction device for an unmanned bus, wherein the device comprises:

[0041] a first acquisition unit configured to acquire road image data in a preset area range corresponding to each station position of the unmanned bus based on the station position;

[0042] a construction unit configured to construct a local map subgraph corresponding to each station position according to road image data in a preset area range corresponding to the station position, the local map subgraph comprising a local point cloud map and a local semantic map;

[0043] an establishment unit configured to establish a geographic fence corresponding to each station position according to the local map subgraph corresponding to the station position, to complete construction of a positioning map for the unmanned bus.

[0044] In a fourth aspect, an embodiment of the present application further provides a positioning device for an unmanned bus, wherein the device comprises:

[0045] a second acquisition unit configured to acquire current positioning information of the unmanned bus and determine a station position corresponding to the current positioning information of the unmanned bus;

[0046] a first determination unit configured to determine a geographic fence corresponding to the station position according to the station position corresponding to the current positioning information of the unmanned bus;

[0047] a third acquisition unit configured to acquire current road image data collected by the unmanned bus in a case where the current positioning information of the unmanned bus enters the geographic fence corresponding to the station position;

[0048] The second determining unit is used to determine the observation and positioning information of the unmanned bus based on the current road image data and the local map sub-map corresponding to the geofence, and to determine the final positioning information of the unmanned bus based on the observation and positioning information.

[0049] The geofence and the local map submap are constructed based on the aforementioned positioning map construction device for driverless buses.

[0050] Fifthly, embodiments of this application also provide an electronic device, including:

[0051] Processor; and

[0052] A memory configured to store computer-executable instructions, which, when executed, cause the processor to perform any of the aforementioned location map construction methods for driverless buses, or any of the aforementioned location methods for driverless buses.

[0053] Sixthly, embodiments of this application also provide a computer-readable storage medium storing one or more programs that, when executed by an electronic device including multiple applications, cause the electronic device to perform any of the aforementioned location map construction methods for unmanned buses, or any of the aforementioned location methods for unmanned buses.

[0054] The above-mentioned technical solutions adopted in the embodiments of this application can achieve the following beneficial effects: The positioning map construction method for unmanned buses in the embodiments of this application first obtains road image data within a preset area corresponding to each station location based on the station locations of the unmanned bus; then, based on the road image data within the preset area corresponding to each station location, a local map sub-map corresponding to each station location is constructed, the local map sub-map including a local point cloud map and a local semantic map; finally, based on the local map sub-maps corresponding to each station location, a geofence corresponding to each station location is established to complete the construction of the positioning map for unmanned buses. The positioning map construction method for unmanned buses in the embodiments of this application utilizes visual technology to construct corresponding local map sub-maps for each station location, including local point cloud maps and local semantic maps. Compared with the method of constructing a global map, this ensures the sparsity of the map and the positioning efficiency and accuracy. Attached Figure Description

[0055] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0056] Figure 1This is a flowchart illustrating a method for constructing a location map for an unmanned bus according to an embodiment of this application.

[0057] Figure 2 This is a flowchart illustrating a positioning method for an unmanned bus according to an embodiment of this application;

[0058] Figure 3 This is a schematic diagram of a positioning map building device for an unmanned bus according to an embodiment of this application;

[0059] Figure 4 This is a schematic diagram of the structure of a positioning device for an unmanned bus according to an embodiment of this application;

[0060] Figure 5 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0062] The technical solutions provided by the various embodiments of this application are described in detail below with reference to the accompanying drawings.

[0063] This application provides a method for constructing a location map for an unmanned bus, such as... Figure 1 The diagram illustrates a flowchart of a method for constructing a location map for an unmanned bus according to an embodiment of this application. The method includes at least the following steps S110 to S130:

[0064] Step S110: Based on the locations of each station of the unmanned bus, obtain road image data within a preset area corresponding to each station location.

[0065] Since the operating routes and travel areas of driverless buses are usually relatively fixed, the positioning map construction method for driverless buses in this application collects road image data within a preset area around each station location that the driverless bus may pass through in the overall operating route. The size and shape of the preset area can be flexibly set according to actual needs. For example, the area within a 2-kilometer radius of each station location can be used as the preset area for each station location. Of course, the specific setting is not specifically limited here.

[0066] S120, constructing a local map subgraph corresponding to each station position according to road image data in a preset area range corresponding to the station position, the local map subgraph including a local point cloud map and a local semantic map.

[0067] After obtaining the road image data in the preset area range corresponding to each station position, a corresponding local map subgraph needs to be constructed for each station position, so that the multiple local map subgraphs corresponding to the multiple stations can constitute a local map subgraph library. In addition, the embodiment of the present application can construct two kinds of local map subgraphs based on the road image data by using different map construction algorithms respectively, which can specifically include a local point cloud map and a local semantic map. Compared with a visual positioning scheme that simply relies on a semantic map, the embodiment of the present application provides more powerful support for improving positioning accuracy.

[0068] S130, establishing a geographic fence corresponding to each station position according to the local map subgraph corresponding to the station position, to complete the construction of the positioning map for the unmanned bus.

[0069] Since the local map subgraphs corresponding to each station position are independent of each other and are respectively used to realize the positioning of the unmanned bus in a certain area range corresponding to each station position, a corresponding geographic fence needs to be further established for the local map subgraph of each station position, so that whether the local map subgraph needs to be called for visual positioning and which local map subgraph of which station needs to be called for visual positioning can be determined according to the geographic fence in the subsequent positioning stage.

[0070] The positioning map construction method for the unmanned bus according to the embodiment of the present application constructs a corresponding local map subgraph for each station position by using a visual technology, including a local point cloud map and a local semantic map. Compared with the way of constructing a global map, the embodiment of the present application guarantees the sparsity of the map and the positioning efficiency and positioning accuracy.

[0071] In an embodiment of the present application, the step of constructing a local map subgraph corresponding to each station position according to road image data in a preset area range corresponding to the station position includes: constructing a local point cloud map corresponding to each station position by using a preset point cloud map construction algorithm according to road image data in a preset area range corresponding to the station position; and constructing a local semantic map corresponding to each station position by using a preset semantic map construction algorithm according to road image data in a preset area range corresponding to the station position.

[0072] The embodiment of the present application can construct the local point cloud map corresponding to each site position by using a preset point cloud map construction algorithm, such as the ORB-SLAM2 algorithm, based on the road image data in the preset area range corresponding to each site position. The ORB-SLAM2 algorithm is a mapping algorithm further supporting a calibrated binocular camera and an RGB-D camera based on the ORB-SLAM (Oriented FAST and Rotated BRIEF-Simultaneous Localization And Mapping). The ORB-SLAM is a feature point-based real-time SLAM system, which can run in large-scale, small-scale, indoor and outdoor environments, is very robust to violent motion, supports wide-baseline loop closure detection and repositioning, and can calculate the camera trajectory in real time and generate a sparse three-dimensional reconstruction result of the scene. The embodiment of the present application can realize the construction of the local point cloud map of each site position by extracting the ORB feature points in the road image data of each site position based on the existing ORB-SLAM2 algorithm.

[0073] The embodiment of the present application can also construct the local semantic map corresponding to each site position by using a preset semantic map construction algorithm based on the road image data in the preset area range corresponding to each site position. The preset semantic map construction algorithm can be realized based on a pre-trained road element recognition model, so as to construct the local semantic map through the recognized road elements.

[0074] Since different sites have different situations, the embodiment of the present application constructs the different road elements contained in the road image data in the preset area range corresponding to different sites, so that the matching positioning based on the local semantic map is not limited by the road surface elements in the high-precision map.

[0075] In an embodiment of the present application, the construction of the local semantic map corresponding to each site position by using the preset semantic map construction algorithm based on the road image data in the preset area range corresponding to each site position includes: identifying the road elements in the preset area range corresponding to each site position by using a preset road element recognition model based on the road image data in the preset area range corresponding to each site position, to obtain a road element recognition result; and constructing the local semantic map corresponding to each site position based on the road element recognition result.

[0076] The preset road element recognition model of the embodiment of the present application can be obtained based on an existing convolutional neural network such as a YOLO V5 network, and the road elements that need to be trained and recognized can specifically include bus stops, billboards, poles around the station, roadside business signs and other information of various station positions. The local semantic map is constructed based on these road elements. Compared with the scheme of constructing a semantic map by only relying on lane lines, arrows and other road surface elements, the problem of being unable to accurately locate due to the inability to well recognize road surface semantic elements in some special periods or special road sections is avoided.

[0077] The selection of the above road elements is based on the pre-trained preset road element recognition model, which ensures the precision of the preset road element recognition model and further ensures the positioning effect.

[0078] In an embodiment of the present application, the road image data is collected based on a collection vehicle, and the local map subgraph corresponding to each station position is constructed based on the road image data in the preset area range corresponding to each station position, which includes: determining the absolute position of each map element in the local map subgraph based on the absolute position of the collection vehicle; and assigning the absolute position of each map element according to the absolute position of each map element in the local map subgraph.

[0079] In the construction phase of the positioning map, the embodiment of the present application can collect road image data in the preset area range corresponding to each station position by a collection vehicle. Since the absolute position of the collection vehicle in the world coordinate system can be determined, the absolute position of each map element in the established local map subgraph can be further determined based on the absolute position of the collection vehicle, and then the absolute position of each map element is assigned.

[0080] The embodiment of the present application can adopt different assignment methods for the constructed local point cloud map and local semantic map. For example, each point in the local point cloud map can be assigned a latitude and longitude, and for the semantic elements in the local semantic map, different assignments can be made according to the different shapes of the semantic elements, such as assigning the four corners of a quadrilateral sign, assigning the upper and lower points of a pole around the station, and so on. Of course, how to assign the absolute position of the map elements in the local map subgraph can be flexibly adjusted according to actual needs by those skilled in the art, and is not limited here.

[0081] In an embodiment of the present application, the establishing of the geographic fence corresponding to each station position according to the local map subgraph corresponding to the station position comprises: assigning a map subgraph identifier to the local map subgraph corresponding to each station position based on the relative order of the station positions; determining a geographic fence region of the local map subgraph corresponding to each station position based on a preset geographic fence construction strategy; and establishing the geographic fence corresponding to each station position according to the geographic fence region of the local map subgraph corresponding to each station position.

[0082] The embodiments of the present application can further edit the constructed local map subgraph corresponding to each station position, for example, assigning or numbering a map subgraph identifier to the local map subgraph corresponding to each station position, the map subgraph identifier corresponding to each station position. Figure 1 In this way, the map subgraph identifier can be represented by a fixed ID, which can be set according to the order of the station positions, for example, the first station position is set as 1, the second station position is set as 2, and so on.

[0083] After the editing of each local map subgraph is completed, the map range corresponding to each local map subgraph, i.e., the geographic fence, needs to be further determined. There are two methods for determining the geographic fence. One method is to set a rectangular frame or any other shaped region frame containing the latitude and longitude of the constructed local map subgraph, and subsequently, whether the corresponding local map subgraph is called for positioning can be determined according to whether the vehicle is located in the region frame. The other method is to set the latitude and longitude information of the start point and end point of the current road segment, and since the road is one-way, whether the vehicle is located in the latitude and longitude between the start point and end point of the current road segment can be determined to determine whether the vehicle enters or exits the local map subgraph.

[0084] The embodiments of the present application also provide a positioning method of an unmanned bus, as shown in Figure 2 The embodiments of the present application also provide a positioning method of an unmanned bus, as shown in

[0085] In step S210, the current positioning information of the unmanned bus is acquired, and a station position corresponding to the current positioning information of the unmanned bus is determined.

[0086] In the positioning method of the unmanned bus according to the embodiments of the present application, the current positioning information of the unmanned bus needs to be acquired first. The current positioning information can be regarded as a preliminary positioning result, for example, the positioning information output by a combined navigation system. Based on the approximate position of the current positioning information, a station position corresponding to the position can be determined. The station can be a station closest to the current position of the unmanned bus and to be reached by the unmanned bus.

[0087] Step S220, determining a geographic fence corresponding to the station position according to the current positioning information of the bus.

[0088] After obtaining the station position corresponding to the current position of the bus, the geographic fence corresponding to the station position can be further obtained according to the station position. The geographic fence corresponding to each station can be established according to the establishment method of each of the foregoing embodiments. The main role is to facilitate the determination of which station the bus enters or is about to enter in the surrounding area during the positioning phase.

[0089] Step S230, obtaining the current road image data collected by the bus in the case that the current positioning information of the bus enters the geographic fence corresponding to the station position.

[0090] The geographic fence of the embodiment of the application can be a region frame surrounding the local map subgraph corresponding to the station, or a road segment containing the latitude and longitude information between the start point and the end point of the road segment where the station is located. If the first form of geographic fence is used, the current position of the bus can be compared with the boundary position of the region frame, so as to determine whether the bus enters or exits the geographic fence corresponding to the station. If the second form of geographic fence is used, the current position of the bus can be compared with the latitude and longitude of the start point and the end point of the road segment, so as to determine whether the bus enters or exits the geographic fence corresponding to the station.

[0091] When the current position of the bus falls within the geographic fence of a station, it indicates that the bus has entered or is about to enter the surrounding area of the station. At this time, the current road image data collected by the bus can be obtained as the basic data for subsequent positioning of the bus.

[0092] Step S240, determining the observation positioning information of the bus according to the current road image data and the local map subgraph corresponding to the geographic fence, and determining the final positioning information of the bus according to the observation positioning information; wherein the geographic fence and the local map subgraph are constructed based on any of the foregoing positioning map construction methods for the bus.

[0093] After determining the station corresponding to the current position of the bus and the geographic fence of the station, the local map subgraph corresponding to the station can be preloaded at the same time. In this way, when the bus enters the geographic fence region of the station subsequently, the preloaded local map subgraph can be directly called for positioning, thereby improving the positioning efficiency.

[0094] Of course, the local map subgraph corresponding to the station can also be loaded after it is determined that the unmanned bus has entered the geographic fence of the station. The specific time for loading the local map subgraph can be flexibly set by a person skilled in the art according to actual needs, and is not specifically limited here.

[0095] When positioning according to the current road image data and the local map subgraph corresponding to the geographic fence, the two can be matched, so that the matching result is provided as observation positioning information to a Kalman filter or an extended Kalman filter for subsequent fusion processing, and then the final positioning result is output.

[0096] The positioning method of the unmanned bus in the embodiment of the application uses a local map subgraph corresponding to each station, which is small in size, to replace a global map for positioning, thereby ensuring positioning efficiency and positioning accuracy.

[0097] In addition, it should be noted that the positioning method of the unmanned bus in the embodiment of the application is essentially a positioning method based on visual technology, which can be applied to a scenario in which the unmanned bus cannot be accurately positioned due to poor GNSS positioning signal quality or lack of road semantic elements when entering or leaving a station or the like. For areas with good GNSS positioning signal quality, fusion positioning can still be directly based on GNSS positioning signals and is not limited by the positioning method based on visual technology in the embodiment of the application.

[0098] In an embodiment of the application, the local map subgraph includes a local point cloud map and a local semantic map, and the determining the observation positioning information of the unmanned bus according to the current road image data and the local map subgraph corresponding to the geographic fence includes: determining first observation positioning information of the unmanned bus by using a first preset matching algorithm according to the current road image data and the local point cloud map corresponding to the geographic fence; determining second observation positioning information of the unmanned bus by using a second preset matching algorithm according to the current road image data and the local semantic map corresponding to the geographic fence; and determining the final observation positioning information according to the first observation positioning information and the second observation positioning information.

[0099] The local map subgraph constructed in advance by the embodiment of the application includes two forms of a local point cloud map and a local semantic map. Two different maps can respectively adopt different matching algorithms to obtain observation positioning information of the unmanned bus. For example, for the local point cloud map, a first preset matching algorithm such as an ORB feature point extracted based on an ORB-SLAM2 algorithm can be adopted for matching, so as to obtain first observation positioning information. In the scene of entering and exiting a station, since the speed of the unmanned bus is relatively low and the driving trajectory is relatively fixed, the local point cloud map can effectively track feature points to realize positioning. For the local semantic map, a second preset matching algorithm such as a road element recognized based on a road element recognition model trained in advance can be adopted for matching, so as to obtain second observation positioning information.

[0100] Due to the influence of factors such as recognition errors of the recognition model, the first observation positioning information and the second observation positioning information are likely to be deviated. Therefore, the embodiment of the application can further perform fusion processing on the first observation positioning information and the second observation positioning information, so as to obtain final observation positioning information.

[0101] In an embodiment of the application, the determining of the final observation positioning information according to the first observation positioning information and the second observation positioning information includes: determining a weight corresponding to the first observation positioning information and a weight corresponding to the second observation positioning information; and performing weighted processing on the first observation positioning information and the second observation positioning information by using a preset weighting algorithm according to the weight corresponding to the first observation positioning information and the weight corresponding to the second observation positioning information, to obtain the final observation positioning information.

[0102] One way of the embodiment of the application for performing fusion processing on the first observation positioning information and the second observation positioning information can be to perform weighted processing on the two by using a preset weighting algorithm to obtain final observation positioning information P. The preset weighting algorithm can be specifically expressed in the following form:

[0103] P = ω1*P1 + ω2*P2, (1)

[0104] ω2 = 1-ω1, (2)

[0105] wherein P1 is the first observation positioning information, P2 is the second observation positioning information, ω1 is the weight of the first observation positioning information P1, and ω2 is the weight of the second observation positioning information P2.

[0106] Based on the above formula (1)-(2), it can be seen that the key of the preset weighting algorithm is to determine the weight of the first observation positioning information and the second observation positioning information. Since the first observation positioning information is obtained based on the ORB feature point matching, it is relatively easy to quantify the number of matching point pairs, and then ω1 can be calculated according to the number of matching point pairs and the total number of extracted feature points, and an upper bound of the number of matching point pairs is set to judge the absolute credibility of the matching result.

[0107] For example, a total of 120 feature points are identified, only 70 pairs of matching points are successful, and the upper bound of the number of matching point pairs is 100, then ω1=(70 / 100)*100%=70%, and correspondingly, ω2=1-ω1=30%.

[0108] Similarly, a lower bound of the number of matching point pairs can also be set to judge whether the first observation positioning information is credible, and a lower bound of the confidence of the matching result of the road element can be set to judge whether the second observation positioning information is credible. If both are not credible, the output will be unable to locate the information, thereby ensuring the positioning accuracy.

[0109] The embodiment of the application also provides a positioning map construction device 300 for an unmanned bus, as shown in Figure 3 The structure schematic diagram of the positioning map construction device for the unmanned bus in the embodiment of the application is provided, and the device 300 comprises a first acquisition unit 310, a construction unit 320 and a establishing unit 330, wherein:

[0110] The first acquisition unit 310 is used for acquiring road image data in a preset area range corresponding to each station position based on each station position of the unmanned bus;

[0111] The construction unit 320 is used for constructing a local map subgraph corresponding to each station position according to the road image data in the preset area range corresponding to each station position, and the local map subgraph comprises a local point cloud map and a local semantic map;

[0112] The establishing unit 330 is used for establishing a geographic fence corresponding to each station position according to the local map subgraph corresponding to each station position, so as to complete the construction of the positioning map for the unmanned bus.

[0113] In an embodiment of the application, the construction unit 320 is specifically used for: constructing a local point cloud map corresponding to each station position by using a preset point cloud map construction algorithm according to the road image data in the preset area range corresponding to each station position; and constructing a local semantic map corresponding to each station position by using a preset semantic map construction algorithm according to the road image data in the preset area range corresponding to each station position.

[0114] In one embodiment of this application, the construction unit 320 is specifically used to: identify road elements within a preset area corresponding to each station location based on road image data within a preset area corresponding to each station location using a preset road element recognition model, and obtain road element recognition results; and construct a local semantic map corresponding to each station location based on the road element recognition results.

[0115] In one embodiment of this application, the road image data is collected by a data acquisition vehicle, and the construction unit 320 is specifically used to: determine the absolute position of each map element in the local map sub-map based on the absolute position of the data acquisition vehicle; and assign values ​​to the absolute positions of each map element according to the absolute positions of each map element in the local map sub-map.

[0116] In one embodiment of this application, the establishment unit 330 is specifically used to: assign map sub-map identifiers to the local map sub-maps corresponding to each site location based on the relative order of each site location; determine the geofence area of ​​the local map sub-maps corresponding to each site location based on a preset geofence construction strategy; and establish geofences corresponding to each site location based on the geofence areas of the local map sub-maps corresponding to each site location.

[0117] It is understood that the above-mentioned positioning map building device for unmanned buses can realize each step of the positioning map building method for unmanned buses provided in the foregoing embodiments. The relevant explanations of the positioning map building method for unmanned buses are applicable to the positioning map building device for unmanned buses, and will not be repeated here.

[0118] This application embodiment also provides a positioning device 400 for an unmanned bus, such as... Figure 4 The diagram shows a schematic representation of a positioning device for an unmanned bus according to an embodiment of this application. The device 400 includes: a second acquisition unit 410, a first determination unit 420, a third acquisition unit 430, and a second determination unit 440, wherein:

[0119] The second acquisition unit 410 is used to acquire the current location information of the unmanned bus and determine the station location corresponding to the current location information of the unmanned bus.

[0120] The first determining unit 420 is used to determine the geofence corresponding to the station location based on the station location corresponding to the current positioning information of the unmanned bus.

[0121] The third acquisition unit 430 is used to acquire the current road image data collected by the unmanned bus when the current location information of the unmanned bus enters the geofence corresponding to the station location.

[0122] The second determining unit 440 is configured to determine observation positioning information of the bus based on the current road image data and the local map subgraph corresponding to the geo-fence, and determine final positioning information of the bus based on the observation positioning information.

[0123] The local map subgraph and the geo-fence are constructed based on the positioning map construction device for the bus.

[0124] In an embodiment of the present application, the local map subgraph includes a local point cloud map and a local semantic map, and the second determining unit 440 is specifically configured to: determine first observation positioning information of the bus based on the current road image data and the local point cloud map corresponding to the geo-fence by using a first preset matching algorithm; determine second observation positioning information of the bus based on the current road image data and the local semantic subgraph corresponding to the geo-fence by using a second preset matching algorithm; and determine the final observation positioning information based on the first observation positioning information and the second observation positioning information.

[0125] In an embodiment of the present application, the second determining unit 440 is specifically configured to: determine a weight corresponding to the first observation positioning information and a weight corresponding to the second observation positioning information; and perform weighted processing on the first observation positioning information and the second observation positioning information by using a preset weighting algorithm based on the weight corresponding to the first observation positioning information and the weight corresponding to the second observation positioning information, to obtain the final observation positioning information.

[0126] It can be understood that the positioning device for the bus can realize each step of the positioning method for the bus provided in the foregoing embodiments, and the related explanations about the positioning method for the bus are all applicable to the positioning device for the bus, which will not be repeated here.

[0127] Figure 5 FIG. 1 is a structural schematic diagram of an electronic device according to an embodiment of the present application. Please refer to Figure 5 At the hardware level, the electronic device includes a processor, and optionally further includes an internal bus, a network interface, and a memory. The memory can include a memory such as a random-access memory (RAM), and can further include a non-volatile memory such as at least one disk memory. Of course, the electronic device can further include other hardware required by a business.

[0128] The processor, the network interface and the memory can be connected with each other through an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation, Figure 5 Only one bidirectional arrow is used to represent the bus, but it does not mean that there is only one bus or only one type of bus.

[0129] The memory is used to store programs. Specifically, the program can include program code including computer operation instructions. The memory can include an internal memory and a non-volatile memory, and provide instructions and data for the processor.

[0130] The processor reads the corresponding computer program from the non-volatile memory into the internal memory and then runs, and forms a positioning map construction device for the unmanned bus at a logical level. The processor executes the program stored in the memory, and is specifically used for executing the following operations:

[0131] Based on the positions of the various sites of the unmanned bus, road image data within a preset area range corresponding to each site position is obtained;

[0132] According to the road image data within the preset area range corresponding to each site position, a local map subgraph corresponding to each site position is constructed, and the local map subgraph includes a local point cloud map and a local semantic map;

[0133] According to the local map subgraph corresponding to each site position, a geographic fence corresponding to each site position is established, so as to complete the construction of the positioning map for the unmanned bus.

[0134] The above as described in the present application Figure 1The method for constructing the positioning map of the unmanned bus disclosed by the embodiment can be applied in a processor or implemented by the processor. The processor can be an integrated circuit chip with signal processing capability. In the implementation process, each step of the above method can be completed by integrated logic circuits of hardware in the processor or instructions in the form of software. The above processor can be a general processor, including a central processing unit (CPU), a network processor (NP), etc. It can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. Each method, step and logic block disclosed in the embodiment of the present application can be implemented or executed. The general processor can be a microprocessor or any conventional processor. The steps of the method disclosed in combination with the embodiment of the present application can be directly embodied as a hardware decoding processor for execution, or a combination of hardware and software modules in the decoding processor for execution. The software module can be located in a mature storage medium in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and combines the hardware to complete the steps of the above method.

[0135] The electronic device can also execute the method for constructing the positioning map of the unmanned bus in the embodiment, and implement the function of the device for constructing the positioning map of the unmanned bus in the embodiment. The embodiment of the present application will not be repeated here. Figure 1 The electronic device can also execute the method for constructing the positioning map of the unmanned bus in the embodiment, and implement the function of the device for constructing the positioning map of the unmanned bus in the embodiment. The embodiment of the present application will not be repeated here. Figure 1 The electronic device can also execute the method for constructing the positioning map of the unmanned bus in the embodiment, and implement the function of the device for constructing the positioning map of the unmanned bus in the embodiment. The embodiment of the present application will not be repeated here.

[0136] The embodiment of the present application also proposes a computer readable storage medium storing one or more programs, the one or more programs including instructions that, when executed by an electronic device including a plurality of applications, can cause the electronic device to execute the method for constructing the positioning map of the unmanned bus in the embodiment, and specifically for executing: Figure 1 The electronic device can also execute the method for constructing the positioning map of the unmanned bus in the embodiment, and implement the function of the device for constructing the positioning map of the unmanned bus in the embodiment. The embodiment of the present application will not be repeated here.

[0137] Based on the positions of each station of the unmanned bus, the road image data in the preset area range corresponding to each station position is obtained;

[0138] According to the road image data in the preset area range corresponding to each station position, a local map subgraph corresponding to each station position is constructed, and the local map subgraph includes a local point cloud map and a local semantic map;

[0139] According to the local map subgraph corresponding to each station position, a geofence corresponding to each station position is established, so as to complete construction of a positioning map for the unmanned bus.

[0140] Those skilled in the art will understand that embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage, etc.) containing computer-usable program code.

[0141] The present application is described with reference to flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, as well as combinations of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing apparatus produce an apparatus that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 means for performing the function specified by one or more blocks.

[0142] These computer program instructions can also be stored in a computer-readable memory that can direct the computer or other programmable data processing apparatus to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including instruction means, which implements the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 one or more blocks.

[0143] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, so that a series of operation steps are performed on the computer or other programmable data processing apparatus to produce a computer-implemented process, so that the instructions executed on the computer or other programmable data processing apparatus provide a process for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 2 one or more blocks.

[0144] In one typical arrangement, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0145] The memory can include non-persistent memory and / or other volatile memory, representing applications instructions stored in the computer readable media for execution by the processing unit. The memory can also include non-volatile memory, such as read only memory (ROM) and / or flash memory, electrically programmable read only memory (EPROM), or electrically erasable programmable read only memory (EEPROM) memory.

[0146] Computer readable media includes permanent and non-permanent, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory (ROM), electrically programmable read only memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital versatile disc (DVD), or other optical storage, magnetic cassette, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible to a computing device. According to the definition herein, computer readable media does not include transitory media such as modulated data signals and carrier waves.

[0147] It should be noted that the electronic device of the embodiments of the present application can also perform the method performed by the positioning device of the unmanned bus and realize the functions of the positioning device of the unmanned bus in the embodiments shown in the above Figure 2 It should be noted that the electronic device of the embodiments of the present application can also perform the method performed by the positioning device of the unmanned bus and realize the functions of the positioning device of the unmanned bus in the embodiments shown in the above ​ It should be noted that the electronic device of the embodiments of the present application can also perform the method performed by the positioning device of the unmanned bus and realize the functions of the positioning device of the unmanned bus in the embodiments shown in the above

[0148] It should be noted that the electronic device of the embodiments of the present application can also perform the method performed by the positioning device of the unmanned bus and realize the functions of the positioning device of the unmanned bus in the embodiments shown in the above

[0149] Those skilled in the art will appreciate that embodiments of the present application can be devised for a method, a system, or a computer program product. Accordingly, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage devices, etc.) embodying computer-readable program code thereon for use by or in connection with an instruction execution system. For the purposes of this description, a computer-usable or computer readable storage medium can be any apparatus that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.

[0150] The foregoing is merely illustrative of the principles of the application and various modifications can be made by persons skilled in the art. The present application is not intended to be limited to the embodiments shown, but is to be accorded the full scope that resides in the art thereof. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A positioning map construction method for an unmanned bus, wherein, The method comprises: Based on the respective station positions of the unmanned bus, road image data within the respective station positions corresponding preset area ranges is acquired; According to the road image data within the respective station positions corresponding preset area ranges, respective station positions corresponding local map subgraphs are constructed, the local map subgraphs comprising local point cloud maps and local semantic maps; According to the respective station positions corresponding local map subgraphs, respective station positions corresponding geofences are established to complete the construction of the positioning map for the unmanned bus; The respective station positions corresponding local map subgraphs are independent of each other and are respectively used to realize the positioning of the unmanned bus within the respective station positions corresponding preset area ranges; The respective station positions corresponding local map subgraphs, the establishment of the respective station positions corresponding geofences comprises: Based on the relative order of the respective station positions, map subgraph identification is assigned to the respective station positions corresponding local map subgraphs; Based on a preset geofence construction strategy, geofence areas of the respective station positions corresponding local map subgraphs are determined; According to the geofence areas of the respective station positions corresponding local map subgraphs, the respective station positions corresponding geofences are established; The map subgraph identification corresponds one-to-one to the respective station positions corresponding local map subgraphs; The preset geofence construction strategy comprises at least one of the following: An area frame containing the longitude and latitude of the constructed local map subgraph is set to determine whether the corresponding local map subgraph is to be called to perform positioning according to whether the positioning of the vehicle is in the area frame; The longitude and latitude information of the starting point and the ending point of a current road section is set to determine whether the corresponding local map subgraph is to be called to perform positioning according to whether the positioning of the vehicle is between the longitude and latitude of the starting point and the ending point of the current road section.

2. The method of claim 1, wherein, The respective station positions corresponding local map subgraphs are constructed according to the road image data within the respective station positions corresponding preset area ranges comprises: According to the road image data within the respective station positions corresponding preset area ranges, the respective station positions corresponding local point cloud maps are constructed by using a preset point cloud map construction algorithm; According to the road image data within the respective station positions corresponding preset area ranges, the respective station positions corresponding local semantic maps are constructed by using a preset semantic map construction algorithm.

3. The method of claim 2, wherein, The respective station positions corresponding local semantic maps are constructed according to the road image data within the respective station positions corresponding preset area ranges by using a preset semantic map construction algorithm comprises: According to the road image data within the respective station positions corresponding preset area ranges, road elements within the respective station positions corresponding preset area ranges are identified by using a preset road element identification model to obtain road element identification results; According to the road element identification results, the respective station positions corresponding local semantic maps are constructed.

4. The method of claim 1, wherein, The road image data is acquired by a collection vehicle, and the respective station positions corresponding local map subgraphs are constructed according to the road image data within the respective station positions corresponding preset area ranges comprises: determine absolute positions of the map elements in the local map subgraph based on the absolute position of the collection vehicle; assign absolute positions to the map elements in the local map subgraph based on the absolute positions of the map elements.

5. A method of positioning an unmanned bus, wherein, The method comprises: obtaining current positioning information of the unmanned bus and determining a station position corresponding to the current positioning information of the unmanned bus; determining a geofence corresponding to the station position according to the station position corresponding to the current positioning information of the unmanned bus; in a case where the current positioning information of the unmanned bus enters the geofence corresponding to the station position, obtaining current road image data collected by the unmanned bus; determining observed positioning information of the unmanned bus according to the current road image data and a local map subgraph corresponding to the geofence, and determining final positioning information of the unmanned bus according to the observed positioning information; wherein the geofence and the local map subgraph are constructed based on the positioning map construction method for the unmanned bus according to any one of claims 1-4.

6. The method of claim 5, wherein, The local map subgraph comprises a local point cloud map and a local semantic map, and the determination of the observed positioning information of the unmanned bus according to the current road image data and the local map subgraph corresponding to the geofence comprises: determining first observed positioning information of the unmanned bus according to the local point cloud map corresponding to the geofence and the current road image data by using a first preset matching algorithm; determining second observed positioning information of the unmanned bus according to the local semantic subgraph corresponding to the geofence and the current road image data by using a second preset matching algorithm; determining the final observed positioning information according to the first observed positioning information and the second observed positioning information.

7. The method of claim 6, wherein, The determination of the final observed positioning information according to the first observed positioning information and the second observed positioning information comprises: determining weights corresponding to the first observed positioning information and weights corresponding to the second observed positioning information; performing weighted processing on the first observed positioning information and the second observed positioning information by using a preset weighting algorithm according to the weights corresponding to the first observed positioning information and the weights corresponding to the second observed positioning information, to obtain the final observed positioning information.

8. A positioning map construction device for an unmanned bus, wherein, The device comprises: a first obtaining unit configured to obtain road image data in a preset area range corresponding to each station position of the unmanned bus based on the station positions of the unmanned bus; a construction unit configured to construct a local map subgraph corresponding to each station position based on the road image data in the preset area range corresponding to the station position, the local map subgraph comprising a local point cloud map and a local semantic map; an establishment unit configured to establish a geofence corresponding to each station position based on the local map subgraph corresponding to the station position, to complete the construction of the positioning map for the unmanned bus; the local map subgraphs corresponding to the station positions are independent of each other and are respectively used to realize positioning of the unmanned bus in the preset area range corresponding to the station position; the establishment unit is specifically configured to: According to the relative order of the site positions, the local map subgraph corresponding to each site position is assigned a map subgraph identifier; According to the preset geographic fence construction strategy, a geographic fence area of the local map subgraph corresponding to each site position is determined; According to the geographic fence area of the local map subgraph corresponding to each site position, a geographic fence corresponding to each site position is established; The map subgraph identifier corresponds to the local map subgraph of each site position one-to-one; The preset geographic fence construction strategy includes at least one of the following: A region box containing the longitude and latitude of the constructed local map subgraph is set to determine whether the corresponding local map subgraph is called for positioning according to whether the positioning of the vehicle is in the region box; The longitude and latitude information of the starting point and the ending point of the current road section is set to determine whether the corresponding local map subgraph is called for positioning according to whether the positioning of the vehicle is between the longitude and latitude of the starting point and the ending point of the current road section.

9. A positioning device for an unmanned bus, wherein, The device comprises: A second acquisition unit is configured to acquire the current positioning information of the unmanned bus and determine the site position corresponding to the current positioning information of the unmanned bus; A first determination unit is configured to determine the geographic fence corresponding to the site position according to the site position corresponding to the current positioning information of the unmanned bus; A third acquisition unit is configured to acquire the current road image data collected by the unmanned bus when the current positioning information of the unmanned bus enters the geographic fence corresponding to the site position; A second determination unit is configured to determine the observation positioning information of the unmanned bus according to the current road image data and the local map subgraph corresponding to the geographic fence, and determine the final positioning information of the unmanned bus according to the observation positioning information; The geographic fence and the local map subgraph are constructed based on the positioning map construction device for unmanned buses of claim 8.

10. An electronic device comprising: a processor; and a memory arranged to store computer executable instructions that, when executed, cause the processor to perform the positioning map construction method for unmanned buses of any one of claims 1-4, or the positioning method of the unmanned bus of any one of claims 5-7.

11. A computer readable storage medium storing one or more programs, which when executed by an electronic device comprising a plurality of applications, cause the electronic device to perform the positioning map construction method for unmanned buses of any one of claims 1-4, or the positioning method of the unmanned bus of any one of claims 5-7. ​

Images