Vector map construction method and device based on driving test subject, and electronic equipment
By combining the data of lidar sensors and inertial measurement units, point cloud maps are generated and vector elements are drawn, and the problem of insufficient accuracy and robustness of traditional positioning and map construction in dynamic environments is solved, achieving more efficient and safe Operation of driverless vehicles.
Patent Information
- Application Number
- CN202411080901.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-07
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-08-07
AI Technical Summary
Traditional GPS positioning is inaccurate, visual SLAM is not robust in dynamic environments, and sensor fusion algorithm based on Kalman filtering is prone to accumulate errors after long-term use, resulting in driverless cars going from point cloud map to vector map The operation is complicated and difficult.
By obtaining the point cloud data and inertial data collected by the target sensor, a point cloud map is generated, and vector elements are drawn based on the environmental characteristics in the point cloud map, a vector map is generated, and a combination of lidar sensors and inertial measurement units is used to improve positioning accuracy and Robustness, avoiding errors that rely on visual features and Kalman filtering algorithms.
It realizes more accurate and stable positioning and map construction in a dynamic environment, generates vector maps containing rich environmental characteristics, improves the understanding and adaptability of driverless vehicles in driving school scenarios, and ensures safer and more efficient driving behavior. , save manpower and material resources for driving schools.
Smart Images

Figure CN119085626A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of positioning and mapping technology, and specifically to a method and device for constructing a vector map based on driving test subjects, and an electronic device. Background Art
[0002] With the advancement of artificial intelligence and computer technology, driverless technology has become a research hotspot and important direction in the field of artificial intelligence, especially its application in driving school scenarios, which can save a lot of manpower and material resources.
[0003] Although there are some methods for autonomous robot positioning and mapping in known environments, traditional GPS (Global Positioning System) positioning is unreliable due to errors, visual SLAM (Simultaneous Localization and Mapping) lacks accuracy and robustness in dynamic environments, and sensor fusion algorithms based on Kalman filtering may accumulate large errors after long-term use. Therefore, driverless cars need to go through a series of complex processes from sensor scanning, generating point cloud maps to making vector maps. These processes are difficult to operate and require a high degree of technical expertise. Summary of the invention
[0004] In view of this, the embodiments of the present application provide a method and device for constructing a vector map based on driving test subjects, and an electronic device.
[0005] In a first aspect, an embodiment of the present application provides a vector map construction method based on driving test subjects, which is applied to a main control computer, and the method includes: obtaining point cloud data and inertial data of a target driving test area collected by a target sensor; generating a point cloud map of the target driving test area based on the point cloud data and inertial data; drawing vector elements based on environmental features contained in the point cloud map; and generating a vector map of the target driving test area based on the vector elements.
[0006] In combination with the first aspect, in some implementations of the first aspect, the environmental features include storage location information corresponding to the target driving test subject. Drawing vector elements based on the environmental features contained in the point cloud map includes: based on the storage location information corresponding to the target driving test subject, drawing multiple navigation points corresponding to the storage location information, each navigation point including semantic information, which is used to mark the geographic coordinates of the operation position in the storage location information.
[0007] In combination with the first aspect, in certain implementations of the first aspect, a vector map of the target driving test area is generated based on vector elements, including: based on semantic information corresponding to each of the multiple navigation points, moving the multiple navigation points to positions corresponding to the storage location information; based on the storage location information, generating an electronic fence corresponding to the storage location information, and obtaining a vector map of the target driving test subject in the target driving test area.
[0008] In combination with the first aspect, in certain implementations of the first aspect, an electronic fence corresponding to the storage location information is generated based on the storage location information, including: generating a virtual boundary based on the boundary information in the storage location information; adding trigger points in the virtual boundary to generate an electronic fence corresponding to the storage location information, wherein the trigger points are used to define key positions of the virtual boundary to trigger specific events.
[0009] In combination with the first aspect, in certain implementations of the first aspect, the target sensor includes a laser radar sensor and an inertial measurement unit, the point cloud data is collected by the laser radar sensor, and the inertial data is collected by the inertial measurement unit. Based on the point cloud data and the inertial data, a point cloud map of the target driving test area is generated, including: synchronizing the point cloud data and the inertial data; obtaining the external parameter data of the laser radar sensor and the inertial measurement unit in the configuration file in the target driver package, and estimating the initial motion trajectory of the target object based on the external parameter data and the synchronized point cloud data and inertial data; based on the initial motion trajectory of the target object, aligning the point cloud data to the standard coordinate system, and accumulating to form a point cloud map of the target driving test area.
[0010] In combination with the first aspect, in certain implementations of the first aspect, after generating a point cloud map of the target driving test area based on the point cloud data and the inertial data, it also includes: obtaining a first parameter value corresponding to a storage parameter in a configuration file in a target driver package; if the first parameter value is true, obtaining a second parameter value corresponding to a path parameter in the configuration file; and storing the point cloud map in a storage path indicated by the second parameter value.
[0011] In combination with the first aspect, in certain implementations of the first aspect, before acquiring the point cloud data and inertial data of the target driving test area, it also includes: testing the target sensor. The testing of the target sensor includes: determining a physical connection with the target sensor through the target port device file; setting communication parameters that match the configuration parameters of the target sensor; receiving a data stream sent by the target sensor, and testing the target sensor based on the data stream.
[0012] In the second aspect, an embodiment of the present application provides a construction device based on a vector map, which is applied to a main control computer, and the device includes: an acquisition module, which is used to acquire point cloud data and inertial data of a target driving test area collected by a target sensor; a first generation module, which is used to generate a point cloud map of the target driving test area based on the point cloud data and inertial data; a drawing module, which is used to draw vector elements based on environmental features contained in the point cloud map; and a second generation module, which is used to generate a vector map of the target driving test area based on the vector elements.
[0013] In a third aspect, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program for executing the vector map construction method based on driving test subjects described in the first aspect.
[0014] In a fourth aspect, an embodiment of the present application provides an electronic device, comprising: a processor; a memory for storing processor executable instructions; the processor is used to execute the vector map construction method based on driving test subjects described in the first aspect.
[0015] In this application, firstly, point cloud data provides the precise spatial structure of the environment, while inertial data provides the motion state of the sensor or vehicle. The combination of the two improves the accuracy and robustness of positioning, as well as the accuracy and reliability of map construction. Secondly, compared with visual SLAM, this solution does not rely on the visual features of the environment, so it can still maintain good performance in dynamic or texture-deficient environments. Furthermore, this solution achieves more stable and reliable positioning and map construction by fusing multiple data, avoiding the problem of cumulative errors that may occur after long-term operation of sensor fusion methods that rely solely on the Kalman filter algorithm. Ultimately, this solution can generate a vector map containing rich environmental features. Through this method, unmanned vehicles can better understand and adapt to driving school scenarios, achieve safer and more efficient driving behaviors, and save manpower and material resources for driving schools. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.
[0017] Figure 1 The figure is a flow chart of a vector map construction method provided in one embodiment of the present application.
[0018] Figure 2The figure is a schematic diagram of a process of drawing vector elements provided by an embodiment of the present application.
[0019] Figure 3 Shown is a schematic diagram of a point cloud map containing storage location information provided by an embodiment of the present application.
[0020] Figure 4 Shown is a schematic diagram of a point cloud map of navigation points containing storage location information provided by an embodiment of the present application.
[0021] Figure 5 Shown is a schematic diagram of a process for generating a vector map of a target driving test area provided in an embodiment of the present application.
[0022] Figure 6 The figure is a schematic diagram of a process for generating an electronic fence corresponding to storage location information provided by an embodiment of the present application.
[0023] Figure 7 Shown is a schematic diagram of an electronic fence including trigger points provided in one embodiment of the present application.
[0024] Figure 8 Shown is a schematic diagram of the process of generating a point cloud map of a target driving test area provided in one embodiment of the present application.
[0025] Fig. 9 Shown is a flow chart of a method for constructing a vector map provided in another embodiment of the present application.
[0026] Fig.10 The figure is a schematic diagram of a process of testing a target sensor provided in an embodiment of the present application.
[0027] Fig.11 Shown is a schematic structural diagram of a vector map construction device based on driving test subjects provided in one embodiment of the present application.
[0028] Fig.12 Shown is a schematic structural diagram of an electronic device provided in one embodiment of the present application. DETAILED DESCRIPTION
[0029] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0030] Figure 1 FIG. 1 is a flow chart of a vector map construction method provided by an embodiment of the present application. For example, the method is applied to a host computer. Figure 1As shown, the method includes the following steps.
[0031] Step S110, obtaining point cloud data and inertial data of the target driving test area collected by the target sensor.
[0032] A point cloud is a collection of points in three-dimensional space, each with coordinates and other information, such as color or intensity. In self-driving vehicles, point cloud data is often generated by lidar sensors. Lidar detects the environment by emitting laser pulses and measuring the time it takes for those pulses to reflect back. These measurements are used to create an accurate three-dimensional map of the surroundings, including obstacles, road boundaries, and other features.
[0033] Inertial data is provided by an IMU (Inertial Measurement Unit), which is used to determine the orientation (attitude) of an object, as well as changes in linear and rotational motion. In an autonomous vehicle, an IMU can help the vehicle estimate its changes in position in space, even indoors where GPS signals are unreliable or unavailable.
[0034] Exemplarily, the target sensors include a lidar sensor and an inertial measurement unit. In this step, first ensure that the lidar and IMU are correctly installed and working properly, for example, by using tools such as cutecom to test whether these two types of sensors can provide data. After confirming that the above sensors are working properly, start the IMU and use relevant instructions to obtain IMU data (i.e., inertial data), and configure the network interface of the lidar sensor to ensure that the relevant data can be received correctly. Subsequently, execute the recording command on the host computer to start collecting real-time data from the lidar and IMU, which will be saved in a file for subsequent processing.
[0035] Step S120, generating a point cloud map of the target driving test area based on the point cloud data and inertial data.
[0036] For example, in this embodiment, point cloud data collected by a lidar sensor is used, which contains the three-dimensional geometric information of the driving test area. Then, the inertial data provided by the IMU is used to correct the point cloud data to ensure the precise alignment and stability of the point cloud. Furthermore, through sensor fusion technology, such as using the LIO-mapping method, the three-dimensional modeling of the environment is performed in real time, and the consistency and accuracy of the point cloud map are maintained in a dynamic environment.
[0037] Step S130: drawing vector elements based on the environmental features contained in the point cloud map.
[0038] Environmental features refer to the identifiable and useful geographical or physical attributes of the target driving test area contained in the point cloud map. These features include but are not limited to: roads and paths, obstacles, traffic signals, terrain features, special areas, etc.
[0039] Vector elements refer to the conversion of these environmental features into graphics and objects that can be represented on a vector map, usually including: lines (representing roads, boundaries, etc.), polygons (representing areas, such as parking spaces or special test areas), marks (representing specific points, such as points of interest or specific locations), and text labels (providing additional information about a feature, such as road names or text on signs). In addition, in the process of drawing vector elements, you can use software or tools such as Unity 3D to convert environmental features into vector elements through analysis of point cloud maps, which are used to guide the path planning, navigation, and decision-making of driverless cars.
[0040] Step S140, generating a vector map of the target driving test area based on the vector elements.
[0041] Vector maps are a type of map that uses mathematical algorithms to represent geospatial data. Unlike raster maps (which are made up of a grid of pixels), vector maps use object-based graphics to represent real-world entities such as roads, buildings, rivers, etc. Each object is made up of geometric elements such as points, lines, and polygons, which are defined by mathematical formulas and can be precisely positioned and resized.
[0042] For example, in this embodiment, the drawn vector elements are exported to a format suitable for a specific application, such as a specific vector map format for autonomous vehicle navigation. The exported vector map will contain all key information of the driving test area, such as the road network, traffic rules, and driving test specific areas, providing necessary navigation and decision support for autonomous vehicles.
[0043] In this embodiment, the point cloud data provides the precise spatial structure of the environment, while the inertial data provides the motion state of the sensor or vehicle. The combination of the two improves the accuracy and robustness of positioning, as well as the accuracy and reliability of map construction. Secondly, compared with visual SLAM, this solution does not rely on the visual features of the environment, so it can still maintain good performance in dynamic or texture-deficient environments. Furthermore, this solution achieves more stable and reliable positioning and map construction by fusing multiple data, avoiding the cumulative error problem that may occur after long-term operation of the sensor fusion method that relies solely on the Kalman filter algorithm. Ultimately, this solution can generate a vector map containing rich environmental features. Through this method, unmanned vehicles can better understand and adapt to driving school scenarios, achieve safer and more efficient driving behaviors, and save manpower and material resources for driving schools.
[0044] Figure 2 FIG. 1 is a flow chart of drawing vector elements provided by an embodiment of the present application. Figure 1 Based on the embodiment shown, Figure 2 The embodiment shown is described below in detail. Figure 2 The embodiment shown and Figure 1 The differences and similarities of the illustrated embodiments are not described in detail.
[0045] In this embodiment, the environmental features include the storage location information corresponding to the target driving test subject. The target driving test subject refers to a specific subject in the driving test, such as subject one, subject two, subject three or subject four. The storage location information refers to the data storage location or the record location in the database related to these subjects. For example, Figure 3 Shown is a schematic diagram of a point cloud map containing storage location information provided by an embodiment of the present application.
[0046] like Figure 2 As shown, drawing vector elements based on environmental features included in the point cloud map includes step S210.
[0047] Step S210: based on the storage location information corresponding to the target driving test subject, draw a plurality of navigation points corresponding to the storage location information.
[0048] Each navigation point contains semantic information, which is used to mark the geographic coordinates of the operating position in the storage location information. Specifically, a navigation point is a specific point on the map that indicates the location that the vehicle should reach or pay attention to during driving. These points can be actual geographic coordinates or virtual marking points used to assist navigation and route planning. In the driving test subjects, the navigation points corresponding to the storage location information are those points that are directly related to the test requirements. For example, if the subject requires the candidate to park at a specific location, then this parking point is a navigation point corresponding to the storage location information. Exemplarily, Figure 4 Shown is a schematic diagram of a point cloud map of navigation points containing storage location information provided by an embodiment of the present application.
[0049] In some embodiments, the navigation points that need to be drawn are first determined according to the requirements of the driving test subjects. Then, precise geographic coordinates are assigned to each point to ensure that the vehicle can accurately navigate to these locations. Next, semantic information is added to each point, which describes the specific role of the point in the test, such as parking points, turn instructions, etc.
[0050] In this embodiment, the accuracy and relevance of the navigation points are ensured by drawing vector elements based on the environmental features of the point cloud map and combining the storage location information to determine the geographic coordinates of the operating position. The semantic information contained in each navigation point not only provides clear instructions about the function and purpose of the point, but also enhances the understanding of the driving test environment, thereby improving the accuracy of path planning and vehicle navigation, and also improving the efficiency and effectiveness of driving tests, providing strong support for the application of autonomous driving technology in the field of training.
[0051] Figure 5 The figure shows a flow chart of generating a vector map of a target driving test area provided by an embodiment of the present application. Figure 2 Based on the embodiment shown, Figure 5 The embodiment shown is described below in detail. Figure 5 The embodiment shown and Figure 2 The differences and similarities of the illustrated embodiments are not described in detail.
[0052] like Figure 5 As shown, in this embodiment, a vector map of the target driving test area is generated based on vector elements, including the following steps.
[0053] Step S510: based on the semantic information corresponding to each of the multiple navigation points, the multiple navigation points are moved to positions corresponding to the storage location information.
[0054] For example, in this embodiment, the semantic information of each navigation point is parsed, and after understanding the semantic information, the corresponding position of the navigation point in the actual driving test area is identified based on the description. Subsequently, the geographic coordinates of the navigation point are adjusted using the semantic information to ensure that they match the positions defined in the storage location information. For example, if the semantic information indicates that a navigation point is a parking point, it is necessary to ensure that the position of the point on the map corresponds to the actual parking area.
[0055] In addition, in some other embodiments, further refinement of the navigation points is included, such as adjusting the size, shape or relative position of the navigation points to other points to ensure that they can clearly express their semantic information on the vector map. In this way, the generated vector map can not only show the geometric features of the target driving test area, but also provide rich information about how to navigate and perform specific operations in this environment.
[0056] Step S520, based on the storage location information, generate an electronic fence corresponding to the storage location information, and obtain a vector map of the target driving test subject in the target driving test area.
[0057] An electronic fence is a virtual boundary system that defines a specific area or space by using geographic coordinates, which can be two-dimensional (longitude and latitude) or three-dimensional (including altitude information). The electronic fence can be circular, rectangular, polygonal, or any other shape, depending on the area to be defined. In this embodiment, the electronic fence generated based on the storage location information is to define the boundaries of the target driving test subject in the target driving test area. Exemplarily, this electronic fence contains all points and areas related to the driving test to ensure that the vehicle does not exceed the predetermined range when taking the driving test.
[0058] For example, in some embodiments, the geometric shape and size of the electronic fence are first determined based on the storage location information and the driving test requirements, and the boundary points of the electronic fence are marked on the map. Then, the semantic information related to the driving test is applied to the boundary points of the electronic fence to ensure that each part of the electronic fence corresponds to the specific driving test requirements. Finally, the generated electronic fence is integrated into the vector map of the target driving test area, making it part of the map, providing a reference for navigation and operation for the autonomous driving system.
[0059] In this embodiment, by moving the navigation points to the positions specified by the storage location information according to their respective semantic information, it is ensured that each point on the vector map accurately reflects its function and importance in the actual driving test environment, which not only improves the accuracy and practicality of the vector map, but also enhances the understanding of the driving test environment. By generating an electronic fence corresponding to the storage location information, the boundaries of the target driving test subject in the target driving test area are defined, ensuring that the vehicle does not exceed the predetermined range during the driving test, thereby enhancing the standardization and safety of the test. At the same time, the electronic fence also provides the necessary spatial restrictions for the autonomous driving system, helping to carry out effective path planning and avoid potential dangerous areas.
[0060] Figure 6 The figure shows a schematic diagram of a process for generating an electronic fence corresponding to storage location information provided by an embodiment of the present application. Figure 5 Based on the embodiment shown, Figure 6 The embodiment shown is described below in detail. Figure 6 The embodiment shown and Figure 5 The differences and similarities of the illustrated embodiments are not described in detail.
[0061] like Figure 6 As shown, in this embodiment, based on the storage location information, an electronic fence corresponding to the storage location information is generated, including the following steps.
[0062] Step S610: Generate a virtual boundary based on the boundary information in the storage location information.
[0063] In the location information, the boundary information defines the outer limits of the driving test area, including the latitude and longitude coordinates, shape, size, etc. of the driving test area. Using the extracted boundary information, a virtual boundary is created to define the geographic scope of the driving test area. In this process, geographic coordinates are used to determine the exact location of the virtual boundary. These coordinates can be two-dimensional or three-dimensional, depending on the complexity of the area to be defined. Based on the boundary information, the system will define the shape of the virtual boundary.
[0064] Step S620: Add trigger points in the virtual boundary to generate an electronic fence corresponding to the storage location information.
[0065] Trigger points are specific geographic locations preset in the electronic fence, which are used to monitor and respond to specific events or behaviors. In other words, trigger points are used to define key locations of the virtual boundary and trigger specific events.
[0066] Specifically, the trigger points are pre-defined points within the electronic fence, such as the entrance and exit of the driving test area, the center point of a specific operation area, etc. When a vehicle or other object enters or leaves these points, a preset event or action can be triggered. For example, when a vehicle enters a certain point, the vehicle's position will be recorded, or a prompt sound of the navigation system will be triggered.
[0067] For example, in some embodiments, firstly, all relevant boundary information and key point information are extracted from the storage location information, including the role of the geographical coordinates of the point in the driving test and the specific events that need to be triggered. Then, using this information, the trigger point is set at the appropriate position of the virtual boundary. For example, Figure 7 Shown is a schematic diagram of an electronic fence including trigger points provided by an embodiment of the present application. Next, corresponding logic and rules are configured for each trigger point, which define the events that should occur when a vehicle or other object interacts with the point. For example, when a vehicle enters a specific point, a navigation prompt or safety warning is triggered. In addition, it is necessary to ensure that these trigger points can be accurately identified and responded to in the system. After the trigger points are set, the electronic fence needs to be tested to ensure that all points work as expected and that the entire system can operate stably. Ultimately, the generated electronic fence not only defines the geographical scope of the driving test area, but also provides a dynamic and interactive monitoring and management method through trigger points.
[0068] Figure 8 The figure shows a schematic diagram of a process for generating a point cloud map of a target driving test area provided by an embodiment of the present application. Figure 1 Based on the embodiment shown, Figure 8 The embodiment shown is described below in detail. Figure 8 The embodiment shown and Figure 1 The differences and similarities of the illustrated embodiments are not described in detail.
[0069] like Figure 8 As shown, in this embodiment, the target sensor includes a laser radar sensor and an inertial measurement unit, the point cloud data is collected by the laser radar sensor, and the inertial data is collected by the inertial measurement unit. Based on the point cloud data and the inertial data, a point cloud map of the target driving test area is generated, including the following steps.
[0070] Step S810: synchronizing the point cloud data and the inertial data.
[0071] For example, in some embodiments, the data collected by the lidar sensor and the inertial measurement unit are first timestamped to ensure that each data point has an accurate time stamp, so that the point cloud data can be accurately matched with the corresponding inertial data. In addition, since the data acquisition frequency of different sensors may be different, data interpolation or resampling is required to ensure the consistency of the data in time. Through the above synchronization process, the point cloud data and the inertial data can be accurately aligned.
[0072] Step S820, in the configuration file in the target drive package, obtain the external parameter data of the lidar sensor and the inertial measurement unit, and estimate the initial motion trajectory of the target object based on the external parameter data and the synchronized point cloud data and inertial data.
[0073] Exemplarily, the relative position and orientation information between the lidar sensor and the IMU, i.e., the external parameter data, is first read from the configuration file to describe the spatial relationship between the above sensors in the vehicle coordinate system. Exemplarily, the external parameter data includes the transformation matrix or transformation parameters between the lidar sensor and the inertial measurement unit, such as rotation and translation vectors.
[0074] It can be understood that point cloud data provides accurate three-dimensional spatial information of the vehicle's surroundings, while inertial data provides the vehicle's own motion state, including speed, acceleration, and attitude changes. By combining external parameter data with these data, the position and motion state of the target object (such as a vehicle) in space can be estimated more accurately.
[0075] Step S830, based on the initial motion trajectory of the target object, align the point cloud data to the standard coordinate system, and accumulate to form a point cloud map of the target driving test area.
[0076] The initial motion trajectory provides the relative position and direction change of the target object in space. The registration process begins by combining the point cloud data collected in real time with the known motion trajectory. By calculating the position of the point cloud data relative to the standard coordinate system, the exact spatial position of each point in the target driving test area can be determined. In addition, this process requires the application of a transformation matrix to transform the point cloud data from the sensor coordinate system to the vehicle coordinate system, and then from the vehicle coordinate system to the standard coordinate system.
[0077] As the vehicle moves, the continuously collected point cloud data is constantly aligned to the standard coordinate system, thereby gradually building a three-dimensional map of the target driving test area. During this accumulation process, the newly collected point cloud data needs to be processed and fused in real time to ensure the continuity and consistency of the map. In addition, the accumulated point cloud map not only provides an accurate environmental model for the autonomous driving vehicle, but is also crucial for tasks such as path planning, obstacle avoidance, and navigation during the driving test. By aligning and accumulating point cloud data, the autonomous driving system can better understand and predict the movement of the vehicle in the driving test area, thereby improving driving safety and efficiency.
[0078] In this embodiment, by synchronizing point cloud data and inertial data, the temporal consistency of data from different sensors is ensured, providing an accurate and synchronized data source for subsequent processing. By using the external reference data in the configuration file, combined with the synchronized data, the initial motion trajectory of the target object is estimated, and the position and motion state of the vehicle can be accurately estimated, providing a basis for subsequent map construction and navigation. Based on the estimated initial motion trajectory, the point cloud data is aligned to the standard coordinate system, and a continuous point cloud map of the target driving test area is formed by accumulating these data, providing a dynamic environmental perception capability for the autonomous driving vehicle, and improving the navigation accuracy and safety of the autonomous driving vehicle. Through the continuously updated point cloud map, the autonomous driving system can better simulate the real driving test environment and provide a more realistic and challenging driving test experience for the driver or the autonomous driving vehicle.
[0079] Fig. 9 FIG. 1 is a flow chart of a method for constructing a vector map provided by another embodiment of the present application. Figure 1 Based on the embodiment shown, Fig. 9 The embodiment shown is described below in detail. Fig. 9 The embodiment shown and Figure 1 The differences and similarities of the illustrated embodiments are not described in detail.
[0080] like Fig. 9 As shown, in this embodiment, after generating a point cloud map of the target driving test area based on point cloud data and inertial data, the following steps are also included.
[0081] Step S910: Obtain a first parameter value corresponding to a storage parameter in a configuration file in a target driver package.
[0082] Storage parameters are a series of settings that control how point cloud maps are stored or managed, such as the map's storage format, compression level, storage location, etc. The first parameter value is a key value in these storage parameters, which determines whether the map needs to be stored or triggers certain specific storage behaviors. For example, if the first parameter value is a Boolean value, it indicates whether the system should save the generated point cloud map to persistent storage.
[0083] Step S920: If the first parameter value is true, obtain a second parameter value corresponding to the path parameter in the configuration file.
[0084] In this embodiment, if the first parameter value is true (for example, if it is a Boolean flag indicating that the storage operation should be performed), a second parameter value corresponding to the path parameter is further obtained from the configuration file. Specifically, the path parameter defines a specific path or directory where the point cloud map should be stored, and the second parameter value is a specific representation of the path, such as a file system path or a network storage location.
[0085] Step S930: store the point cloud map in the storage path indicated by the second parameter value.
[0086] In this embodiment, the point cloud map is stored in the storage path indicated by the second parameter value, ensuring that the point cloud map is saved to an appropriate location according to the predefined path parameters, thereby facilitating subsequent access, analysis or use. Exemplarily, the storage process includes operations such as writing the point cloud data to a file, a database, or transmitting it to a remote server via a network.
[0087] In this embodiment, by obtaining the first parameter value of the storage parameter from the configuration file, it is possible to determine whether a storage operation needs to be performed, thereby avoiding unnecessary storage operations and improving efficiency. Based on the truth or falsity of the first parameter value, the second parameter value of the path parameter is further obtained from the configuration file, ensuring the flexibility and configurability of the storage path, allowing the user or system administrator to specify different storage locations as needed. Finally, storing the point cloud map in the path specified by the second parameter value not only ensures the persistence of the data, but also facilitates the subsequent use and analysis of the data.
[0088] In some embodiments of the present application, before acquiring the point cloud data and inertial data of the target driving test area, it also includes: testing the target sensor. Fig.10 FIG. 1 is a schematic diagram of a process for testing a target sensor according to an embodiment of the present application. Figure 1 Based on the embodiment shown, Fig.10The embodiment shown is described below in detail. Fig.10 The embodiment shown and Figure 1 The differences and similarities of the illustrated embodiments are not described in detail.
[0089] like Fig.10 As shown, in this embodiment, testing the target sensor includes the following steps.
[0090] Step S1010: Determine physical connection with the target sensor through the target port device file.
[0091] A port device file is a file used to represent and access a specific hardware device in a computer operating system. In Unix-like systems, including Linux, device files are usually located in the / dev directory. Each device file corresponds to a hardware device in the system, allowing users or programs to interact with the hardware device through a standard file operation interface.
[0092] In this embodiment, the target port device file specifically refers to the device file corresponding to the target sensor. Through this file, the test system can identify the sensor and establish communication with it. For example, if the target sensor is a USB device connected to the computer, then the target port device file is a device file representing the USB device, such as / dev / ttyUSB0. After determining the physical connection, you can interact with the target sensor by reading or writing the target port device file, such as sending instructions, configuring parameters, or receiving data.
[0093] Step S1020: setting communication parameters that match the configuration parameters of the target sensor.
[0094] The communication parameters matching the configuration parameters of the target sensor refer to a series of communication protocols and parameters that need to be set in order to ensure that the target sensor can correctly exchange data with the test system or control unit. These parameters must be consistent with the technical specifications and requirements of the target sensor. Exemplarily, the communication parameters include: Baud Rate: refers to the rate of data transmission, that is, the number of bits transmitted per unit time; Data Bits: refers to the number of bits sent in each transmission; Stop Bits: The stop bit is used to mark the end of the data packet, which can be 1 or 2 bits; Parity Bit: It is an error detection mechanism, which can be none, odd or even; Interface Type: For example, the sensor communicates through interfaces such as SPI, I2C, UART, USB, etc., and each interface has its own specific communication protocol; Address: In some communication protocols, such as I2C, each device has a unique address, which must be set correctly to ensure that data can be sent to the correct device; Clock Rate: For some interfaces, such as SPI, the clock rate needs to be set to synchronize data transmission.
[0095] Step S1030: receiving a data stream sent by a target sensor, and testing the target sensor based on the data stream.
[0096] Data stream refers to a continuous sequence of data sent by a target sensor in real time or at regular intervals, including various environmental parameters or status information measured by the target sensor. For example, the received data stream is parsed and converted into a format that can be further analyzed. By real-time monitoring and recording of the data stream, the output of the target sensor can be monitored to see if it is stable and compared with the expected performance parameters.
[0097] In addition, the test system will perform quality checks on the data stream, including data integrity, accuracy, and whether it is within a preset range. Through statistical analysis, such as calculating the average, standard deviation, maximum and minimum values, the accuracy and stability of the target sensor are evaluated. In addition, by comparing the readings in the data stream with known standards or reference values, the accuracy of the sensor can be evaluated.
[0098] During the test, if anomalies or deviations are found in the data stream, the test system triggers a fault diagnosis procedure to identify possible causes of the problem, such as sensor failure, signal interference or other environmental factors. Through this analysis, the target sensor can be calibrated or repaired to restore its performance.
[0099] In this embodiment, the physical connection with the target sensor is determined through the target port device file, ensuring that the target sensor can be identified and a communication link is established. Next, communication parameters that match the configuration parameters of the target sensor are set to ensure that data can be transmitted between the target sensor and the test system in the correct format and rate. Finally, based on the test results of the data stream, the target sensor can be monitored in real time, performance evaluated, and fault diagnosed to ensure that the target sensor can provide high-quality data output under various working conditions.
[0100] The above describes in detail the implementation process of the vector map construction method based on driving test subjects. Next, we will briefly analyze the vector map generation process with specific examples.
[0101] First, connect and test the sensors to ensure that the LiDAR and IMU are correctly installed and data transmission is normal. Then, use the tool cutecom to test the sensor data by entering the corresponding command in the terminal, selecting the correct device and baud rate, and observing the data flow of the above sensors. Next, start the IMU and LiDAR sensors respectively, and use the relevant commands to check the data flow of the IMU and LiDAR sensors to ensure that the data is received correctly.
[0102] Then, record the data, using the rosbag tool to record the point cloud data collected by the lidar and the inertial data of the IMU. During the recording process, the above sensor devices need to be fully moved in the environment where the map is required to be built to capture complete environmental information. After the recording is completed, use the rosbag play command to play the recorded data set so that the mapping process can be observed in rviz.
[0103] After the data is prepared, use the LIO-mapping method to process the recorded data and generate a point cloud map in pcd format. Set the correct parameters in the LIO-mapping configuration file, including the external parameter data of the IMU relative to the lidar sensor, and the topic name. By modifying the configuration file, enable the automatic saving function of the point cloud data and specify the save path.
[0104] After the point cloud map is generated, it needs to be further processed in Unity to generate a vector map. First, create a new project in Unity and use CreateMap to draw a vector map of the driving school scene. According to the actual point cloud map of the driving school, drag the pre-defined storage locations to the corresponding navigation points to complete the drawing of the point map. After the drawing is completed, export the storage location file, which will contain the vector map information of the driving school.
[0105] Finally, to generate an electronic fence, continue to use the electronic fence function in CreateMap. Add the trigger points of the fence in Unity, generate a set of fence points according to the requirements, and export the electronic fence file. In this way, the entire process from sensor connection to electronic fence generation is completed, providing the necessary map support for the application of driverless cars in driving school scenarios.
[0106] Combination of the above Figures 1 to 10 , describes in detail the embodiment of the vector map construction method based on the driving test subjects of the present application, and the following is combined with Fig.11 , the embodiment of the vector map construction device based on the driving test subjects of the present application is described in detail. It should be understood that the description of the embodiment of the vector map construction method based on the driving test subjects corresponds to the description of the embodiment of the vector map construction device based on the driving test subjects. Therefore, the part not described in detail can refer to the previous method embodiment.
[0107] Fig.11 FIG. 1 is a schematic diagram of a structure of a vector map construction device based on driving test subjects provided by an embodiment of the present application. Fig.11 As shown, the vector map construction device 110 based on driving test subjects provided in the embodiment of the present application includes:
[0108] An acquisition module 1110 is used to acquire point cloud data and inertial data of a target driving test area collected by a target sensor;
[0109] A first generating module 1120, for generating a point cloud map of a target driving test area based on the point cloud data and the inertial data;
[0110] A drawing module 1130, for drawing vector elements based on environmental features included in the point cloud map;
[0111] The second generating module 1140 is used to generate a vector map of the target driving test area based on the vector elements.
[0112] In one embodiment of the present application, the environmental features include the storage location information corresponding to the target driving test subject. The drawing module 1130 is also used to draw a plurality of navigation points corresponding to the storage location information based on the storage location information corresponding to the target driving test subject, each navigation point containing semantic information for marking the geographic coordinates of the operation position in the storage location information.
[0113] In one embodiment of the present application, the second generation module 1140 is also used to move multiple navigation points to positions corresponding to the storage location information based on the semantic information corresponding to each of the multiple navigation points; based on the storage location information, generate an electronic fence corresponding to the storage location information, and obtain a vector map of the target driving test subject in the target driving test area.
[0114] In one embodiment of the present application, the second generation module 1140 generates a virtual boundary based on the boundary information in the storage location information; in the virtual boundary, trigger points are added to generate an electronic fence corresponding to the storage location information, wherein the trigger points are used to define key positions of the virtual boundary and trigger specific events.
[0115] In one embodiment of the present application, the target sensor includes a laser radar sensor and an inertial measurement unit, the point cloud data is collected by the laser radar sensor, and the inertial data is collected by the inertial measurement unit. The first generation module 1120 is also used to synchronize the point cloud data and the inertial data; in the configuration file in the target driving package, the external parameter data of the laser radar sensor and the inertial measurement unit are obtained, and based on the external parameter data and the synchronized point cloud data and inertial data, the initial motion trajectory of the target object is estimated; based on the initial motion trajectory of the target object, the point cloud data is aligned to the standard coordinate system, and accumulated to form a point cloud map of the target driving test area.
[0116] In one embodiment of the present application, the first generation module 1120 is also used to obtain a first parameter value corresponding to a storage parameter in a configuration file in a target driver package; if the first parameter value is true, obtain a second parameter value corresponding to a path parameter in the configuration file; and store the point cloud map in a storage path indicated by the second parameter value.
[0117] In one embodiment of the present application, the acquisition module 1110 is also used to determine the physical connection with the target sensor through the target port device file; set communication parameters that match the configuration parameters of the target sensor; receive the data stream sent by the target sensor, and test the target sensor based on the data stream.
[0118] Below, reference Fig.12 To describe an electronic device according to an embodiment of the present application. Fig.12 Shown is a schematic structural diagram of an electronic device provided by an exemplary embodiment of the present application.
[0119] like Fig.12 As shown, the electronic device 120 includes one or more processors 1201 and a memory 1202 .
[0120] The processor 1201 may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device 120 to perform desired functions.
[0121] The memory 1202 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory (cache), etc. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 1201 may run the program instructions to implement the vector map construction method based on driving test subjects and / or other desired functions of the various embodiments of the present application described above. Various contents such as point cloud data, inertial data, point cloud maps, vector elements, vector maps, etc. may also be stored in the computer-readable storage medium.
[0122] In one example, the electronic device 120 may further include: an input device 1203 and an output device 1204 , and these components are interconnected via a bus system and / or other forms of connection mechanisms (not shown).
[0123] The input device 1203 may include, for example, a keyboard, a mouse, etc.
[0124] The output device 1204 can output various information to the outside, including point cloud data, inertial data, point cloud map, vector elements, vector map, etc. The output device 1204 can include, for example, a display, a speaker, a printer, a communication network and its connected remote output device, etc.
[0125] Of course, to simplify, Fig.12 Only some of the components related to the present application in the electronic device 120 are shown, and components such as a bus, an input / output interface, etc. are omitted. In addition, according to specific application situations, the electronic device 120 may also include any other appropriate components.
[0126] In addition to the above-mentioned methods and devices, an embodiment of the present application may also be a computer program product, which includes computer program instructions, which, when executed by a processor, enable the processor to execute the steps of the vector map construction method based on driving test subjects according to various embodiments of the present application described above in this specification.
[0127] The computer program product may be written in any combination of one or more programming languages to write program codes for performing the operations of the embodiments of the present application, including object-oriented programming languages, such as Java, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as an independent software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.
[0128] In addition, an embodiment of the present application may also be a computer-readable storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, the processor executes the steps of the vector map construction method based on driving test subjects according to various embodiments of the present application described above in this specification.
[0129] The computer readable storage medium can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can include, for example, but is not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.
[0130] The basic principles of the present application are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present application are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present application. In addition, the specific details disclosed above are only for the purpose of illustration and ease of understanding, not for limitation, and the above details do not limit the present application to being implemented by adopting the above specific details.
[0131] The block diagrams of the devices, apparatuses, equipment, and systems involved in this application are only illustrative examples and are not intended to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagram. As will be appreciated by those skilled in the art, these devices, apparatuses, equipment, and systems can be connected, arranged, and configured in any manner. Words such as "including", "comprising", "having", etc. are open words, referring to "including but not limited to", and can be used interchangeably with them. The words "or" and "and" used here refer to the words "and / or" and can be used interchangeably with them, unless the context clearly indicates otherwise. The words "such as" used here refer to the phrase "such as but not limited to", and can be used interchangeably with them.
[0132] It should also be noted that in the apparatus, device and method of the present application, each component or each step can be decomposed and / or recombined. Such decomposition and / or recombination should be regarded as equivalent solutions of the present application.
[0133] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present application. Therefore, the present application is not intended to be limited to the aspects shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.
[0134] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present application to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.
Claims
1. A method for constructing a vector map based on driving test subjects, characterized in that: Applied to a host control computer, the method comprises: Obtain point cloud data and inertial data of the target driving test area collected by the target sensor; Based on the point cloud data and the inertial data, generating a point cloud map of the target driving test area; Drawing vector elements based on environmental features contained in the point cloud map; Based on the vector elements, a vector map of the target driving test area is generated.
2. The method according to claim 1, characterized in that The environmental characteristics include storage location information corresponding to the target driving test subject; Drawing vector elements based on the environmental features contained in the point cloud map includes: Based on the storage location information corresponding to the target driving test subject, multiple navigation points corresponding to the storage location information are drawn, and each of the navigation points contains semantic information for marking the geographic coordinates of the operation position in the storage location information.
3. The method according to claim 2, characterized in that The step of generating a vector map of the target driving test area based on the vector elements includes: Based on the semantic information corresponding to each of the multiple navigation points, the multiple navigation points are moved to positions corresponding to the storage location information; Based on the storage location information, an electronic fence corresponding to the storage location information is generated to obtain a vector map of the target driving test subject in the target driving test area.
4. The method according to claim 3, characterized in that The step of generating an electronic fence corresponding to the storage location information based on the storage location information includes: Generate a virtual boundary based on the boundary information in the storage location information; In the virtual boundary, trigger points are added to generate an electronic fence corresponding to the storage location information, wherein the trigger points are used to define key positions of the virtual boundary and trigger specific events.
5. The method according to any one of claims 1 to 4, characterized in that: The target sensor includes a laser radar sensor and an inertial measurement unit, the point cloud data is collected by the laser radar sensor, and the inertial data is collected by the inertial measurement unit; wherein, generating a point cloud map of the target driving test area based on the point cloud data and the inertial data includes: Synchronizing the point cloud data and the inertial data; In a configuration file in a target drive package, external parameter data of the laser radar sensor and the inertial measurement unit are obtained, and an initial motion trajectory of a target object is estimated based on the external parameter data and the synchronized point cloud data and the inertial data; Based on the initial motion trajectory of the target object, the point cloud data is aligned to a standard coordinate system and accumulated to form a point cloud map of the target driving test area.
6. The method according to any one of claims 1 to 4, characterized in that: After generating a point cloud map of the target driving test area based on the point cloud data and the inertial data, the method further includes: In a configuration file in a target driver package, obtain a first parameter value corresponding to a storage parameter; If the first parameter value is true, obtaining a second parameter value corresponding to the path parameter in the configuration file; The point cloud map is stored in the storage path indicated by the second parameter value.
7. The method according to any one of claims 1 to 4, characterized in that: Before acquiring the point cloud data and inertial data of the target driving test area, the method further includes: testing the target sensor; Wherein, the testing of the target sensor includes: Determine physical connection with the target sensor through the target port device file; Setting communication parameters that match configuration parameters of the target sensor; A data stream sent by the target sensor is received, and based on the data stream, the target sensor is tested.
8. A vector map construction device based on driving test subjects, characterized in that: Applicable to the host computer, including: An acquisition module, used to acquire point cloud data and inertial data of a target driving test area collected by a target sensor; A first generating module, configured to generate a point cloud map of the target driving test area based on the point cloud data and the inertial data; A drawing module, used for drawing vector elements based on environmental features contained in the point cloud map; The second generating module is used to generate a vector map of the target driving test area based on the vector elements.
9. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and the computer program is used to execute the vector map construction method based on driving test subjects as described in any one of claims 1 to 7 above.
10. An electronic device, characterized in that: include: processor; a memory for storing instructions executable by the processor; The processor is used to execute the vector map construction method based on driving test subjects as described in any one of claims 1 to 7 above.
Citation Information
Patent Citations
Area judging method based on virtual sensor
CN102998689A
Intelligent robot system for motor vehicle driver training
CN103971558A
Driver training platform
CN104504514A
Intelligent driving training system and method based on vehicle safety control
CN106781824A
Coach assessment method and device in driving training
CN107331247A
