Driving navigation method, device, computer equipment, storage medium and computer program product

Abstract

The present application relates to a driving navigation method, device, computer equipment, storage medium and computer program product, which are applied to the field of maps. The method comprises: obtaining driving data collected by a target vehicle on a driving road; identifying the driving data to obtain lane-level data of the driving road; the lane-level data comprises lane line data, drivable lane information and vehicle information; determining the lane position information of the target vehicle based on the lane line data; determining the merging lane information of the target vehicle at the next intersection based on the lane position information, the vehicle information data of the driving road in the map, the drivable lane information and the vehicle information; navigating the target vehicle based on the lane position information and the merging lane information. The use of this method can effectively improve the accuracy of driving road navigation data.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a driving navigation method, device, computer equipment, storage medium and computer program product. Background Art

[0002] With the development of computer technology and Internet technology, the application of intelligent driving in different business scenarios has been implemented, making intelligent driving widely concerned by the public. Due to the emergence of various special road conditions, it is necessary to update all road data in the road network data in real time and build corresponding driving navigation data.

[0003] However, in the current driving navigation method, standard maps are usually used for driving navigation. Since the accuracy of standard maps has errors of more than meters and the positioning accuracy of the Global Positioning System (GPS) is low, when standard maps are used for driving navigation, the accuracy of the obtained road navigation data is easily low. Summary of the invention

[0004] Based on this, it is necessary to provide a driving navigation method, device, computer equipment, computer readable storage medium and computer program product that can effectively improve the accuracy of driving road navigation data in response to the above technical problems.

[0005] In the first aspect, the present application provides a driving navigation method. The method includes: acquiring driving data collected by a target vehicle on a driving road; identifying the driving data to obtain lane-level data of the driving road; the lane-level data includes lane line data, drivable lane information and vehicle information; determining the lane position information of the target vehicle based on the lane line data; determining the merging lane information of the target vehicle at the next intersection based on the lane position information, the vehicle information data of the driving road in the map, the drivable lane information and the vehicle information; navigating the target vehicle based on the lane position information and the merging lane information.

[0006] In the second aspect, the present application also provides a driving navigation device. The device includes: an acquisition module, which is used to acquire driving data collected by the target vehicle on the driving road; an identification module, which is used to identify the driving data and obtain lane-level data of the driving road; the lane-level data includes lane line data, drivable lane information and vehicle information; a determination module, which is used to determine the lane position information of the target vehicle based on the lane line data; according to the lane position information, the vehicle information data of the driving road in the map, the drivable lane information and the vehicle information, determine the merging lane information of the target vehicle at the next intersection; and a navigation module, which is used to navigate the target vehicle based on the lane position information and the merging lane information.

[0007] In a third aspect, the present application also provides a computer device. The computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program: obtaining driving data collected by the target vehicle on the driving road; identifying the driving data to obtain lane-level data of the driving road; the lane-level data includes lane line data, drivable lane information and vehicle information; determining the lane position information of the target vehicle based on the lane line data; determining the merging lane information of the target vehicle at the next intersection based on the lane position information, the vehicle information data of the driving road in the map, the drivable lane information and the vehicle information; navigating the target vehicle based on the lane position information and the merging lane information.

[0008] In a fourth aspect, the present application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the following steps are implemented: obtaining driving data collected by a target vehicle on a driving road; identifying the driving data to obtain lane-level data of the driving road; the lane-level data includes lane line data, drivable lane information, and vehicle information; determining the lane position information of the target vehicle based on the lane line data; determining the mergible lane information of the target vehicle at the next intersection based on the lane position information, the vehicle information data of the driving road in the map, the drivable lane information, and the vehicle information; navigating the target vehicle based on the lane position information and the mergible lane information.

[0009] In a fifth aspect, the present application also provides a computer program product. The computer program product includes a computer program, which implements the following steps when executed by a processor: obtaining driving data collected by a target vehicle on a driving road; identifying the driving data to obtain lane-level data of the driving road; the lane-level data includes lane line data, drivable lane information, and vehicle information; determining the lane position information of the target vehicle based on the lane line data; determining the merging lane information of the target vehicle at the next intersection based on the lane position information, the vehicle information data of the driving road in the map, the drivable lane information, and the vehicle information; navigating the target vehicle based on the lane position information and the merging lane information.

[0010] The above-mentioned driving navigation method, device, computer equipment, storage medium and computer program product obtain driving data collected by the target vehicle on the driving road, identify the driving data, and obtain lane-level data of the driving road; the lane-level data includes lane line data, drivable lane information and vehicle information; the lane position information of the target vehicle is determined based on the lane line data, and the mergible lane information of the target vehicle at the next intersection is determined according to the lane position information, vehicle signal data of the driving road in the map, drivable lane information and vehicle information, and the target vehicle is navigated based on the lane position information and the mergible lane information. Since the driving data of the target vehicle is collected when it is driving and the lane-level data is obtained by identifying the driving data, the lane position of the target vehicle can be determined based on the lane-level data, so that lane-level driving navigation can be performed according to the lane position of the target vehicle and the mergible lane information, thereby realizing lane-level guidance based on standard map data and effectively improving the accuracy of road navigation data. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 A diagram showing an application environment of a driving navigation method in an embodiment;

[0012] Figure 2 A schematic diagram of a flow chart of a driving navigation method in one embodiment;

[0013] Figure 3 is a schematic diagram of lane-level driving navigation in one embodiment;

[0014] Figure 4 A flowchart of a step of determining information about a mergible lane for a target vehicle at a next intersection according to lane position information, vehicle information data of a driving road in a map, drivable lane information, and vehicle information in one embodiment;

[0015] Figure 5 A schematic diagram of vehicle signal data in an embodiment;

[0016] Figure 6 A schematic diagram of the types of maneuvering points in one embodiment;

[0017] Figure 7 A schematic diagram of performing lane guidance according to the type of maneuvering point in one embodiment;

[0018] Figure 8 is a flowchart of lane-level guidance in one embodiment;

[0019] Fig. 9 A schematic diagram of navigating by using a path line navigation method in one embodiment;

[0020] Fig.10 A schematic diagram of navigating using a lane-level navigation method in one embodiment;

[0021] Fig.11 is a schematic diagram of a directed graph in one embodiment;

[0022] Fig.12 is a schematic diagram of a standard map or road network in one embodiment;

[0023] Fig.13 is a structural block diagram of a driving navigation device in one embodiment;

[0024] Fig.14 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0026] The driving navigation method provided in the embodiment of the present application can be applied to Figure 1In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other servers. The terminal 102 can obtain the vehicle signal data of the driving road in the map from the server 104, and the terminal 102 can also obtain the vehicle signal data of the driving road in the map from the local database. The terminal 102 can obtain the driving data collected by the target vehicle on the driving road, and the terminal 102 identifies the driving data to obtain the lane-level data of the driving road; the lane-level data includes lane line data, drivable lane information and vehicle information; the terminal 102 determines the lane position information of the target vehicle based on the lane line data, and the terminal 102 determines the merging lane information of the target vehicle at the next intersection according to the lane position information, the vehicle signal data of the driving road in the map, the drivable lane information and the vehicle information; the terminal 102 navigates the target vehicle based on the lane position information and the merging lane information.

[0027] The terminal 102 may be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, IoT devices and portable wearable devices. IoT devices may be smart speakers, smart TVs, smart air conditioners, smart voice interaction devices, smart home appliances, smart vehicle-mounted devices and other Internet of Vehicles devices. Portable wearable devices may be smart watches, smart bracelets, head-mounted devices and the like.

[0028] The server 104 can be implemented with an independent server or a server cluster composed of multiple servers. It can be understood that the server 104 provided in the embodiment of the present application can also be a service node in the blockchain system. The service nodes in the blockchain system form a peer-to-peer (P2P) network. The P2P protocol is an application layer protocol running on the Transmission Control Protocol (TCP) protocol.

[0029] Intelligent Traffic System (ITS), also known as Intelligent Transportation System, is an effective and comprehensive application of advanced science and technology (information technology, computer technology, data communication technology, sensor technology, electronic control technology, automatic control theory, operations research, artificial intelligence, etc.) to transportation, service control and vehicle manufacturing, strengthening the connection between vehicles, roads and users, thus forming a comprehensive transportation system that ensures safety, improves efficiency, improves the environment and saves energy. Or;

[0030] Intelligent Vehicle Infrastructure Cooperative Systems (IVICS), referred to as IVICS, is a development direction of Intelligent Transportation Systems (ITS). IVICS uses advanced wireless communications and new generation Internet technologies to implement all-round dynamic real-time information interaction between vehicles and roads, and conducts active vehicle safety control and road cooperative management based on the collection and integration of dynamic traffic information in all time and space, fully realizing the effective coordination of people, vehicles and roads, ensuring traffic safety, and improving traffic efficiency, thus forming a safe, efficient and environmentally friendly road traffic system.

[0031] Computer Vision Technology (CV) Computer vision is a science that studies how to make machines "see". To put it more specifically, it refers to machine vision that uses cameras and computers to replace human eyes to identify, record and measure targets, and further performs graphic processing to make computer processing into images that are more suitable for human eye observation or transmission to instruments for detection. As a scientific discipline, computer vision studies related theories and technologies, and attempts to establish an artificial intelligence system that can obtain information from images or multi-dimensional data. Computer vision technology usually includes image processing, image recognition, image semantic understanding, image retrieval, OCR, video processing, video semantic understanding, video content / behavior recognition, three-dimensional object reconstruction, 3D technology, virtual reality, augmented reality, simultaneous positioning and map construction, automatic driving, smart transportation and other technologies, as well as common biometric recognition technologies such as face recognition and fingerprint recognition.

[0032] With the research and advancement of artificial intelligence technology, artificial intelligence technology has been studied and applied in many fields, such as common smart homes, smart wearable devices, virtual assistants, smart speakers, smart marketing, unmanned driving, automatic driving, drones, robots, smart medical care, smart customer service, Internet of Vehicles, automatic driving, smart transportation, etc. I believe that with the development of technology, artificial intelligence technology will be applied in more fields and play an increasingly important role.

[0033] Machine Learning (ML) is a multi-disciplinary subject that involves probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory and other disciplines. It specializes in studying how computers simulate or implement human learning behavior to acquire new knowledge or skills and reorganize existing knowledge structures to continuously improve their performance. Machine learning is the core of artificial intelligence and the fundamental way to make computers intelligent. Its applications are spread across all areas of artificial intelligence. Machine learning and deep learning usually include artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and self-learning.

[0034] Deep learning is a branch of machine learning. It is an algorithm that attempts to use multiple processing layers containing complex structures or composed of multiple nonlinear transformations to perform high-level abstraction on data. Deep learning is an algorithm in machine learning that is based on learning to represent data. So far, there are several deep learning frameworks, such as convolutional neural networks, deep belief networks, and recursive neural networks, which have been applied in computer vision, speech recognition, natural language processing, audio recognition, and bioinformatics, and have achieved excellent results.

[0035] In one embodiment, Figure 2 As shown, a driving navigation method is provided, which is applied to Figure 1 The terminal in is used as an example to illustrate, including the following steps:

[0036] Step 202, obtaining driving data collected by the target vehicle on the driving road.

[0037] The target vehicle refers to a vehicle traveling on the road. For example, vehicles 1, 2, and 3 traveling on road A can all be target vehicles. The target vehicle in the embodiment of the present application can be the vehicle traveling on the current road, and the vehicle is equipped with a vehicle terminal, a camera, and a sensor.

[0038] The driving road refers to the road on which the target vehicle is traveling. For example, if vehicle 1 is currently traveling on road A, then the driving road is road A. The vehicle-mounted terminal installed in vehicle 1 can obtain the driving data collected while traveling on road A in real time.

[0039] Driving data refers to the data collected by the target vehicle during driving. For example, if a camera is installed in the target vehicle, the vehicle terminal in the target vehicle can collect driving data of the target vehicle during driving by calling the camera. Driving data can include various types of data, for example, driving data is real-time collected driving picture data, driving video data, driving audio data, etc.

[0040] Specifically, if a terminal device is installed in the target vehicle, or the target vehicle has an association relationship with a terminal device, the terminal can obtain the driving data collected by the target vehicle on the driving road. The driving road can be the road that the target vehicle is currently driving on, or the road that the target vehicle was driving on at the last moment. The terminal device installed in the vehicle includes but is not limited to a vehicle-mounted terminal device, a mobile terminal device, and a portable terminal device.

[0041] For example, the target vehicle A is associated with the terminal A, then the terminal A can obtain the driving data A collected by the target vehicle A on the driving road A at the current moment. Assuming that the driving data A is the image information taken by the camera installed in the target vehicle, the image information may include the lane line information, sign pole information, lane landmark information (left turn or right turn), zebra crossing information, red light information and vehicle information within the field of view in front of the target vehicle in the current moment of the driving road A.

[0042] Step 204, identifying the driving data to obtain lane-level data of the driving road; the lane-level data includes lane line data, drivable lane information and vehicle information.

[0043] Among them, lane-level data refers to data containing lane information or related to lane information. For example, lane-level data may include lane line data, drivable lane information, and information about vehicles traveling in each lane.

[0044] Lane line data refers to lane line information marked on the road. Lane line data can contain multiple types of lane line data, for example, lane line data can include yellow solid lines, white solid lines, white dashed solid lines, yellow dashed solid lines, guide lines, and yellow grid lines. Among them, yellow solid lines include double yellow lines and single yellow lines. Double yellow lines are drawn on the road section to separate traffic flows traveling in opposite directions. Single yellow lines are drawn on the road section to separate traffic flows traveling in opposite directions or as special bus and school bus stop markings; when single yellow lines are drawn on the roadside or on the curb, it means that roadside parking is prohibited.

[0045] Solid white line: When drawn in the middle of a road section, it is used to separate motor vehicles and non-motor vehicles traveling in the same direction, or to indicate the edge of the road; when drawn at an intersection, it is used as a guide lane line or stop line, or to guide the driving trajectory of vehicles.

[0046] Guide lines are mainly one or several white V-shaped lines or diagonal line areas set according to the terrain of the intersection. They are used at intersections that are too wide, irregular, or have complex driving conditions, at the ramp entrances of grade-separated intersections, or other special locations. They indicate that vehicles must follow the prescribed routes and must not drive over or across the lines.

[0047] White dashed and solid lines are used to indicate the edge of the roadway where vehicles can temporarily cross the line. Vehicles are allowed to temporarily cross on the dashed line side, and vehicles are prohibited from crossing on the solid line side. Yellow dashed and solid lines: When drawn in the middle of a road section, they are used to separate traffic flows traveling in opposite directions. Vehicles are prohibited from crossing the line on the solid line side, and vehicles are allowed to temporarily cross the line on the dashed line side.

[0048] The drivable lane information refers to the drivable area on the road where vehicles are allowed to drive. For example, the drivable lane information may include the drivable area divided by lane lines, and the non-drivable area may include an obstacle area, a construction area, etc.

[0049] Vehicle information refers to vehicle information within the field of view in front of a target vehicle on the road. For example, vehicle information may include vehicle position information and vehicle quantity information, i.e., lane position information of each vehicle on the road and vehicle quantity information within the field of view in front of the target vehicle.

[0050] Specifically, after the terminal obtains the driving data collected by the target vehicle on the driving road, the terminal can use the trained machine learning model to identify the driving data and obtain lane-level data of the driving road, which may include lane line data, drivable lane information and vehicle information. The machine learning model can be a machine learning model trained by classification using sample data in advance, and the machine learning model trained by classification is deployed in the driving navigation system to realize the function of automatically and quickly identifying driving data.

[0051] For example, after the terminal obtains the driving data A collected by the target vehicle A on the driving road A, the terminal can use the trained classification model to identify the driving data A to obtain the lane-level data A of the driving road A. The lane-level data A may include 5 lane line information, 2 drivable lane information and 10 vehicle information within the front field of view in the driving direction of the target vehicle.

[0052] Step 206: Determine lane position information of the target vehicle based on the lane line data.

[0053] The lane position information refers to the lane position of the target vehicle itself. For example, the lane position information may be the first lane where the target vehicle is traveling on road A. The first lane is the lane position information of the target vehicle itself.

[0054] Specifically, the terminal uses the trained machine learning model to identify the driving data, and after obtaining the lane-level data of the driving road, the terminal can determine the lane position information of the target vehicle based on the lane line data in the lane-level data. The lane position information can be lane information identified in different ways, for example, the lane position information can be the first lane or the third lane, and the lane position information can also be the left lane or the right lane.

[0055] For example, the terminal can identify the edge of the driving road by using the trained machine learning model to identify the driving data. Furthermore, the terminal can determine the lane position information of the target vehicle based on the road edge and lane line data. Figure 3 As shown, it is a schematic diagram of lane-level driving navigation. The vehicle is in the first lane from the left, that is, the lane position information of the vehicle is: lane 1.

[0056] Step 208, determining the merging lane information of the target vehicle at the next intersection based on the lane position information, the vehicle information data of the driving road in the map, the drivable lane information and the vehicle information.

[0057] The vehicle signal data refers to identification data used to mark the lane type. For example, the vehicle signal data may be a vehicle signal icon displayed in the navigation interface, which identifies the type of the next intersection or maneuvering point.

[0058] The mergible lane information refers to lanes that can be driven and merged. For example, there are some lanes on the actual driving road that allow vehicles to drive, but if the lane line is a solid line, the vehicle cannot merge into another lane, and it is impossible to drive from the current lane.

[0059] Specifically, after the terminal determines the lane position information of the target vehicle based on the lane line data in the lane-level data, the terminal can determine the merging lane information of the target vehicle at the next intersection based on the lane position information of the vehicle, the vehicle signal data of the driving road in the map, the drivable lane information and the vehicle information. That is, the terminal can determine the drivable lane information of the target vehicle at the next intersection based on the lane position information of the vehicle and the vehicle signal data of the driving road in the map. Further, the terminal determines the merging lane information of the target vehicle at the next intersection based on the identified drivable lane information and vehicle information in the driving road and based on different driving strategies. Among them, the driving strategy can be the shortest time strategy, the shortest distance strategy, etc. It can be understood that the driving strategy in the embodiment of the present application includes but is not limited to the above-mentioned driving strategy, and can also be other driving strategies set by user.

[0060] For example, assuming that the driving strategy set in the driving navigation system of the target vehicle is the shortest distance strategy, the terminal determines that the lane position information of the target vehicle A is the second lane based on the lane line data in the lane-level data. The terminal can determine that the target vehicle's drivable lane information at the next intersection is the left lane based on the lane position information of the vehicle itself as the second lane and the vehicle information data of the driving road in the map as the two-way road entering the main road (H-shaped intersection). Further, the terminal determines that the target vehicle's drivable lane information at the next intersection is the left lane based on the identified drivable lane information in the driving road and the vehicle information in each lane based on the shortest distance strategy. If the driving strategy set in the driving navigation system of the target vehicle is the shortest time strategy, and there are a large number of vehicles in the left lane at the current moment, and there is congestion, the terminal determines that the target vehicle's drivable lane information at the next intersection may not be the left lane based on the shortest time strategy, that is, the terminal determines that the target vehicle's drivable lane information at the next intersection is other non-congested lanes.

[0061] Step 210 , navigating the target vehicle based on the lane position information and the mergible lane information.

[0062] Among them, navigation refers to navigating the driving route of the target vehicle. For example, when performing driving navigation for the target vehicle in this application, the terminal can use path line navigation for navigation, that is, the terminal displays the corresponding path-level driving guidance lines in the navigation interface according to the path-level guidance signs; the terminal can also use lane line navigation for navigation, that is, the terminal displays the corresponding lane-level driving guidance lines in the navigation interface according to the lane-level guidance signs.

[0063] Specifically, after the terminal determines the mergible lane information of the target vehicle at the next intersection based on the lane position information of the own vehicle, the vehicle signal data of the driving road in the map, the drivable lane information and the vehicle information, the terminal can navigate the target vehicle based on the lane position information and the mergible lane information of the own vehicle. Among them, the navigation method may include a lane-level navigation method and a path-level navigation method. The path-level navigation method refers to guidance based on the path line, that is, the terminal draws and paves the road along the direction of the path line based on the guidance of the path line level in the directed graph when encountering an intersection, that is, the corresponding path-level driving guidance line is displayed according to the path-level guidance mark. The lane-level navigation method refers to guidance based on the lane line, that is, the guidance method adopted when the terminal has determined the lane position of the own vehicle and the target lane position.

[0064] For example, when the target vehicle is traveling on the road, the terminal can display the path-level guidance sign along the direction of the path line in the current road. Assuming that when the target vehicle travels to 300m away from the next intersection or the next maneuvering point, the vehicle information data of the next intersection or the type of the next maneuvering point can be obtained. When the target vehicle travels to 300m away from the next intersection or the next maneuvering point, the terminal determines the merging lane information of the target vehicle at the next intersection based on the lane position information of the vehicle at the current moment, the vehicle information data of the road in the map, the drivable lane information and the vehicle information, and generates a lane-level guidance sign based on the lane position information of the vehicle at the current moment and the merging lane information. The terminal displays the lane-level guidance sign along the direction of the target lane line in the road so that the target vehicle can travel according to the displayed lane-level guidance sign.

[0065] In the above-mentioned driving navigation method, by acquiring driving data collected by the target vehicle on the driving road, the driving data is identified to obtain lane-level data of the driving road; the lane-level data includes lane line data, drivable lane information and vehicle information; the lane position information of the target vehicle is determined based on the lane line data, and the merging lane information of the target vehicle at the next intersection is determined according to the lane position information, vehicle signal data of the driving road in the map, drivable lane information and vehicle information, and the target vehicle is navigated based on the lane position information and the merging lane information. Since the driving data of the target vehicle is collected when it is driving, and the lane-level data is obtained by identifying the driving data, the lane position of the target vehicle can be determined based on the lane-level data, so that lane-level driving navigation can be performed according to the lane position of the target vehicle and the merging lane information, thereby realizing lane-level guidance based on standard map data and effectively improving the accuracy of road navigation data.

[0066] In one embodiment, the lane-level data is obtained by identifying driving data through a machine learning model; before obtaining the driving data collected by the target vehicle on the driving road, the method further includes:

[0067] Obtaining training data collected by a sample vehicle while driving on a sample road;

[0068] The training data is labeled with lane lines and drivable lanes to obtain labeled training data;

[0069] The pre-trained machine learning model is trained based on the labeled training data to obtain a machine learning model for identifying driving data.

[0070] The labeled training data refers to the labeled training data.

[0071] The machine learning model refers to a model obtained by training the initial model using different training methods, that is, a machine learning model used to identify driving data.

[0072] Specifically, before the terminal obtains the driving data collected by the target vehicle on the driving road, the terminal can obtain the training data collected by the sample vehicle during the driving process of the sample road, and mark the lane lines and drivable lanes on the training data to obtain marked training data. Furthermore, the terminal can train the pre-training machine learning model based on the marked training data to obtain a machine learning model for identifying driving data.

[0073] For example, the terminal can use deep learning to train and deploy the model. First, the terminal can pre-train the initial model. The training steps may include:

[0074] 1. Data labeling, that is, labeling the selected sample cars, lane lines, and drivable areas;

[0075] 2. Build the model. You can use the backbone (main network) + branch (branch network) method to build the initial model;

[0076] 3. Use the training framework to perform classification training on the initial model that has been built;

[0077] 4. Model quantization, that is, obtaining a machine learning model for identifying the driving data.

[0078] Secondly, the terminal can deploy the trained machine learning model, that is, the trained machine learning model can be deployed in the navigation system or the APP (Application) running environment, for example, using the TNN framework of a specific embedded platform for deployment.

[0079] In this embodiment, by training the pre-training machine learning model with labeled training data, a machine learning model for identifying driving data can be obtained, and the machine learning model is deployed in the driving navigation system, so that during the vehicle driving process, the collected driving data can be automatically identified through the trained machine learning model to obtain more accurate lane-level data, thereby providing more accurate lane-level data for subsequent lane-level positioning.

[0080] In one embodiment, the step of determining lane position information of a target vehicle based on lane line data includes:

[0081] Determine the lane markings of each lane on the driving road according to the lane line data;

[0082] Based on the lane markings of each lane, the lane position information of the target vehicle is determined.

[0083] Among them, the lane identification refers to the identification information used to uniquely identify each lane. For example, road A contains 4 lanes. Starting from the left edge of road A, the first lane corresponds to lane 1, the second lane corresponds to lane 2, the third lane corresponds to lane 3, and the fourth lane corresponds to lane 4.

[0084] Specifically, the terminal obtains the driving data collected by the target vehicle on the driving road, identifies the driving data, and obtains the lane-level data of the driving road. Then, the terminal can determine the lane markings of each lane in the driving road according to the lane line data in the lane-level data; further, the terminal can determine the lane position information of the target vehicle based on the lane markings of each lane.

[0085] For example, the terminal obtains the driving data collected by the target vehicle A on the driving road A as image A. The terminal uses the trained machine learning model to identify image A. After obtaining the lane-level data A of the driving road A, the terminal can determine that there are three lanes in the driving road A and the lane markings of each lane are lane 1, lane 2, and lane 3 based on the lane line data in the lane-level data A as three white solid lines; further, the terminal can determine the lane position information of the target vehicle as lane 2 based on the lane markings of each lane. As a result, the lane markings of each lane in the driving road can be determined based on the lane line data, and the lane position of the target vehicle can be determined based on the lane markings of each lane, thereby achieving accurate lane-level positioning.

[0086] In one embodiment, Figure 4 As shown, the vehicle signal data includes the number of lanes and the lane type of each lane; the step of determining the merging lane information of the target vehicle at the next intersection according to the lane position information, the vehicle signal data of the driving road in the map, the drivable lane information and the vehicle information includes:

[0087] Step 402, obtaining the driving direction of the target vehicle on the driving road.

[0088] Step 404, obtaining the number of lanes of the driving road in the driving direction and the lane type of each lane in the map.

[0089] Step 406, determining the merging lane information of the target vehicle at the next intersection based on the lane position information, the number of lanes, the lane type, the drivable lane information and the vehicle information.

[0090] Specifically, the terminal obtains the driving data collected by the target vehicle on the driving road, identifies the driving data, obtains the lane-level data of the driving road, and determines the lane position information of the target vehicle based on the lane line data in the lane-level data. Then, the terminal can determine the merging lane information of the target vehicle at the next intersection based on the lane position information of the vehicle, the vehicle signal data of the driving road in the map, the drivable lane information and the vehicle information.

[0091] SD map refers to a standard map, and the map accuracy has an error of more than meter level (5-10 meters). In the SD map, when the target vehicle is driving near an intersection or a maneuvering point, the terminal can obtain the lane type saved in the SD map, such as Figure 5 As shown in the figure, it is a schematic diagram of vehicle signal data. That is, there is no complete lane-level data in the SD map. Only when the vehicle drives to the nearest intersection or maneuvering point will the number of lanes and the road type of each lane be available, such as Figure 5The vehicle signal icon shown. That is, the terminal obtains the driving direction of the target vehicle on the driving road, and obtains the number of lanes of the driving road in the driving direction and the lane type of each lane in the map. It can be understood that the vehicle signal data in the embodiment of the present application includes but is not limited to Figure 5 The vehicle signal data mentioned in the description may also be other vehicle signal data.

[0092] Since not all sections of the SD map have lane-level data, only the map data of the section of the road near the intersection reaches the lane level. Therefore, in the embodiment of the present application, it is only necessary to solve the positioning problem of the vehicle during driving to determine the lane information of the vehicle. That is, in AR navigation, since the target vehicle is equipped with a camera, the terminal can call the camera to collect driving data in real time, and train a machine learning model for identifying driving data through deep learning, so as to identify the drivable area, vehicle information and lane line information in the driving data. The terminal can determine the edge of the current driving road based on the above-identified lane line information. Further, the terminal can determine the lane to which the vehicle belongs based on the edge of the current driving road.

[0093] Furthermore, after the terminal determines the lane position information to which the vehicle belongs, the terminal can determine the merging lane information of the target vehicle at the next intersection based on the lane position information to which the vehicle belongs, the vehicle signal data of the driving road in the map, the drivable lane information and the vehicle information. Alternatively, the terminal can also determine the merging lane information of the target vehicle at the next intersection based on the lane position information to which the vehicle belongs, the maneuvering point type of the driving road in the map, the drivable lane information and the vehicle information.

[0094] Among them, the vehicle information data and maneuvering point type of the driving road in the map are data stored in the map. When the vehicle travels to the next intersection or the next maneuvering point at a preset distance, the vehicle information data of the next intersection or the relevant data of the next maneuvering point can be obtained from the map database. The maneuvering point type is used to mark the type of the intersection ahead, such as Figure 6 As shown, it is a schematic diagram of the maneuvering point type. It can be understood that the maneuvering point types in the embodiment of the present application include but are not limited to Figure 6 The maneuvering point type mentioned in may also be other maneuvering point types.

[0095] After the terminal determines the mergible lane information of the target vehicle at the next intersection, the terminal can display the corresponding driving guidance route in the navigation interface. It can be understood that the GPS navigation system of the target vehicle in the embodiment of the present application can be a navigation system based on a vehicle-mounted GPS receiver, combined with other navigation means to obtain carrier position data, and matched with a navigation map database, to display the carrier position in real time and provide road guidance.

[0096] In this embodiment, the target vehicle's merging lane information at the next intersection can be determined based on the lane position of the vehicle, the vehicle information data of the driving road in the map, the drivable lane information and the vehicle information, thereby realizing lane-level driving route guidance, thereby effectively improving the accuracy of driving road navigation data.

[0097] In one embodiment, the step of navigating the target vehicle based on the lane position information and the mergible lane information includes:

[0098] When the lane position information and the mergible lane information belong to the same lane, the first driving guidance route is displayed in the navigation interface of the map;

[0099] When the lane position information and the mergible lane information belong to different lanes, a second driving guidance route and lane change reminder information are displayed in the navigation interface of the map; the lane change reminder information is used to instruct the target vehicle to switch to the lane corresponding to the mergible lane information.

[0100] Among them, the driving guidance route refers to the driving guidance route used to instruct the target vehicle to drive along the planned driving guidance route. In the embodiment of the present application, the first driving guidance route refers to the driving guidance route generated when the lane to which the vehicle belongs and the mergible lane belong to the same lane, and the second driving guidance route refers to the driving guidance route generated when the lane to which the vehicle belongs and the mergible lane belong to different lanes.

[0101] The lane change reminder information is used to prompt the target vehicle to change lanes according to the driving guidance route. For example, the lane change reminder information is: instructing the target vehicle to drive from the current lane to the left lane.

[0102] Specifically, after the terminal determines the mergible lane information of the target vehicle at the next intersection based on the lane position information of the vehicle, the vehicle information data of the driving road in the map, the drivable lane information and the vehicle information, the terminal can navigate the target vehicle based on the lane position information and the mergible lane information. If the terminal detects that the lane position information of the vehicle and the mergible lane information belong to the same lane, the terminal displays the first driving guidance route in the navigation interface of the map, wherein the first driving guidance route can be a driving guidance route that continues to guide the target vehicle to drive in the current lane. If the terminal detects that the lane position information of the vehicle and the mergible lane information belong to different lanes, the terminal displays the second driving guidance route and lane change reminder information in the navigation interface of the map; wherein the second driving guidance route can be a driving guidance route that guides the target vehicle to drive from the current lane into the target lane, and the lane change reminder information is used to instruct the target vehicle to switch to the target lane corresponding to the mergible lane information.

[0103] For example, Figure 7As shown in FIG. 1 , it is a schematic diagram of lane line guidance according to the type of maneuvering point. Assuming that the target vehicle is currently traveling on the driving road A, when the target vehicle travels to the front intersection B 300 meters away from the driving direction of the target vehicle, the terminal can obtain from the map data that the type of the front intersection B is a three-way intersection, so go left. Figure 7 As shown in FIG. 1 , assuming that the terminal detects that the lane position information of the vehicle is lane 2 and the target lane is lane 1, that is, when the terminal detects that the lane position information of the vehicle and the merging lane information belong to different lanes, the terminal displays the second driving guidance route and lane change reminder information in the navigation interface of the map, that is, guides the target vehicle to drive to the leftmost lane 1, as shown in FIG. Figure 7 The black arrow in the middle is the correct guidance route, namely the second driving guidance route. Thus, it is possible to generate a corresponding lane-level guidance route based on the vehicle information data in the standard map or the type of maneuvering point, thereby effectively improving the accuracy of driving road navigation data.

[0104] In one embodiment, the step of determining lane position information of a target vehicle based on lane line data includes:

[0105] The lane line data, drivable lane information and vehicle information are recorded and processed by the recorder to obtain driving environment information;

[0106] Based on the driving environment information, determine the lane line data of the driving road at the current moment;

[0107] Determine the lane markings of each lane on the current driving road according to the lane line data of the current driving road;

[0108] Based on the lane markings of each lane, the lane position information of the target vehicle at the current moment is determined.

[0109] The recorder refers to a model for dynamically estimating the state of the system during driving, that is, dynamically estimating the state of the vehicle body environment at each moment of driving. For example, the recorder in the embodiment of the present application can be an EKF, an extended Kalman filter, a high-efficiency recursive filter, which can dynamically estimate the state of the driving system from a series of measurements that do not completely contain noise.

[0110] Driving environment information refers to the vehicle body environment corresponding to each moment when the target vehicle is driving on the road. For example, the driving environment information may be that at 12:10, vehicle A is driving on section AB, the distance to the next intersection is 1,200 meters, and the vehicle information in each lane within the 50m field of view ahead exists.

[0111] Specifically, the terminal obtains the driving data collected by the target vehicle on the driving road, identifies the driving data, and obtains the lane-level data of the driving road. The terminal can determine the lane position information of the target vehicle based on the lane line data in the lane-level data. Since the lane line data corresponding to different times may be the same or different; or the camera's shooting angle is blocked in some sections, that is, the camera's shooting angle is blocked or the target vehicle cannot collect effective driving data when driving, the terminal can record and process the lane line data, drivable lane information and vehicle information through a recorder to obtain the driving environment information at each time. Based on the driving environment information, the terminal determines the lane line data of the driving road at the current time, and determines the lane markings of each lane in the driving road at the current time based on the lane line data of the driving road at the current time. Further, the terminal can determine the lane position information of the target vehicle at the current time based on the lane markings of each lane.

[0112] For example, the terminal obtains the driving data A collected by the target vehicle A on the driving road A at 12:10, and identifies the driving data A. After obtaining the lane-level data A of the driving road, the terminal can record and process the lane line data, drivable lane information and vehicle information contained in the lane-level data A through a recorder to obtain the driving environment information at each time after 12:10. Based on the driving environment information, the terminal determines that there are 3 lane line data of the driving road at the current time 12:12, and according to the lane line data of the driving road at the current time 12:12, the lane markings of each lane in the driving road at the current time are determined to be lane 1 and lane 2. Furthermore, the terminal can determine that the lane position information of the target vehicle at the current time 12:12 is lane 1 based on the lane markings of each lane. That is, the purpose of using a recorder in this application is to record the driving environment information at every moment while driving, and each frame of driving environment data obtained at each moment is continuous. Because sometimes the camera angle may not be able to capture valid driving data, the lane line data, drivable lane information and vehicle information are recorded and processed by the recorder, thereby avoiding the problem of inaccurate collected driving data due to the fact that the camera angle cannot capture valid data, and can make the obtained driving environment information globally unique, while also effectively improving the accuracy of vehicle positioning.

[0113] In one embodiment, the step of recording and processing lane line data, drivable lane information, and vehicle information by a recorder to obtain driving environment information includes:

[0114] Input lane line data, drivable lane information and vehicle information into the recorder for recording and processing to obtain current driving environment information;

[0115] Get the target time to be queried;

[0116] The driving environment information corresponding to the target time is searched in the current driving environment information and the historical driving environment information.

[0117] The historical driving environment information refers to the driving environment information corresponding to the road section that has been traveled. For example, if vehicle A is currently traveling to road section BC, the driving environment information obtained in the road section AB that has been traveled is the historical driving environment information.

[0118] Specifically, the terminal obtains the driving data collected by the target vehicle on the driving road, identifies the driving data, and after obtaining the lane-level data of the driving road, the terminal can record and process the lane line data, drivable lane information and vehicle information contained in the lane-level data through a recorder to obtain the driving environment information at each moment, that is, the terminal inputs the lane line data, drivable lane information and vehicle information contained in the lane-level data into the recorder for recording and processing to obtain the current driving environment information. Furthermore, the terminal can obtain the target moment to be queried input by the driver, and the terminal queries the driving environment information corresponding to the target moment in the current driving environment information and the historical driving environment information.

[0119] For example, at 12:10, the terminal obtains the driving data A collected by the target vehicle A on the driving road A, and identifies the driving data A. After obtaining the lane-level data A of the driving road, the terminal inputs the lane line data, the drivable lane information and the vehicle information contained in the lane-level data A into the recorder for recording and processing, and obtains the driving environment information at 12:10. Further, the terminal can obtain the target time to be queried input by the driver as 12:05 and 12:10, and the terminal queries the driving environment information corresponding to the target time 12:05 and 12:10 in the current driving environment information and the historical driving environment information. As a result, the driving environment information at each time during driving can be recorded and processed by the recorder for the lane line data, the drivable lane information and the vehicle information. Each frame of driving environment data recorded at each time is continuous, which effectively ensures the global uniqueness of the driving environment information, and also effectively improves the accuracy of determining the lane positioning of the vehicle based on the driving environment information.

[0120] In one embodiment, the driving data is collected by a camera or a sensor; the step of navigating the target vehicle based on the lane position information and the mergible lane information includes:

[0121] When the number of lanes covered by the camera or sensor when collecting data is greater than or equal to the number of lanes in the driving road, a lane-level guidance mark is generated based on the lane position information and the information of the mergable lane, and the target vehicle is navigated according to the lane-level guidance mark;

[0122] When the number of lanes covered when the camera or sensor collects data is less than the number of lanes in the driving road, path-level guidance signs are generated based on lane position information and mergible lane information, and the target vehicle is navigated according to the path-level guidance signs.

[0123] The number of lanes covered refers to the number of lanes in the road that can be covered when collecting data. For example, at 12:10, the camera collects the driving data of the target vehicle A on the driving road A as image A. Five lanes can be identified in the driving image A, so the number of lanes covered is 5.

[0124] The lane-level guidance mark refers to identification information used to uniquely identify a lane-level driving route. For example, the lane-level guidance mark includes a lane-level driving guidance route. The lane-level driving guidance route refers to the relevant information of each lane included in the driving guidance route.

[0125] The path-level guidance mark refers to identification information used to uniquely identify the path-level driving route. For example, the path-level guidance mark includes a path-level driving guidance route. The path-level driving guidance route means that the driving guidance route does not contain relevant information of each lane, but only contains path-level information.

[0126] Specifically, when the terminal navigates the target vehicle based on the lane position information and the mergeable lane information, when the number of lanes covered when the camera or sensor collects data is greater than or equal to the number of lanes in the current driving road, the terminal can generate a lane-level guidance mark based on the lane position information and the mergeable lane information, and navigate the target vehicle according to the lane-level guidance mark; when the number of lanes covered in the data collected by the camera or sensor is less than the number of lanes in the driving road, the terminal can generate a path-level guidance mark based on the lane position information and the mergeable lane information, and navigate the target vehicle according to the path-level guidance mark. The driving data can be collected by a camera or a sensor.

[0127] For example, the terminal can locate the lane-level data identified by the machine learning model. After obtaining the lane where the vehicle is located, the terminal can obtain the vehicle signal data from the server's map database or the local map database, and perform lane line guidance for different scenarios. That is, the terminal can adopt different navigation strategies according to the parameters of the front camera of the target vehicle. When the front camera can cover 5 lanes (i.e., an 18-meter area from left to right) between 10 meters and 20 meters in front of the car, when the terminal detects that the number of vehicle signals is less than or equal to 5 lanes, lane-level guidance is used for navigation; when the terminal detects that the number of vehicle signals is greater than 5 lanes, path line guidance is used for navigation. As a result, different navigation strategies can be dynamically adopted for driving navigation based on the parameters of the camera, thereby effectively improving the accuracy of road navigation data.

[0128] In one embodiment, the lane-level guidance mark includes a lane-level driving guidance route; the steps of generating the lane-level guidance mark based on the lane position information and the mergable lane information, and navigating the target vehicle according to the lane-level guidance mark include:

[0129] Get the maneuver point type of the next intersection;

[0130] Generate driving guidance routes based on lane position information, mergible lane information and maneuver point types;

[0131] The driving guidance route is displayed on the navigation interface of the map so that the target vehicle can travel according to the driving guidance route.

[0132] Among them, the maneuver point type refers to the type used to mark the intersection ahead, such as Figure 6 The following is a schematic diagram of the maneuver point types.

[0133] Specifically, the terminal generates a lane-level guidance sign based on the lane position information and the mergable lane information. When navigating the target vehicle according to the lane-level guidance sign, the terminal can obtain the maneuvering point type of the next intersection, and generate a driving guidance route based on the lane position information, the mergable lane information and the maneuvering point type. Furthermore, the terminal displays the driving guidance route on the navigation interface of the map so that the target vehicle can travel according to the driving guidance route.

[0134] For example, Figure 7 As shown, the terminal obtains the maneuver point type of the next intersection as the left side of the three-way intersection shown in the figure. The goal is to guide the target vehicle to the leftmost lane, that is, lane 1, as shown in Figure 7 The route indicated by the black arrow is the correct guidance route. That is, the terminal generates a driving guidance route based on the vehicle's lane position information of lane 2, the target lane information of lane 1, and the maneuver point type of the three-way intersection. Figure 7The route indicated by the black arrow shown in the figure. Furthermore, the terminal displays the driving guidance route on the navigation interface of the map, so that the target vehicle changes lanes according to the driving guidance route. As a result, the corresponding lane-level guidance route can be generated based on the vehicle information data in the standard map or the type of maneuvering point, thereby effectively improving the accuracy of driving road navigation data.

[0135] In one embodiment, the method further comprises:

[0136] When the target vehicle is traveling on the driving road, a path-level guidance mark is displayed along the direction of the path line on the driving road;

[0137] Navigating the target vehicle based on lane position information and mergeable lane information includes:

[0138] When the target vehicle travels to a preset distance value from the next intersection, a lane-level guidance sign is generated based on the lane position information and the mergable lane information, and the lane-level guidance sign is displayed along the direction of the target lane line in the driving road.

[0139] Specifically, when the target vehicle is traveling on the driving road, the terminal displays path-level guidance signs along the direction of the path line in the driving road. That is, when the target vehicle is traveling on the road section, the terminal can use path-level navigation for driving navigation, that is, cruise guidance. Cruise guidance refers to the route guidance that should be provided when no maneuvers are encountered.

[0140] When the target vehicle travels to a preset distance value from the next intersection or the next maneuvering point, the terminal can generate a lane-level guidance sign based on the lane position information of the vehicle and the information of the mergable lane. The terminal displays the lane-level guidance sign along the direction of the target lane line in the driving road, that is, when the target vehicle travels to an adjacent intersection or maneuvering point, maneuvering guidance needs to be provided in advance. For example, when the target vehicle travels to a preset distance value from the next intersection or the next maneuvering point, the terminal can use lane-level navigation for driving navigation, so that the vehicle can change lanes or routes in advance.

[0141] Maneuvering guidance refers to guidance that should be provided in advance when encountering one of the following situations during route guidance: when encountering an intersection on the route, it is not a straight-through intersection, or it is necessary to switch to a road with a different road grade from the current one; when encountering a roundabout on the route. When the target vehicle is traveling along the pre-calculated route, the terminal can provide the driver with cruise guidance information, maneuvering guidance information and prompt information based on the real-time position of the target vehicle on the route. For example, during maneuvering guidance, the terminal can provide the following information in the navigation interface: when encountering an intersection on the route, if it is not a straight-through intersection, or it is necessary to switch to a road with a different road grade from the current one, the terminal can provide the turning direction (left turn or right turn), turning angle or fast or slow, or U-turn at the intersection in the navigation interface; when encountering a roundabout on the route, the terminal can provide information on the number of intersections on the roundabout to leave the roundabout in the navigation interface. During cruise guidance, the terminal can provide the maneuvering method of the next maneuver ahead of the route in the navigation interface, including the turning direction, angle or roundabout. Among them, the timing of providing maneuvering guidance information and prompt information can be pre-set, and the terminal can provide corresponding information before the maneuvering and prompting position. For example, the preset distance can be set to 300m in advance on an ordinary road or street, and can be set to 1000m in advance on a highway; if the distance between two consecutive maneuvering or prompting positions is less than 100m on an ordinary road, or less than 300m on a highway, while providing guidance information for the previous maneuvering or prompting position, the terminal can also prompt that the next maneuver or prompt is very close.

[0142] For example, assuming that the preset distance value is 50m, when the target vehicle is traveling on a section of the driving road, the terminal can display a path-level guidance sign along the direction of the path line in the driving road; when the target vehicle travels to 50m from the next intersection, the terminal can generate a lane-level guidance sign based on the lane position information and the mergeable lane information, and display the lane-level guidance sign along the direction of the target lane line in the driving road. In addition, when the terminal displays the map guidance screen in the navigation interface, the terminal can switch between the maneuvering guidance screen and the cruise guidance screen, that is, the terminal can switch between the lane-level guidance screen and the path-level guidance screen. As a result, different driving navigation methods can be used according to the dynamic driving environment during driving, especially when approaching an intersection or near a maneuvering point, the corresponding lane-level driving guidance route can be displayed in advance, providing users with a more flexible and accurate driving navigation route.

[0143] In one embodiment, the driving data includes vehicle data; the vehicle data is the position information and quantity information of each vehicle within the field of view when the target vehicle is driving on the driving road; the method further includes:

[0144] If the distance between the location where the target vehicle is traveling and the next intersection is greater than a preset distance value, the target lane information in the drivable lane information is determined based on the location information and quantity information of each vehicle;

[0145] The target lane information is displayed in the navigation interface of the map; the target lane information is used to prompt lane change when driving on the driving road.

[0146] Specifically, if the distance between the location of the target vehicle and the next intersection is greater than the preset distance value, the terminal can determine the target lane information in the drivable lane information based on the location information and quantity information of each vehicle, and display the target lane information in the navigation interface of the map. The target lane information is used to prompt the driver to change lanes when driving on the driving road. That is, when there is no adjacent intersection or maneuvering point, the terminal can also re-plan the best lane-level driving guidance route based on the preset navigation strategy.

[0147] For example, assuming that the preset distance value is 50m, when the target vehicle is driving on a road section, the terminal can display a path-level guidance mark along the direction of the path line in the driving road; if the distance between the location where the target vehicle is driving and the next intersection is greater than 50m, because there are many vehicles in lane 1 at the current moment and there is congestion, the terminal can determine that the target lane in the drivable lane information is lane 2 based on the shortest time strategy and the location information and quantity information of each vehicle in the front field of view at the current moment, and display the target lane 2 in the navigation interface of the map. The target lane 2 is used to prompt the driver to change from the current lane to lane 2 when driving on the driving road. As a result, the corresponding lane-level driving guidance route can be displayed in the navigation interface, providing users with a more accurate driving navigation route, thereby effectively improving the accuracy of driving road navigation data.

[0148] In one embodiment, the method further comprises:

[0149] When there is no merging lane information for the target vehicle at the next intersection, the target lane information that can be driven at the next moment is re-determined based on the lane position information and destination position information of the target vehicle at the current moment.

[0150] Specifically, when the terminal determines that there is no merging lane information for the target vehicle at the next intersection based on the lane position information of the vehicle itself, the vehicle signal data of the driving road in the map, the drivable lane information and the vehicle information, the terminal can re-determine the drivable target lane information at the next moment based on the lane position information of the target vehicle at the current moment and the destination position information.

[0151] For example, in some special cases, when the road ahead is temporarily closed due to an accident or construction, the terminal can re-determine the target lane information that can be driven at the next moment based on the lane position information and destination position information of the target vehicle at the current moment. This makes it possible to display the corresponding lane-level driving guidance route in advance according to the dynamic driving environment during driving, providing users with a more flexible and accurate driving navigation route.

[0152] The present application also provides an application scenario, and the application scenario applies the above-mentioned driving navigation method. Specifically, the application of the driving navigation method in the application scenario is as follows:

[0153] When the navigation system in the vehicle needs to provide real-time route guidance to the driver, the above-mentioned driving navigation method can be used, that is, after the driver logs in to the navigation system in the vehicle terminal, the vehicle terminal can automatically obtain the driving data collected by the target vehicle on the driving road, and identify the driving data to obtain the lane-level data of the driving road. The vehicle terminal can determine the lane position of the vehicle based on the lane line data contained in the lane-level data, and determine the target lane of the vehicle at the next intersection based on the lane position of the vehicle, the vehicle information data of the driving road in the map, the drivable lane information and the vehicle information in the front field of view, and navigate the driving route of the vehicle based on the lane position of the vehicle and the target lane. Among them, the navigation system in the vehicle terminal can be implemented by an independent server or a server cluster composed of multiple servers.

[0154] The method provided in the embodiment of the present application can be applied to the lane-level guidance scenario based on SD map data in the AR navigation scenario. The following takes the lane-level guidance scenario based on SD map data in the AR navigation scenario as an example to illustrate the driving navigation method provided in the embodiment of the present application.

[0155] In the traditional way, SD map is a standard map, and the map accuracy has an error of more than meter level. SD map navigation cannot achieve lane-level navigation, but only path-level navigation. The reasons are as follows:

[0156] 1. GPS positioning accuracy is low, generally around 10 meters;

[0157] 2. SD maps do not have complete lane-level data. Only the number of lanes and the road type of each lane are available at the adjacent intersections.

[0158] Therefore, when using SD maps for driving navigation, it is easy to cause the accuracy of the obtained road navigation data to be low. In order to solve the above problems, the present application provides a lane-level guidance method based on SD map data in an AR navigation scenario, that is, in the embodiment of the present application, in the AR navigation scenario, lane-level positioning can be achieved, that is, the camera is used to locate which lane the vehicle is in, and lane-level guidance can also be performed, that is, different guidance strategies are supported, so that AR navigation has the ability to guide lane lines in sections near intersections.

[0159] On the technical side, Figure 8 Shown is a flowchart of lane-level guidance.

[0160] Among them, Deep Neural Networks is a branch of machine learning. It is an algorithm for compressing data with high correlation. It can automatically and efficiently parameterize data features to avoid manual design of data features. Multiple sensors include GPS, IMU, camera, wheel speed, and magnetometer. GPS, Global Positioning System, can measure the current longitude and latitude. IMU, Inertial measurement unit, can measure the current acceleration, rotation speed, etc. Camera, generally records the current image at a frame rate of 30 frames per second.

[0161] The specific steps are as follows:

[0162] 1) The terminal uses a deep neural network model to detect the collected driving data and can identify the drivable area (free space), lane lines (lane) and cars;

[0163] 2) The terminal inputs the data identified in step 1 into the recorder to perform EKF (Extended Kalman Filter) tracking;

[0164] 3) The terminal can query the corresponding driving environment status according to the timestamp, and determine the road edge according to the query result, and then determine the lane where the vehicle is located based on the road edge;

[0165] 4) The terminal generates corresponding lane-level guidance based on the lane where the vehicle is located and the vehicle information data in the SD map, or the type of maneuvering point.

[0166] Model detection and lane-level localization:

[0167] a. The output types of the deep neural network model that the terminal needs to train are: car detection, lane line detection, and drivable area detection;

[0168] b. The terminal inputs the output of the deep neural network model trained in step a into the tracker for tracking. The client, i.e. the user, can query the output results at any time by means of timestamp query, and dynamically construct the vehicle body environment based on the three types of outputs of the above-mentioned vehicle detection, lane line detection, and drivable area detection, i.e., know where the drivable area in front of the car is, where the lane is, and where there are other cars;

[0169] c. From step b, we can know which lane belongs to the road edge. The terminal can determine which lane the vehicle is in based on the number of times the road edge passes. For example, the second lane from the left can get the lane-level positioning of the vehicle.

[0170] The terminal can use deep learning to train and deploy the model. First, the terminal can pre-train the initial model. The training steps may include:

[0171] 1. Data labeling, that is, labeling the selected sample cars, lane lines, and drivable areas;

[0172] 2. Build the model. You can use the backbone (main network) + branch (branch network) method to build the initial model;

[0173] 3. Use the training framework to perform classification training on the initial model that has been built;

[0174] 4. Model quantization, that is, obtaining a machine learning model for identifying the driving data.

[0175] Secondly, the terminal can deploy the trained machine learning model, that is, the trained machine learning model can be deployed in the navigation system or the APP (Application) running environment, for example, using the TNN framework of a specific embedded platform for deployment.

[0176] Lane Level Guidance:

[0177] 1. The terminal obtains the lane where the vehicle is located based on the lane-level data and positioning identified by the aforementioned machine learning model;

[0178] 2. After the terminal obtains the vehicle information data from the map side, it provides lane line guidance for different scenarios

[0179] Specifically, there are different strategies based on the parameters of the front camera of the car. For example, if the lane width is 3.5 meters, the 5-lane area is 17.5 meters, and the 3-lane area is 10.5 meters.

[0180] 1) When the front camera can cover 5 lanes (i.e., 17.5 meters from left to right) between 10 meters and 20 meters in front of the car, when the terminal detects that the number of vehicle signals is less than or equal to 5 lanes, lane-level guidance is used for navigation; when the terminal detects that the number of vehicle signals is greater than 5 lanes, path line guidance is used for navigation.

[0181] 2) When the front camera can cover 3 lanes (i.e., 10.5 meters from left to right) between 10 meters and 20 meters in front of the car, when the terminal detects that the number of vehicle signals is less than or equal to 3 lanes, lane-level guidance is used for navigation; when the terminal detects that the number of vehicle signals is greater than 3 lanes, path line guidance is used for navigation.

[0182] Lane-level guidance example:

[0183] Example 1, such as Fig. 9 As shown in FIG. 1 , it is a schematic diagram of navigation using a path line navigation method, that is, Fig. 9 Lane-level navigation guidance is not used in Fig. 9 The path line is used for paving, but the paving points to the wrong direction, that is, the paving points to lane 2. It is necessary to change lanes because according to the data from the vehicle signal, the vehicle is in the correct lane, but the actual paving points to the wrong direction; Fig. 9 The paving based on the path line is ambiguous and points to the right lane turning intersection.

[0184] like Fig.10 As shown, it is a schematic diagram of navigation using lane-level navigation. Fig.10 Lane-grade paving is used in Fig.10 It is shown that the vehicle is on the correct road. The terminal continues to draw and pave the road along lane 1. If the vehicle is on the wrong lane, a lane change reminder should be given.

[0185] Example 2, such as Figure 7 As shown, it is a schematic diagram of lane line guidance according to the type of maneuvering point. Figure 7 The maneuver point type is obtained as left at the three-way intersection. The goal is to guide the target vehicle to the leftmost lane, that is, to guide the target vehicle to the leftmost lane 1. Figure 7 The black arrow is the correct guide route.

[0186] Among them, SD maps, or high-precision maps, have corresponding production specifications. Generally speaking, SD-level maps are built based on LINK-level data. The LINK level can be simply understood as data with only one link, that is, all lanes merge into one path. When using SD-level maps, there is only car signal data when approaching an intersection, and this car signal data is a supplement to the SD data.

[0187] It can be understood that the map data in this application is composed of a directed graph, and each node represents a key GPS location, for example Fig.11 As shown in the figure, it is a schematic diagram of a directed graph, and the lines between nodes are directed graph structures. Fig.12 The figure shows a standard map or a schematic diagram of a road network. A node refers to an intersection of roads or a point that distinguishes different parts of a road. A road segment refers to a road between two nodes. A road network refers to a road network composed of nodes and road segments.

[0188] In this embodiment, in the AR navigation scenario, lane-level positioning can be achieved, that is, the camera is used to locate which lane the vehicle is in. At the same time, lane-level guidance can be performed, that is, different guidance strategies are supported, so that AR navigation has the ability to provide lane line guidance on roads near intersections.

[0189] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0190] Based on the same inventive concept, the embodiment of the present application also provides a driving navigation device for implementing the driving navigation method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more driving navigation device embodiments provided below can refer to the limitations of the driving navigation method above, and will not be repeated here.

[0191] In one embodiment, Fig.13 As shown, a driving navigation device is provided, including: an acquisition module 1302, an identification module 1304, a determination module 1306 and a navigation module 1308, wherein:

[0192] The acquisition module 1302 is used to acquire driving data collected by the target vehicle on the driving road.

[0193] The identification module 1304 is used to identify the driving data and obtain lane-level data of the driving road; the lane-level data includes lane line data, drivable lane information and vehicle information.

[0194] The determination module 1306 is used to determine the lane position information of the target vehicle based on the lane line data; and determine the merging lane information of the target vehicle at the next intersection according to the lane position information, the vehicle information data of the driving road in the map, the drivable lane information and the vehicle information.

[0195] The navigation module 1308 is used to navigate the target vehicle based on the lane position information and the mergible lane information.

[0196] In one embodiment, the device also includes: an acquisition module for acquiring training data collected by the sample vehicle during driving on the sample road; a labeling module for labeling lane lines and drivable lanes on the training data to obtain labeled training data; and a training module for training a pre-training machine learning model based on the labeled training data to obtain a machine learning model for identifying the driving data.

[0197] In one embodiment, the determination module is further used to determine the lane markings of each lane in the driving road according to the lane line data; and determine the lane position information of the target vehicle based on the lane markings of each lane.

[0198] In one embodiment, the acquisition module is also used to obtain the driving direction of the target vehicle on the driving road; obtain the number of lanes of the driving road in the driving direction and the lane type of each lane in the map; the determination module is also used to determine the merging lane information of the target vehicle at the next intersection based on the lane position information, the number of lanes, the lane type, the drivable lane information and the vehicle information.

[0199] In one embodiment, the device also includes: a display module, which is used to display a first driving guidance route in the navigation interface of the map when the lane position information and the mergible lane information belong to the same lane; when the lane position information and the mergible lane information belong to different lanes, display a second driving guidance route and lane change reminder information in the navigation interface of the map; the lane change reminder information is used to instruct the target vehicle to switch to the lane corresponding to the mergible lane information.

[0200] In one embodiment, the device also includes: a recording and processing module, which is used to record and process the lane line data, the drivable lane information and the vehicle information through a recorder to obtain driving environment information; the determination module is also used to determine the lane line data of the driving road at the current moment based on the driving environment information; according to the lane line data of the driving road at the current moment, determine the lane markings of each lane in the driving road at the current moment; based on the lane markings of each lane, determine the lane position information of the target vehicle at the current moment.

[0201] In one embodiment, the device also includes: a recording processing module is also used to input the lane line data, the drivable lane information and the vehicle information into a recorder for recording and processing to obtain current driving environment information; the acquisition module is also used to obtain the target time to be queried; and the query module is used to query the driving environment information corresponding to the target time in the current driving environment information and the historical driving environment information.

[0202] In one embodiment, the device also includes: a generation module, which is used to generate a lane-level guidance mark based on the lane position information and the mergible lane information when the number of lanes covered when the camera or the sensor collects data is greater than or equal to the number of lanes in the driving road; the navigation module is also used to navigate the target vehicle according to the lane-level guidance mark; the generation module is also used to generate a path-level guidance mark based on the lane position information and the mergible lane information when the number of lanes covered when the camera or the sensor collects data is less than the number of lanes in the driving road; the navigation module is also used to navigate the target vehicle according to the path-level guidance mark.

[0203] In one embodiment, the acquisition module is also used to obtain the maneuvering point type of the next intersection; the generation module is also used to generate a driving guidance route based on the lane position information, the mergible lane information and the maneuvering point type; the display module is also used to display the driving guidance route on the navigation interface of the map, so that the target vehicle can travel according to the driving guidance route.

[0204] In one embodiment, the display module is also used to display a path-level guidance sign along the direction of the path line in the driving road when the target vehicle is traveling on the driving road; the generation module is also used to generate a lane-level guidance sign based on the lane position information and the mergible lane information when the target vehicle travels to a preset distance value from the next intersection; the display module is also used to display the lane-level guidance sign along the direction of the target lane line in the driving road.

[0205] In one embodiment, the determination module is also used to determine the target lane information in the drivable lane information based on the position information and the quantity information of each vehicle if the distance between the position where the target vehicle is traveling and the next intersection is greater than a preset distance value; the display module is also used to display the target lane information in the navigation interface of the map; the target lane information is used to prompt a lane change when driving on the driving road.

[0206] Each module in the above-mentioned driving navigation device can be implemented in whole or in part by software, hardware or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0207] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Fig.14 As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected via a system bus, and the communication interface, the display unit and the input device are connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a driving navigation method is implemented. The display unit of the computer device is used to form a visually visible image, and can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covered on the display screen, or a button, trackball or touchpad set on the computer device casing, or an external keyboard, touchpad or mouse, etc.

[0208] Those skilled in the art will understand that Fig.14 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0209] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0210] In one embodiment, a computer-readable storage medium is provided, storing a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0211] In one embodiment, a computer program product or computer program is provided, the computer program product or computer program includes computer instructions, the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the steps in the above-mentioned method embodiments.

[0212] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0213] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0214] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0215] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A driving navigation method, characterized in that: The method comprises: Acquire driving data collected by the target vehicle on the driving road; Identify the driving data to obtain lane-level data of the driving road; the lane-level data includes lane line data, drivable lane information and vehicle information; Determining lane position information of the target vehicle based on the lane line data; Determine the lane information of the target vehicle at the next intersection that can be merged according to the lane position information, the vehicle signal data of the driving road in the map, the drivable lane information and the vehicle information; the vehicle signal data includes the number of lanes; The target vehicle is navigated based on the lane position information and the mergible lane information, including: the driving data is collected by a sensor, and when the number of lanes covered when the sensor collects data is greater than or equal to the number of lanes in the driving road in the map, a lane-level guidance mark is generated based on the lane position information and the mergible lane information, and the target vehicle is navigated according to the lane-level guidance mark; when the number of lanes covered when the sensor collects data is less than the number of lanes in the driving road in the map, a path-level guidance mark is generated based on the lane position information and the mergible lane information, and the target vehicle is navigated according to the path-level guidance mark.

2. The method according to claim 1, characterized in that The lane-level data is obtained by identifying the driving data through a machine learning model; before obtaining the driving data collected by the target vehicle on the driving road, the method further includes: Obtaining training data collected by a sample vehicle while driving on a sample road; Performing lane line marking and drivable lane marking on the training data to obtain marked training data; The pre-trained machine learning model is trained based on the labeled training data to obtain a machine learning model for identifying the driving data.

3. The method according to claim 1, characterized in that The determining the lane position information of the target vehicle based on the lane line data includes: Determining lane markings of lanes on the driving road according to the lane line data; Based on the lane markings of the lanes, lane position information of the target vehicle is determined.

4. The method according to claim 1, characterized in that: The vehicle signal data also includes the lane type of each lane; Determining the merging lane information of the target vehicle at the next intersection according to the lane position information, the vehicle information data of the driving road in the map, the drivable lane information and the vehicle information includes: Obtaining the driving direction of the target vehicle on the driving road; Acquire the number of lanes of the driving road in the driving direction and the lane type of each lane in the map; The merging lane information of the target vehicle at the next intersection is determined according to the lane position information, the number of lanes, the lane type, the drivable lane information and the vehicle information.

5. The method according to claim 4, characterized in that The navigating the target vehicle based on the lane position information and the mergible lane information includes: When the lane position information and the mergible lane information are information of the same lane, displaying a first driving guidance route in the navigation interface of the map; When the lane position information and the mergible lane information belong to information of different lanes, a second driving guidance route and lane change reminder information are displayed in the navigation interface of the map; the lane change reminder information is used to instruct the target vehicle to switch to the lane corresponding to the mergible lane information.

6. The method according to claim 1, characterized in that The determining the lane position information of the target vehicle based on the lane line data includes: The lane line data, the drivable lane information and the vehicle information are recorded and processed by a recorder to obtain driving environment information; Based on the driving environment information, determining lane line data of the driving road at the current moment; Determining the lane markings of each lane in the driving road at the current moment according to the lane line data of the driving road at the current moment; Based on the lane markings of the lanes, the lane position information of the target vehicle at the current moment is determined.

7. The method according to claim 6, characterized in that The lane line data, the drivable lane information and the vehicle information are recorded and processed by a recorder to obtain the driving environment information, including: Inputting the lane line data, the drivable lane information and the vehicle information into a recorder for recording and processing to obtain current driving environment information; Get the target time to be queried; The driving environment information corresponding to the target time is searched in the current driving environment information and the historical driving environment information.

8. The method according to claim 1, characterized in that The lane-level guidance mark includes a lane-level driving guidance route; Generating a lane-level guidance mark based on the lane position information and the mergible lane information, and navigating the target vehicle according to the lane-level guidance mark includes: Obtaining the maneuver point type of the next intersection; Generate a driving guidance route based on the lane position information, the mergible lane information and the maneuvering point type; A driving guidance route is displayed on the navigation interface of the map so that the target vehicle travels according to the driving guidance route.

9. The method according to claim 1, characterized in that: The method further comprises: When the target vehicle is traveling on the driving road, a path-level guidance mark is displayed along the direction of the path line on the driving road; The navigating the target vehicle based on the lane position information and the mergible lane information includes: When the target vehicle travels to a preset distance value from the next intersection, a lane-level guidance sign is generated based on the lane position information and the mergable lane information, and the lane-level guidance sign is displayed along the direction of the target lane line in the driving road.

10. The method according to claim 1, characterized in that The driving data includes vehicle data; the vehicle data is the position information and quantity information of each vehicle within the field of view when the target vehicle is driving on the driving road; the method further includes: If the distance between the position where the target vehicle is traveling and the next intersection is greater than a preset distance value, determining the target lane information in the drivable lane information based on the position information and the quantity information of each of the vehicles; The target lane information is displayed in the navigation interface of the map; the target lane information is used to prompt a lane change when driving on the driving road.

11. A driving navigation device, characterized in that: The device comprises: An acquisition module is used to acquire driving data collected by the target vehicle on the driving road; An identification module, used to identify the driving data to obtain lane-level data of the driving road; the lane-level data includes lane line data, drivable lane information and vehicle information; A determination module, configured to determine the lane position information of the target vehicle based on the lane line data; and determine the merging lane information of the target vehicle at the next intersection according to the lane position information, the vehicle signal data of the driving road in the map, the drivable lane information and the vehicle information; the vehicle signal data includes the number of lanes; A navigation module, used for navigating the target vehicle based on the lane position information and the mergible lane information, comprising: the driving data is collected by a sensor, when the number of lanes covered when the sensor collects data is greater than or equal to the number of lanes in the driving road in the map, a lane-level guidance mark is generated based on the lane position information and the mergible lane information, and the target vehicle is navigated according to the lane-level guidance mark; when the number of lanes covered when the sensor collects data is less than the number of lanes in the driving road in the map, a path-level guidance mark is generated based on the lane position information and the mergible lane information, and the target vehicle is navigated according to the path-level guidance mark.

12. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 10 are implemented.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 10 are implemented.

14. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 10 are implemented.

Images