Lane center line determination method and device, equipment and storage medium

Abstract

The invention relates to a lane center line determination method and device, equipment and a storage medium, and the method comprises the steps: obtaining a current lane collected for multiple times, segmenting the current lane to obtain a plurality of segmented lanes, generating a first lane center line corresponding to the segmented lanes, connecting the first lane center line corresponding to the segmented lanes of the current lane, and generating a second lane center line of the current lane, connecting the second lane center line of the current lane and the second lane center line of the next lane according to the pre-acquired topological positions of the current lane and the next lane, and performing smoothing processing to generate a target lane center line of the target lane. According to the method, the lane center lines of the small-granularity segmented lanes which are subjected to data collection for multiple times at the same position are independently calculated and then are subjected to fitting connection smoothness, the lane center line manufacturing efficiency and the accuracy of the target lane center line are effectively improved, the accuracy of the lane center line is effectively improved, the accuracy of a high-precision map is further guaranteed, and the accuracy of the target lane center line is improved. And the vehicle intelligent driving effect is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent driving, and particularly relates to a method, device, equipment and storage medium for determining a lane center line. Background Art

[0002] As an important support for intelligent driving, a high-precision map contains information with higher precision and richer details compared to a traditional navigation map. Among them, the lane center line with complete topological connection is the key information required by an unmanned driving system. The lane center line is a characteristic line formed by sequentially connecting the center points in the lane width direction from the starting point to the ending point of the lane, which can reflect the curvature change of the lane and be used as a marking line for vehicle driving.

[0003] Currently, the production of the lane center line in an existing high-precision map is to generate discrete lane center lines through the collected actual road data. Since the road data collected each time is discrete and each section of the road is processed independently, there are large deviations. The finally produced lane center line deviates from the actual lane, and there is a problem that the curvature of the lane center line is too large, resulting in insufficient smoothness and distortion, so that the accuracy of the high-precision map cannot be effectively guaranteed, further affecting the intelligent driving effect of the vehicle. Summary of the Invention

[0004] One of the purposes of the present invention is to provide a method for determining a lane center line to solve the problems of low production accuracy and smoothness of the lane center line in existing vehicle intelligent driving; the second purpose is to provide a device for determining a lane center line; the third purpose is to provide an electronic device; the fourth purpose is to provide a readable storage medium.

[0005] To achieve the above purposes, the technical solutions adopted by the present invention are as follows:

[0006] According to the first aspect of the present invention, there is provided a method for determining a lane center line, the method including:

[0007] Obtain the current lane collected multiple times, segment the current lane to obtain a plurality of segmented lanes, and generate a first lane center line corresponding to the segmented lanes;

[0008] Connect the first lane center lines corresponding to the segmented lanes of the current lane to generate a second lane center line of the current lane;

[0009] According to the pre-obtained topological positions of the current lane and the next lane, connect the second lane center lines of the current lane and the next lane and perform smoothing processing to generate a target lane center line of the target lane.

[0010] Optionally, obtaining the current lane collected multiple times, segmenting the current lane to obtain multiple segmented lanes, and generating the first lane centerline corresponding to the segmented lanes includes:

[0011] Pre-divide the target lane at a preset interval to determine the current lane for which the lane centerline is to be generated; wherein, the target lane includes the current lane and the next lane;

[0012] Collect data of the current lane a preset number of times to obtain the lane data of the current lane, wherein the lane data includes the lane length;

[0013] Determine the segmentation interval according to the lane length, uniformly segment the current lane collected multiple times, and generate the initial centerlines corresponding to the multiple segmented lanes;

[0014] Perform fitting processing on the initial centerlines generated from the segmented lanes collected multiple times to generate the first lane centerline corresponding to the segmented lanes.

[0015] Optionally, connecting the first lane centerlines corresponding to the segmented lanes of the current lane to generate the second lane centerline of the current lane includes:

[0016] Obtain the start parameters and end parameters of the first lane centerline corresponding to the segmented lanes;

[0017] Connect the first lane centerlines of the multiple segmented lanes according to the start parameters and end parameters of the first lane centerline;

[0018] Determine the connected first lane centerline as the second lane centerline of the current lane.

[0019] Optionally, according to the topological positions of the current lane and the next lane, connecting and smoothing the second lane centerlines of the current lane and the next lane to generate the target lane centerline of the target lane includes:

[0020] Determine the association relationship between the second lane centerlines corresponding to the current lane and the next lane according to the topological positions of the current lane and the next lane;

[0021] Connect the second lane centerlines of the current lane and the next lane according to the association relationship;

[0022] Smooth the connected second lane centerline to generate the target lane centerline of the target lane.

[0023] Optionally, the step of smoothing the connected second lane centerline to generate the target lane centerline of the target lane includes:

[0024] Performing iterative processing on the connection position of the second lane centerline by using a preset smoothing function to determine the lane center point of the target lane;

[0025] Fitting the lane center point to generate the target lane centerline of the target lane.

[0026] Optionally, before connecting and smoothing the second lane centerlines of the current lane and the next lane according to the topological positions of the current lane and the next lane to generate the target lane centerline of the target lane, the method further includes:

[0027] In response to the determination of the second lane centerline of the current lane, sequentially determining the second lane centerline of the next lane in the target lane.

[0028] According to a second aspect of the present invention, there is provided a lane centerline determination device, the device includes:

[0029] An acquisition data module, configured to acquire the current lane collected multiple times, segment the current lane to obtain a plurality of segmented lanes, and generate a first lane centerline corresponding to the segmented lanes;

[0030] A first generation module, configured to connect the first lane centerlines corresponding to the segmented lanes of the current lane to generate a second lane centerline of the current lane;

[0031] A second generation module, configured to connect and smooth the second lane centerlines of the current lane and the next lane according to the topological positions of the current lane and the next lane acquired in advance to generate a target lane centerline of the target lane.

[0032] Optionally, the acquisition data module includes:

[0033] A division sub-module, configured to pre-divide the target lane at a preset interval to determine the current lane for which the lane centerline is to be generated; wherein the target lane includes the current lane and the next lane;

[0034] An acquisition sub-module, configured to perform data acquisition on the current lane a preset number of times to acquire lane data of the current lane, where the lane data includes the lane length;

[0035] A segmentation sub-module, configured to determine a segmentation interval according to the lane length, uniformly segment the current lane collected multiple times, and generate an initial centerline corresponding to a plurality of segmented lanes;

[0036] The first generation sub-module is used to perform fitting processing on the initial centerlines generated by collecting the segmented lanes multiple times, and generate the first lane centerline corresponding to the segmented lanes.

[0037] According to another aspect of the present invention, there is also provided an electronic device, including:

[0038] A processor;

[0039] A memory for storing instructions executable by the processor;

[0040] Wherein, the processor is configured to execute the instructions to implement the lane centerline determination method as described above.

[0041] According to another aspect of the present invention, there is also provided a readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the lane centerline determination method as described above are implemented.

[0042] Advantages of the present invention:

[0043] The lane centerline determination method provided by the embodiments of the present invention obtains the current lane collected multiple times, segments the current lane to obtain multiple segmented lanes, generates the first lane centerline corresponding to the segmented lanes, connects the first lane centerlines corresponding to the segmented lanes of the current lane to generate the second lane centerline of the current lane, and according to the pre-acquired topological positions of the current lane and the next lane, connects and smooths the second lane centerlines of the current lane and the next lane to generate the target lane centerline of the target lane. The present invention collects lane data multiple times and adopts a segmented processing method, divides the target lane into small-grained segmented lanes, fits and connects the lane centerlines corresponding to the segmented lanes to obtain the lane centerline of the current lane, and according to the topological positions of the current lane and the next lane, connects and smooths the lane centerlines of the current lane and the next lane to determine the target lane centerline of the entire target lane, effectively improving the accuracy of the lane centerline. By separately calculating the lane centerlines of the small-grained segmented lanes that collect data multiple times at the same position and then fitting, connecting and smoothing, the production efficiency of the lane centerline and the accuracy of the target lane centerline are effectively improved. Due to the effective improvement of the accuracy of the lane centerline, the accuracy of the high-precision map is further ensured, and the vehicle intelligent driving effect is improved. Description of the Drawings

[0044] Figure 1 is a flowchart of the steps of a lane centerline determination method provided by an embodiment of the present invention;

[0045] Figure 2 is Figure 1 a flowchart of step 101 of a lane centerline determination method provided by an embodiment of the present invention in

[0046] Figure 3 Yes Figure 1 It is the flowchart of step 102 of a method for determining a lane center line provided by an embodiment of the present invention;

[0047] Figure 4 Yes Figure 1 It is the flowchart of step 103 of a method for determining a lane center line provided by an embodiment of the present invention;

[0048] Figure 5 It is the flowchart of steps of another method for determining a lane center line provided by an embodiment of the present invention;

[0049] Figure 6 It is the scenario schematic of a method for determining a lane center line provided by an embodiment of the present invention Figure 1 ;

[0050] Figure 7 It is the scenario schematic of a method for determining a lane center line provided by an embodiment of the present invention Figure 2 ;

[0051] Figure 8 It is the structural block diagram of a device for determining a lane center line provided by an embodiment of the present invention;

[0052] Figure 9 It is the structural schematic diagram of an electronic device provided by an embodiment of the present invention. Detailed implementation manners

[0053] The following will illustrate the implementation manners of the present invention with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific implementation manners. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention, rather than for limiting the protection scope of the present invention.

[0054] Refer to Figure 1 , which shows the flowchart of steps of a method for determining a lane center line provided by an embodiment of the present invention. The method may include:

[0055] Step 101, obtaining the current lane collected multiple times, segmenting the current lane to obtain multiple segmented lanes, and generating a first lane center line corresponding to the segmented lanes.

[0056] To solve the problem of low accuracy and smoothness in the production of lane centerlines in current vehicle intelligent driving, in the embodiments of the present invention, lane data is collected multiple times, and the lane is processed in a segmented manner. After generating the lane centerline of a small-granularity lane, a fitting and smoothing process is performed. Finally, an accurate lane centerline of the target lane is obtained. Here, the target lane is the lane for which a smooth lane centerline is to be generated. After determining the target lane to be processed in this embodiment, according to the lane data of the target lane, the target lane is horizontally segmented, and the originally large-volume data target lane is divided into multiple lanes with smaller granularities, so as to process the current lane in sequence. Among them, the lane data of the lane includes data such as lane length and lane boundary lines, which is used to determine the division or segmentation granularity of the lane.

[0057] It should be noted that the execution entity for generating the lane centerline in the embodiments of the present invention can be a vehicle terminal or any map data processing device or system. Specifically, it is not limited here. The embodiments of the present invention will be described by taking the precision map processing device as the execution entity as an example.

[0058] In the embodiments of the present invention, the vehicle terminal can obtain lane boundary line data through sensors such as lidar, cameras, and GPS, or directly retrieve the map to obtain lane boundary line data. The lane boundary line data refers to the data information of the lane edge lines marked on the road surface. Each lane is composed of at least two lane boundary lines. According to the driving direction of the vehicle, the lane boundary lines of each lane are divided into a left boundary line and a right boundary line. The lane boundary lines on each side may be composed of single lines or double lines. In the embodiments of the present invention, the target lane is pre-divided into multiple lanes with small segment distances, so as to sequentially determine the lane centerlines of each small-segment lane starting from the current lane.

[0059] Specifically, the current lane collected multiple times is obtained, the current lane is segmented to obtain multiple segmented lanes, and a first lane centerline corresponding to the segmented lane is generated. When collecting lane data, in order to improve the accuracy of the lane data and avoid large deviations in the lane data collected once affecting the accuracy of the lane centerline, in this embodiment, the lane data of the target lane is collected multiple times to obtain the current lane collected multiple times, and then the current lane is evenly segmented, that is, the lane centerline is generated in units of segmented lanes. The target lane is divided into multiple lanes, and the lanes are classified into the current lane and the next lane according to the processing sequence. The current lane is segmented into multiple evenly segmented lanes, so as to generate the first lane centerline of the segmented lane.

[0060] In this embodiment, the segmentation of the current lane can be performed in a way of average length. Specifically, the segmentation interval of the current lane can be determined according to the lane length in the lane data of the current lane, the current lane collected multiple times is evenly segmented, and the initial centerlines corresponding to multiple segmented lanes are generated. Then, the initial centerlines of the segmented lanes divided after each collection are fitted to generate the first lane centerline corresponding to the segmented lane.

[0061] It should be noted that due to multiple collections when collecting the lane, for the same segmented lane in reality, after multiple collections and divisions, there will be multiple segmented lane data with deviations. Therefore, the initial centerlines of the same segmented lane collected multiple times are fitted, so that the error of the first lane centerline of the generated segmented lane is smaller and the accuracy is higher. Among them, the first lane centerline can be generated by any one of the production methods of the lane centerline, which will not be elaborated here one by one.

[0062] Step 102: Connect the first lane centerlines corresponding to the segmented lanes of the current lane to generate the second lane centerline of the current lane.

[0063] In the embodiment of the present invention, the current lane is segmented into multiple segmented lanes. After determining the first lane centerlines corresponding to the segmented lanes, the first lane centerlines corresponding to the multiple segmented lanes of the current lane are connected to generate the second lane centerline of the current lane. Among them, the second lane centerline is generated by connecting the first lane centerlines.

[0064] Specifically, since the lengths of each segmented lane are determined, the lengths of the generated first lane centerlines are also determined. Then, the two end points of the first lane centerline to be connected can be identified, that is, the starting point parameters and ending point parameters of the first lane centerline corresponding to the segmented lane are obtained. According to the starting point parameters and ending point parameters of the first lane centerline, the first lane centerlines of multiple segmented lanes are connected, and the connected first lane centerline is determined as the second lane centerline of the current lane. It should be noted that the starting point parameters and ending point parameters can be the position coordinates of the two end points of the first lane centerline, which are used to connect the first lane centerlines in sequence according to the position of the lane centerline, so as to generate the second lane centerline of the current lane.

[0065] It should be noted that if there are uneven connection traces between the connection end points of the first lane centerline, in this embodiment, a loss function or other repair tools can also be used to adjust the connected first lane centerline to reduce the unevenness of the connection end points of the first lane centerline, so as to improve the smoothness of the second lane centerline of the current lane, which will not be elaborated here.

[0066] Step 103: According to the pre-acquired topological positions of the current lane and the next lane, connect the second lane centerlines of the current lane and the next lane and perform smoothing processing to generate the target lane centerline of the target lane.

[0067] In the embodiment of the present invention, the target lane is divided into multiple lanes, which are distinguished as the current lane and the next lane, that is, the target lane is composed of multiple lanes. To ensure the accuracy and smoothness of the lane centerline of the finally determined target lane, it is necessary to consider the actual positions of the lanes divided in the target lane, and use this position relationship as a reference to fit the second lane centerlines of each lane. Therefore, the topological positions of the current lane and the next lane can be pre-acquired, and according to the topological positions of the current lane and the next lane, the second lane centerlines of the current lane and the next lane are connected and smoothed to generate the target lane centerline of the target lane.

[0068] Specifically, based on the actual position of the target lane, obtain the topological positions of the divided current lane and the next lane, determine the association relationship between the second lane centerlines corresponding to the current lane and the next lane respectively, connect the second lane centerlines of the current lane and the next lane according to the association relationship, and perform smoothing processing on the connected second lane centerlines to generate the target lane centerline of the target lane. Among them, the association relationship of the second lane centerlines is used to represent the position association between the current lane and the next lane, so that the connection of the second lane centerlines conforms to the actual lane logic and improves the accuracy of the lane centerline.

[0069] Exemplarily, referring to Figure 6 shows a schematic scenario of a method for determining a lane centerline provided by an embodiment of the present invention Figure 1 Taking the example that the target lane is divided into three segments, the initial centerlines of the same segmented lane collected multiple times are clustered and fitted to generate the first lane centerlines of each segmented lane, the first lane centerlines of the multiple segmented lanes are connected and smoothed to obtain the second lane centerline of the current lane. Finally, after traversing and obtaining all the second lane centerlines of the current lane, according to the pre-acquired topological positions of the current lane and the next lane, the second lane centerlines of the current lane and the next lane are connected and smoothed to generate the target lane centerline of the target lane.

[0070] The lane centerline determination method provided by the embodiment of the present invention obtains the current lane collected multiple times, segments the current lane to obtain multiple segmented lanes, generates the first lane centerline corresponding to the segmented lanes, connects the first lane centerlines corresponding to the segmented lanes of the current lane to generate the second lane centerline of the current lane, and according to the pre-acquired topological positions of the current lane and the next lane, connects and smooths the second lane centerlines of the current lane and the next lane to generate the target lane centerline of the target lane. The present invention collects lane data multiple times and adopts a segmented processing method to divide the target lane into small-grained segmented lanes, fits and connects the lane centerlines corresponding to the segmented lanes to obtain the lane centerline of the current lane, and based on the topological positions of the current lane and the next lane, connects and smooths the lane centerlines of the current lane and the next lane to determine the target lane centerline of the entire target lane, effectively improving the accuracy of the lane centerline. By separately calculating the lane centerlines of the small-grained segmented lanes with data collected multiple times at the same position and then fitting, connecting, and smoothing, the production efficiency of the lane centerline and the accuracy of the target lane centerline are effectively improved. Due to the effective improvement of the accuracy of the lane centerline, the accuracy of the high-precision map is further ensured, and the intelligent driving effect of the vehicle is enhanced.

[0071] Further, referring to Figure 2 , Figure 2 is Figure 1 the flowchart of step 101 of the lane centerline determination method provided by the embodiment of the present invention in

[0072] Step 201, pre-divide the target lane at a preset interval to determine the current lane for which the lane centerline is to be generated; wherein, the target lane includes the current lane and the next lane.

[0073] Specifically, to improve the accuracy of the lane centerline, in this embodiment, the target lane is pre-divided at a preset interval to divide out lanes with a smaller granularity, and according to the production order of the lane centerline, the divided lanes are distinguished into the current lane and the next lane, that is, the target lane includes the current lane and the next lane, so as to first determine the current lane for which the lane centerline is to be generated. The present invention embodiment does not make specific limitations on the determination of the next lane and the current lane, and the sequential production order can be adopted, which will not be elaborated here one by one.

[0074] Step 202, perform data collection on the current lane a preset number of times to obtain the lane data of the current lane, where the lane data includes the lane length.

[0075] In the embodiment of the present invention, in order to improve the accuracy of lane data and avoid large deviations in the lane data collected once, this embodiment adopts a method of collecting lane data multiple times for the same section of the lane, so as to perform clustering fitting on the lane centerlines generated from the lane data after each collection. Therefore, before segmenting the lane, this embodiment performs a preset number of data collections on the current lane to obtain the lane data of the current lane. The lane data includes the lane length. According to the lane length of the current lane, the current lane is horizontally segmented, and the originally large-data-processing current lane is segmented into multiple smaller-granularity segmented lanes.

[0076] Step 203: Determine the segmentation interval according to the lane length, uniformly segment the currently collected lane multiple times, and generate initial centerlines corresponding to multiple segmented lanes.

[0077] In this embodiment, the segmentation interval is determined according to the lane length, the currently collected lane is uniformly segmented multiple times, and the average segmentation method is adopted to obtain multiple segmented lanes, thereby generating the initial centerlines corresponding to the segmented lanes. The initial centerline is the lane centerline of the segmented lane, and its generation method will not be elaborated here one by one.

[0078] Step 204: Perform fitting processing on the initial centerlines generated by collecting the segmented lanes multiple times to generate the first lane centerline corresponding to the segmented lanes.

[0079] Specifically, perform fitting processing on the initial centerlines generated by collecting the segmented lanes multiple times to generate the first lane centerline corresponding to the segmented lanes. Specifically, cluster fitting is performed on the initial centerlines according to the positions of the segmented lanes. That is, the initial centerlines of the segmented lanes collected multiple times are fitted into a centerline close to the actual segmented lane to generate the first lane centerline corresponding to the segmented lanes.

[0080] In this embodiment, the target lane is determined by dividing it at a preset interval in advance to obtain the current lane. The current lane is collected a preset number of times, the currently collected lane is uniformly segmented multiple times, and initial centerlines corresponding to multiple segmented lanes are generated. The initial centerlines generated by collecting the segmented lanes multiple times are subjected to fitting processing to obtain the first lane centerline corresponding to the segmented lanes. By separately calculating the lane centerlines of the small-granularity segmented lanes where data is collected multiple times at the same position and then fitting and connecting them, the production efficiency of the lane centerline is effectively improved.

[0081] Further, referring to Figure 4 , Figure 4 is Figure 1 The method flowchart of step 103 of the lane centerline determination method provided in the embodiment of the present invention. Step 102 connecting the segmented lanes corresponding to the first lane centerline of the current lane to generate the second lane centerline of the current lane includes:

[0082] Step 301: Obtain the starting parameters and ending parameters of the first lane centerline corresponding to the segmented lanes.

[0083] It should be noted that, to determine the lane centerline of the current lane, in this embodiment, the method of connecting the lane centerlines of multiple segmented lanes is adopted. Specifically, the two end points of the first lane centerline to be connected can be identified, that is, the starting parameters and ending parameters of the first lane centerline corresponding to the segmented lanes are obtained, so as to connect the first lane centerlines of multiple segmented lanes according to the starting parameters and ending parameters of the first lane centerline. It should be noted that the starting parameters and ending parameters can be the position coordinates of the two end points of the first lane centerline, and are used to connect the first lane centerlines in sequence according to the position of the lane centerline, so as to generate the second lane centerline of the current lane.

[0084] Step 302: Connect the first lane centerlines of multiple segmented lanes according to the starting parameters and ending parameters of the first lane centerline.

[0085] Step 303: Determine the connected first lane centerline as the second lane centerline of the current lane.

[0086] Exemplarily, referring to Figure 7 shows a schematic scenario of a method for determining a lane centerline provided by an embodiment of the present invention. Figure 2 Taking the number of data collection times as 3 times as an example, the preset number of times for data collection of the current lane in actual production can be set to 10 times or any other number, which is used to improve the accuracy of collecting lane data. The specific number is set according to the actual lane complexity or production requirements and is not specifically limited herein. As Figure 7 shown, there are some deviations in the initial centerlines generated by the segmented lanes each time, and they do not completely overlap, that is, it indicates that the initially generated centerlines for a single time are distorted. In the embodiment of the present invention, the method of clustering and fitting after multiple collections is adopted to fit the multiple initial centerlines of the segmented lanes to obtain the first lane centerline of the segmented lanes, and then connect the first lane centerlines of multiple segmented lanes according to the starting parameters and ending parameters of the first lane centerline, and the second lane centerline of the current lane is obtained after connection.

[0087] Furthermore, referring to Figure 4 , Figure 4 is Figure 1 The flowchart of step 103 of the method for determining a lane centerline provided by an embodiment of the present invention in

[0088] Step 401: Determine the association relationship between the current lane and the next lane corresponding to the second lane centerline according to the topological positions of the current lane and the next lane.

[0089] In the embodiment of the present invention, since the target lane is pre-divided into multiple current lanes and next lanes, and the positions of the current lane and the next lane after division reflect the positions of the actual lanes. Therefore, after segmentally generating the lane centerlines of each current lane and next lane, it is necessary to determine the association relationship between the current lane and the next lane corresponding to the second lane centerline according to the topological positions of the current lane and the next lane, so as to connect the lane centerlines based on the association relationship.

[0090] Step 402: Connect the second lane centerlines of the current lane and the next lane according to the association relationship.

[0091] Step 403: Smooth the connected second lane centerlines to generate the target lane centerline of the target lane.

[0092] It should be noted that the embodiment of the present invention does not specifically limit the smoothing process of the lane centerline. The smoothing process for the lane centerline to be smoothed may include: using the association relationship of each lane centerline as the initialization input parameter of the preset smoothing function, and continuously iterating based on the constraint conditions to minimize the function value of the preset smoothing function. Each time an iteration is performed, the positions of the lane centerlines to be optimized are optimized once. Stop iterating until the preset smoothing function converges, and obtain the finally optimized lane centerline. Fit the finally optimized lane centerline with a curve function to obtain the smoothed lane centerline.

[0093] Specifically, step 403 for smoothing the connected second lane centerlines to generate the target lane centerline of the target lane specifically includes the following steps:

[0094] First, use the preset smoothing function to perform iterative processing on the connection positions of the second lane centerlines to determine the lane center point of the target lane;

[0095] Second, fit the lane center point to generate the target lane centerline of the target lane.

[0096] It should be noted that in the above steps, use the initialization input parameter of the preset smoothing function, continuously iterate to minimize the function value of the preset smoothing function. Each time an iteration is performed, determine the lane center point of the target lane to be optimized, fit the lane center point, and stop iterating until the preset smoothing function converges to obtain the target lane centerline of the smoothed target lane. Among them, the preset smoothing function can be any kind of smoothing function to automatically and real-time complete the smoothing process and obtain accurate and reasonable lane centerlines.

[0097] Reference Figure 5 , which shows a flowchart of steps of another method for determining a lane center line provided by an embodiment of the present invention. This method is basically the same as the method for determining a lane center line provided by the first embodiment of the present invention. The difference is that the method may further include:

[0098] Step 101: Obtain the current lane collected multiple times, segment the current lane to obtain multiple segmented lanes, and generate a first lane center line corresponding to the segmented lanes.

[0099] Step 102: Connect the first lane center lines corresponding to the segmented lanes of the current lane to generate a second lane center line of the current lane.

[0100] Step 104: In response to the determination of the second lane center line of the current lane, sequentially determine the second lane center lines of the next lanes in the target lane.

[0101] In the embodiment of the present invention, in response to the determination of the second lane center line of the current lane, the second lane center lines of the next lanes in the target lane are sequentially determined. Among them, the generation method of the second lane center line of the next lane is basically the same as the generation method of the second lane center line of the current lane. Specifically, by obtaining the next lane collected multiple times, segmenting the next lane to obtain multiple segmented lanes, generating a first lane center line corresponding to the segmented lanes, and connecting the first lane center lines corresponding to the segmented lanes of the next lane, a second lane center line of the next lane is generated.

[0102] Step 103: According to the pre-obtained topological positions of the current lane and the next lane, connect and smooth the second lane center lines of the current lane and the next lane to generate a target lane center line of the target lane.

[0103] The above steps 101 to 103 are as described in the previous text and will not be elaborated here.

[0104] On the basis of achieving the beneficial effects of the first embodiment, in the embodiment of the present invention, in response to the determination of the second lane center line of the current lane, the second lane center lines of the next lanes in the target lane are sequentially determined, that is, each divided lane of the target lane is iteratively traversed. After determining the lane center lines of each current lane and the next lane, the second lane center lines of the current lane and the next lane are connected and smoothed to generate a target lane center line of the target lane, effectively improving the accuracy of the lane center line. By separately calculating the lane center lines of the small-grained segmented lanes obtained by collecting data multiple times at the same position and then fitting, connecting, and smoothing, the production efficiency of the lane center line and the accuracy of the target lane center line are effectively improved. Due to the effective improvement of the accuracy of the lane center line, the accuracy of the high-precision map is further ensured, and the vehicle intelligent driving effect is improved.

[0105] Reference Figure 8 , which shows a structural block diagram of a lane centerline determination device provided by an embodiment of the present invention. The device may include:

[0106] An acquisition data module 501, configured to acquire the current lane collected multiple times, segment the current lane to obtain multiple segmented lanes, and generate a first lane centerline corresponding to the segmented lanes;

[0107] A first generation module 502, configured to connect the first lane centerlines corresponding to the segmented lanes of the current lane to generate a second lane centerline of the current lane;

[0108] A second generation module 503, configured to connect and smooth the second lane centerlines of the current lane and the next lane according to the pre-acquired topological positions of the current lane and the next lane, and generate a target lane centerline of the target lane.

[0109] Further, the acquisition data module 501 includes:

[0110] A division sub-module, configured to pre-divide the target lane at a preset interval to determine the current lane for which the lane centerline is to be generated; wherein, the target lane includes the current lane and the next lane;

[0111] An acquisition sub-module, configured to perform data acquisition on the current lane a preset number of times to acquire lane data of the current lane, where the lane data includes the lane length;

[0112] A segmentation sub-module, configured to determine a segmentation interval according to the lane length, uniformly segment the current lane collected multiple times, and generate an initial centerline corresponding to the multiple segmented lanes;

[0113] A first generation sub-module, configured to perform fitting processing on the initial centerlines generated by the multiple segmented lanes to generate a first lane centerline corresponding to the segmented lanes.

[0114] Further, the first generation module 502 includes:

[0115] A second acquisition sub-module, configured to acquire start parameters and end parameters of the first lane centerline corresponding to the segmented lanes;

[0116] A first connection sub-module, configured to connect the first lane centerlines of the multiple segmented lanes according to the start parameters and end parameters of the first lane centerline;

[0117] A first determination sub-module, configured to determine the connected first lane centerline as the second lane centerline of the current lane.

[0118] Further, the second generation module 503 includes:

[0119] A second determination sub-module, configured to determine the association relationship between the current lane and the second lane centerline corresponding to the current lane and the next lane according to the topological positions of the current lane and the next lane;

[0120] A second connection sub-module, configured to connect the second lane centerlines of the current lane and the next lane according to the association relationship;

[0121] A second generation sub-module, configured to smooth the connected second lane centerlines to generate the target lane centerline of the target lane.

[0122] Further, the second generation sub-module includes:

[0123] A processing unit, configured to iteratively process the connection positions of the second lane centerlines by using a preset smoothing function to determine the lane center points of the target lane;

[0124] A fitting unit, configured to fit the lane center points to generate the target lane centerline of the target lane.

[0125] Further, the device further includes:

[0126] A determination module, configured to, in response to the determination of the second lane centerline of the current lane, sequentially determine the second lane centerlines of the next lanes in the target lane.

[0127] The lane centerline determination device provided by the embodiment of the present invention obtains the current lane collected multiple times, segments the current lane to obtain multiple segmented lanes, generates a first lane centerline corresponding to the segmented lanes, connects the first lane centerlines corresponding to the segmented lanes of the current lane, generates a second lane centerline of the current lane, and connects and smooths the second lane centerlines of the current lane and the next lane according to the pre-acquired topological positions of the current lane and the next lane to generate the target lane centerline of the target lane. By collecting lane data multiple times and adopting a segmented processing method, the present invention divides the target lane into small-grained segmented lanes, fits and connects the lane centerlines corresponding to the segmented lanes to obtain the lane centerline of the current lane, and connects and smooths the lane centerlines of the current lane and the next lane according to the topological positions of the current lane and the next lane to determine the target lane centerline of the entire target lane, effectively improving the accuracy of the lane centerline. By separately calculating the lane centerlines of the small-grained segmented lanes with data collected multiple times at the same position and then fitting, connecting and smoothing, the production efficiency of the lane centerline and the accuracy of the target lane centerline are effectively improved. Due to the effective improvement of the accuracy of the lane centerline, the accuracy of the high-precision map is further ensured, and the intelligent driving effect of the vehicle is improved.

[0128] Referring to Figure 9 , the embodiment of the present invention also provides an electronic device, as Figure 9 shown, including a processor 601, a communication interface 602, a memory 603, and a communication bus 604. Among them, the processor 601, the communication interface 602, and the memory 603 communicate with each other through the communication bus 604.

[0129] The memory 603 is used to store a computer program;

[0130] When the processor 601 is used to execute the program stored on the memory 603, the following steps are implemented:

[0131] Obtain the current lane collected multiple times, segment the current lane to obtain multiple segmented lanes, and generate a first lane centerline corresponding to the segmented lanes;

[0132] Connect the first lane centerlines corresponding to the segmented lanes of the current lane to generate a second lane centerline of the current lane;

[0133] According to the pre-acquired topological positions of the current lane and the next lane, connect and smooth the second lane centerlines of the current lane and the next lane to generate the target lane centerline of the target lane.

[0134] The communication bus mentioned in the above terminal may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. This communication bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of convenience of representation, only a thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0135] The communication interface is used for communication between the above terminal and other devices.

[0136] The memory may include a Random Access Memory (RAM), or may also include a non-volatile memory, such as at least one disk memory. Optionally, the memory may also be at least one storage device located far from the aforementioned processor.

[0137] The above-mentioned processor may be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0138] In another embodiment provided by the present invention, a computer-readable storage medium is also provided. Instructions are stored in the computer-readable storage medium, and when it runs on a computer, it causes the computer to execute the lane centerline determination method described in any one of the above embodiments.

[0139] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present invention are generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid state disk (SSD)).

[0140] It should be noted that, in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including", or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, article, or device that includes a series of elements includes not only those elements but also other elements that are not explicitly listed, or further includes elements that are inherent to such process, method, article, or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, article, or device that includes the element.

[0141] Each embodiment in this specification is described in a related manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method embodiment.

[0142] The above are only the preferred embodiments of the present invention and are not intended to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention are all included within the protection scope of the present invention.

Claims

1. A method for determining a lane center line, characterized in that, The method includes: Obtaining the current lane collected multiple times, segmenting the current lane to obtain a plurality of segmented lanes, and generating a first lane centerline corresponding to the segmented lanes; Connecting the segmented lanes of the current lane corresponding to the first lane centerline to generate a second lane centerline of the current lane; According to the topological positions of the current lane and the next lane obtained in advance, connecting the second lane centerlines of the current lane and the next lane and performing smoothing processing to generate a target lane centerline of the target lane.

2. The method according to claim 1, characterized in that, The obtaining the current lane collected multiple times, segmenting the current lane to obtain a plurality of segmented lanes, and generating a first lane centerline corresponding to the segmented lanes includes: Pre-dividing the target lane at a preset interval to determine the current lane for which the lane centerline is to be generated; wherein the target lane includes the current lane and the next lane; Performing data collection on the current lane a preset number of times to obtain lane data of the current lane, wherein the lane data includes the lane length; Determining a segmentation interval according to the lane length, uniformly segmenting the current lane collected multiple times, and generating initial centerlines corresponding to the plurality of segmented lanes; Performing fitting processing on the initial centerlines generated by collecting the segmented lanes multiple times to generate a first lane centerline corresponding to the segmented lanes.

3. The method according to claim 1, characterized in that, The connecting the segmented lanes of the current lane corresponding to the first lane centerline to generate a second lane centerline of the current lane includes: Obtaining the starting point parameters and ending point parameters of the first lane centerline corresponding to the segmented lanes; Connecting the first lane centerlines of the plurality of segmented lanes according to the starting point parameters and ending point parameters of the first lane centerline; Determining the connected first lane centerline as the second lane centerline of the current lane.

4. The method according to claim 1, characterized in that, The connecting the second lane centerlines of the current lane and the next lane according to the topological positions of the current lane and the next lane and performing smoothing processing to generate a target lane centerline of the target lane includes: Determining the association relationship between the second lane centerlines corresponding to the current lane and the next lane according to the topological positions of the current lane and the next lane; Connecting the second lane centerlines of the current lane and the next lane according to the association relationship; Performing smoothing processing on the connected second lane centerline to generate a target lane centerline of the target lane.

5. The method according to claim 4, characterized in that, The performing smoothing processing on the connected second lane centerline to generate a target lane centerline of the target lane includes: Performing iterative processing on the connection position of the second lane centerline using a preset smoothing function to determine the lane center point of the target lane; Performing fitting on the lane center point to generate a target lane centerline of the target lane.

6. The method according to claim 1, characterized in that, Before the connecting the second lane centerlines of the current lane and the next lane according to the topological positions of the current lane and the next lane and performing smoothing processing to generate a target lane centerline of the target lane, it further includes: In response to the determination of the second lane center line of the current lane, the second lane center line of the next lane in the target lane is determined in sequence.

7. A device for determining a lane center line, characterized in that, The device includes: A data acquisition module, configured to acquire the current lane collected multiple times, segment the current lane to obtain a plurality of segmented lanes, and generate a first lane center line corresponding to the segmented lanes; A first generation module, configured to connect the first lane center lines corresponding to the segmented lanes of the current lane to generate a second lane center line of the current lane; A second generation module, configured to connect and smooth the second lane center lines of the current lane and the next lane according to the pre-acquired topological positions of the current lane and the next lane, and generate a target lane center line of the target lane.

8. The device according to claim 7, characterized in that, The data acquisition module includes: A division sub-module, configured to pre-divide the target lane at a preset interval to determine the current lane for which the lane center line is to be generated; wherein the target lane includes the current lane and the next lane; An acquisition sub-module, configured to perform data acquisition on the current lane a preset number of times to acquire lane data of the current lane, wherein the lane data includes the lane length; A segmentation sub-module, configured to determine a segmentation interval according to the lane length, uniformly segment the current lane collected multiple times, and generate initial center lines corresponding to the plurality of segmented lanes; A first generation sub-module, configured to perform fitting processing on the initial center lines generated from the segmented lanes collected multiple times to generate a first lane center line corresponding to the segmented lanes.

9. An electronic device, characterized in that, It includes: A processor; A memory for storing instructions executable by the processor; Wherein, the processor is configured to execute the instructions to implement the lane center line determination method according to any one of claims 1 to 6.

10. A readable storage medium, characterized in that, A computer program is stored on the readable storage medium, and when the computer program is executed by the processor, the lane center line determination method according to any one of claims 1 to 6 is implemented.

Images