Real-time map generation method and device, electronic equipment and storage medium

Abstract

The application provides a real-time map generation method and device, electronic equipment and storage medium; wherein the method comprises: acquiring environment map information and real-time position information detected by an autonomous vehicle in real time; based on the real-time position information, determining at least one target reachable key point of the autonomous vehicle at the real-time position through a preset first-level map, and extracting target map elements within a target map element indication range of the at least one target reachable key point from a preset second-level map to obtain target map information; and fusing the environment map information and the target map information to obtain real-time map information. The method reduces the running space occupied by real-time map generation through a hierarchical map mechanism.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic driving, and in particular to a real-time map generation method and device, electronic equipment and a storage medium. BACKGROUND

[0002] In the technical field of automatic driving, the map required for navigation has a high accuracy requirement. An offline map generated by manual annotation is difficult to ensure consistency with the real-time driving environment. Therefore, the navigation map required for automatic driving is usually generated in real time.

[0003] The existing real-time map generation technology usually fuses the environment map detected by the camera with the offline map in real time. However, since the offline map occupies a large running space, there is a technical problem of occupying too much running space when fusing with the environment map. SUMMARY

[0004] Therefore, the purpose of the present application is to provide a real-time map generation method, device, electronic equipment and storage medium to reduce the running space occupied by real-time map generation.

[0005] In a first aspect, the present application provides a real-time map generation method, which comprises: acquiring environment map information and real-time position information detected by an automatic driving vehicle in real time; determining at least one target reachable key point of the automatic driving vehicle at the real-time position based on the real-time position information through a preset first-level map, and extracting target map elements within the target reachable key point indication range from a preset second-level map to obtain target map information; and fusing the environment map information and the target map information to obtain real-time map information.

[0006] Optionally, the environment map information comprises real-time map elements; and the step of fusing the environment map information and the target map information to obtain real-time map information comprises: comparing the real-time map elements and the target map elements to obtain a comparison result, the comparison result being used to indicate whether the target map information has changed; and if the comparison result indicates that the target map information has changed, fusing the environment map information and the target map information through the real-time map elements and the target map elements to obtain real-time map information.

[0007] Optionally, the real-time map element and the target map element comprise lane line elements; and the comparing the real-time map element and the target map element to obtain a comparison result comprises: acquiring a confidence degree corresponding to a lane line element in the real-time map element; comparing the lane line elements in the real-time map element and the target map element, and determining whether the lane line element in the target map information is changed according to the confidence degree to obtain the comparison result.

[0008] Optionally, the comparing the real-time map element and the target map element to obtain a comparison result comprises: judging whether a deviation between a current frame lane line element and at least one historical frame lane line element is greater than a preset deviation threshold, the current frame lane line element being used to indicate a lane line element in a real-time map element of a current frame, and the historical frame lane line element being used to indicate a lane line element in a real-time map element of a historical frame; if the deviation between the current frame lane line element and the at least one historical frame lane line element is greater than the preset deviation threshold, determining a lane line element in a real-time map element of a previous frame as a lane line element in a real-time map element of the current frame; and comparing the lane line element in the real-time map element of the current frame and a lane line element in the target map element to obtain the comparison result.

[0009] Optionally, the real-time map element and the target map element comprise lane center line elements; and the fusing the environment map information and the target map information to obtain real-time map information comprises: sampling lane center line elements at overlapping and non-overlapping positions of the environment map information and the target map information to obtain overlapping position sampling results and non-overlapping position sampling results; performing curve fitting on the overlapping position sampling results and the non-overlapping position sampling results to obtain fitting results; and performing lane center line smoothing update on the target map information according to the fitting results to obtain the real-time map information.

[0010] Optionally, after the fusing the environment map information and the target map information to obtain real-time map information, the method further comprises: performing road curvature degree calculation on lane line elements outside a preset distance in the real-time map information according to the real-time position information to obtain a target road curvature degree; judging whether the target road curvature degree is less than a preset degree threshold; and if the target road curvature degree is less than the preset degree threshold, performing road shape point extraction on the lane line elements outside the preset distance in the real-time map information, and removing road indication points outside the extracted road shape points in the real-time map information to obtain compressed real-time map information.

[0011] Optionally, the acquiring the environment map information and the real-time position information detected by the autonomous vehicle in real time comprises: acquiring the real-time position information and real-time environment data collected by at least one environment perception device on the autonomous vehicle in real time during the autonomous driving of the autonomous vehicle, and generating an environment map element through the real-time environment data; performing lane center line detection on the environment map element through a preset road center line detection model to obtain a real-time detected lane center line element; and combining the environment map element and the real-time detected lane center line element to obtain the environment map information.

[0012] Optionally, the determining the at least one target reachable key point at the real-time position of the autonomous vehicle based on the real-time position information through the preset first-level map comprises: performing road reachability search on the real-time position of the autonomous vehicle through the preset first-level map based on the real-time position information to obtain all reachable key points at the real-time position of the autonomous vehicle, wherein the preset first-level map comprises key point information at all lane branch points and key point information at preset lane spacings, and the key point information comprises key point position information; and performing key point screening of a preset reachability distance on all the reachable key points to obtain the at least one target reachable key point at the real-time position of the autonomous vehicle.

[0013] Optionally, the extracting the target map element in the target map element indication range of the at least one target reachable key point from the preset second-level map to obtain the target map information comprises: judging whether the preset range of the autonomous vehicle contains the at least one target reachable key point; if the preset range of the autonomous vehicle does not contain the at least one target reachable key point, extracting map information in a range of a nearest reachable key point from the preset second-level map to obtain the target map information, wherein the nearest reachable key point is used to indicate a target reachable key point closest to the preset range; and if the preset range of the autonomous vehicle contains the at least one target reachable key point, extracting map information in the preset range from the preset second-level map to obtain the target map information.

[0014] In a second aspect, an embodiment of the present application provides a real-time map generation device, which comprises: an acquisition module configured to acquire environment map information and real-time position information detected by an autonomous vehicle in real time; an extraction module configured to determine at least one target reachable key point at a real-time position of the autonomous vehicle based on the real-time position information through a preset first-level map, and extract a target map element in a target map element indication range of the at least one target reachable key point from a preset second-level map to obtain target map information; and a fusion module configured to fuse the environment map information and the target map information to obtain real-time map information.

[0015] Optionally, the environment map information comprises real-time map elements; the fusion module comprises: a comparison unit, configured to compare the real-time map elements and the target map elements to obtain a comparison result, the comparison result being used to indicate whether the target map information is changed; and a fusion unit, configured to fuse the environment map information and the target map information through the real-time map elements and the target map elements to obtain real-time map information, if the comparison result indicates that the target map information is changed.

[0016] Optionally, the real-time map elements and the target map elements comprise lane line elements; the comparison unit is further configured to: acquire a confidence degree corresponding to a lane line element in the real-time map elements; compare the lane line elements in the real-time map elements and the target map elements, and determine whether the lane line elements in the target map information are changed according to the confidence degree to obtain the comparison result.

[0017] Optionally, the comparison unit is further configured to: determine whether a deviation between a current frame lane line element and at least one historical frame lane line element is greater than a preset deviation threshold, the current frame lane line element being used to indicate a lane line element in a real-time map element of a current frame, and the historical frame lane line element being used to indicate a lane line element in a real-time map element of a historical frame; if the deviation between the current frame lane line element and the at least one historical frame lane line element is greater than the preset deviation threshold, determine a lane line element in a real-time map element of a previous frame as a lane line element in a real-time map element of the current frame; and compare the lane line element in the real-time map element of the current frame and a lane line element in the target map elements to obtain the comparison result.

[0018] Optionally, the real-time map elements and the target map elements comprise lane center line elements; the fusion module is further configured to: sample lane center line elements at an overlapping position and a non-overlapping position of the environment map information and the target map information to obtain an overlapping position sampling result and a non-overlapping position sampling result; perform curve fitting of the lane center line elements on the overlapping position sampling result and the non-overlapping position sampling result, and perform lane center line smoothing update on the target map information through a fitting result to obtain real-time map information.

[0019] Optionally, the device further comprises a calculation module configured to calculate a road curvature degree of the lane line element beyond the preset distance in the real-time map information according to the real-time position information, to obtain a target road curvature degree; a judgment module configured to judge whether the target road curvature degree is less than a preset degree threshold; and a compression module configured to, if the target road curvature degree is less than the preset degree threshold, extract road shape points of the lane line element beyond the preset distance in the real-time map information, and remove road indicating points beyond the extracted road shape points in the real-time map information, to obtain compressed real-time map information.

[0020] Optionally, the acquisition module is further configured to acquire real-time position information and real-time environment data collected by at least one environment perception device on the autonomous vehicle in an autonomous driving process of the autonomous vehicle, and generate environment map elements through the real-time environment data; perform lane center line detection on the environment map elements through a preset road center line detection model, to obtain real-time detected lane center line elements; and combine the environment map elements and the real-time detected lane center line elements, to obtain environment map information.

[0021] Optionally, the extraction module is further configured to perform road reachability search on a real-time position where the autonomous vehicle is located through a preset first-level map based on the real-time position information, to obtain all reachable key points of the autonomous vehicle at the real-time position, the preset first-level map comprising key point information at all lane branch points and key point information at preset lane spacings, the key point information comprising key point position information; and perform key point screening of a preset reachability distance on all reachable key points, to obtain at least one target reachable key point of the autonomous vehicle at the real-time position.

[0022] Optionally, the extraction module is further configured to judge whether the at least one target reachable key point is contained in a preset range of the autonomous vehicle; if the at least one target reachable key point is not contained in the preset range of the autonomous vehicle, extract map information within a range of a nearest reachable key point from a preset second-level map, to obtain target map information, the nearest reachable key point being used to indicate a target reachable key point closest to the preset range; and if the at least one target reachable key point is contained in the preset range of the autonomous vehicle, extract map information within the preset range from the preset second-level map, to obtain target map information.

[0023] In a third aspect, an electronic device is provided, which comprises a processor and a memory, the memory storing machine executable instructions capable of being executed by the processor, and the processor executes the machine executable instructions to implement the real-time map generation method.

[0024] In a fourth aspect, an embodiment of the present application provides a computer readable storage medium, which stores computer executable instructions. When the computer executable instructions are invoked and executed by a processor, the computer executable instructions cause the processor to implement the real-time map generation method.

[0025] The embodiments of the present application bring the following beneficial effects:

[0026] The real-time map generation method, device, electronic equipment and storage medium described above first acquire real-time position information of the autonomous vehicle, determine at least one target reachable key point from a preset first-level map according to the real-time position information, then extract map elements within an indication range of the at least one target reachable key point from a preset second-level map to obtain offline target map information, and finally fuse the target map information with environment map information detected by the autonomous vehicle in real time to obtain real-time map information. The method reduces the running space occupied by real-time map generation through a hierarchical map mechanism.

[0027] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the present application. The objects and other advantages of the present application will be realized and achieved by the structure particularly pointed out in the description, claims and drawings.

[0028] In order to make the above-mentioned objects, features and advantages of the present application more apparent, the following will specifically describe a preferred embodiment, and the accompanying drawings will be described in detail as follows. BRIEF DESCRIPTION OF DRAWINGS

[0029] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the following will briefly introduce the drawings needed to be used in the specific embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0030] Figure 1 An embodiment flow chart of the real-time map generation method in the embodiment of the present application;

[0031] Figure 2 Another embodiment flow chart of the real-time map generation method in the embodiment of the present application;

[0032] Figure 3 Another embodiment flow chart of the real-time map generation method in the embodiment of the present application;

[0033] Figure 4 A schematic diagram of a real-time map generation device provided by the embodiment of the present application;

[0034] Figure 5 A schematic diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0035] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions of the present application will be clearly and completely described below with reference to the drawings. Obviously, the described embodiments are only some of the embodiments of the present application, but not all of the embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0036] The terms “first”, “second”, “third”, “fourth” and the like (if any) in the description, claims and above drawings of the present application are used to distinguish similar objects, and do not necessarily indicate a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms “include” or “have” and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not necessarily have to include only those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to the process, method, product or device.

[0037] For the sake of understanding, the specific flow of the embodiments of the present application will be described below. Please refer to Figure 1 One embodiment of the real-time map generation method in the embodiments of the present application includes:

[0038] Step S10, acquiring environment map information and real-time position information detected by the autonomous vehicle in real time;

[0039] It should be noted that in the process of autonomous driving of the autonomous vehicle, the autonomous vehicle will collect environment information in real time, and generate environment map information according to the environment information collected in real time. The environment map information detected by the autonomous vehicle in real time is also called online map information. Different from offline map information, the online map information has real-time performance, can reflect the actual driving road conditions of the autonomous vehicle, and can improve the real-time performance of the navigation map generation and the safety of the autonomous driving when used for real-time navigation of the autonomous driving.

[0040] In an embodiment, the environment information collected by the environment perception device installed on the autonomous vehicle is used to generate the environment map information detected by the autonomous vehicle in real time. The environment perception device can include, but is not limited to, at least one of a camera, a laser radar, an ultrasonic radar, a millimeter wave radar, and a laser scanner, without limitation. As an example but not limitation, the real-time environment image collected by the camera installed on the autonomous vehicle and the real-time point cloud collected by the laser radar are obtained, and the environment map information detected by the autonomous vehicle in real time is generated based on the real-time environment image and the real-time point cloud.

[0041] It can be understood that the environment map information is used to indicate the road information in the real-time driving environment of the autonomous vehicle. In an embodiment, the environment map information includes real-time map elements, wherein the real-time map elements include, but are not limited to, at least one of lane lines, curbs, guardrails, lane centerlines, traffic lights, traffic signs, sidewalk lines, stop lines, and other road elements, without limitation.

[0042] In the embodiment, the real-time position information can be obtained by a positioning device / system of the autonomous vehicle. For example, the positioning system can be a Global Navigation Satellite System (GNSS), a GNSS combined with an inertial motion unit (IMU), or the like, without limitation.

[0043] In step S20, based on the real-time position information, at least one target reachable key point of the autonomous vehicle at the real-time position is determined by using a preset first-level map, and target map elements within a range of the at least one target reachable key point are extracted from a preset second-level map to obtain target map information.

[0044] It should be noted that the preset first-level map and the preset second-level map are both offline maps, that is, maps that are not detected in real time. It can be understood that, in order to reduce the running memory space occupied by real-time map generation, the embodiment adopts a hierarchical map mechanism, and specifically divides the map into a preset first-level map and a preset second-level map. The preset second-level map is a complete high-precision map including relevant information of all map elements, and the preset first-level map only contains key point information extracted from the complete high-precision map. In an embodiment, the preset first-level map and the preset second-level map can both be a map in at least one preset administrative region, for example, a map of Shenzhen, a map of Guangdong Province, or a map of Guangdong Province and Guangxi Zhuang Autonomous Region, and the like, and the specific embodiments are not limited herein. Since the preset first-level map only contains key point information, the memory space occupied is small, and the preset first-level map can be resident in the memory. Since the memory space occupied by the high-precision map is too large, the embodiment extracts target map elements in a target reachable key point indication range from the preset second-level map after determining at least one target reachable key point of the real-time position of the autonomous vehicle according to the preset first-level map, loads the target map elements into the memory to obtain target map information, and uses the target map information for real-time map generation and real-time navigation, thereby avoiding the technical problem of high running space occupation caused by full loading of the high-precision map.

[0045] In the embodiment, first, at least one target reachable key point of the real-time position of the autonomous vehicle is determined from the preset first-level map according to the real-time position of the autonomous vehicle. Then, target map elements are extracted from the preset second-level map according to a map range indicated by all target reachable key points, so that the high-precision preset second-level map only needs to be incrementally loaded into the memory to obtain target map information. The target map information contains relevant information of all map elements in the map range indicated by all target reachable key points, and is used for subsequent fusion with environmental map information. The embodiment can avoid high running space occupation caused by full loading of the high-precision map, and also ensure the accuracy of real-time map generation.

[0046] It can be understood that the key point information in the preset first-level map refers to information of road key positions, wherein the road key positions can be intersection positions, positions of changes in driving directions (such as positions of turnable positions and positions of U-turn positions), positions of changes in driving lanes (such as positions of changes from single-lane to multi-lane and positions of changes from multi-lane to single-lane), and the like, which are not limited herein. It should be noted that the key point information includes position information of the key points. In an embodiment, the target reachable key point refers to a nearest reachable key point of the autonomous vehicle at a real-time position. In another embodiment, the at least one target reachable key point refers to reachable key points of the autonomous vehicle within a preset range of the real-time position. It can be understood that the reachable key point is related to road connectivity, and a road key point having road connectivity with the real-time position of the autonomous vehicle can be determined as a reachable key point.

[0047] In step S30, the environment map information and the target map information are fused to obtain real-time map information.

[0048] In this step, the environment map information is fused into the target map information to obtain the real-time map information. It can be understood that the environment map information has real-time performance, which can make up for the low real-time performance of the target map information extracted from the offline map. Therefore, by fusing the environment map information and the target map information, the real-time map information for autonomous driving navigation can be obtained.

[0049] In an embodiment, when the environment map information and the target map information are fused, first, difference detection is performed on the environment map information and the target map information to obtain a difference detection result, and then the target map information is corrected in difference according to the difference detection result by using the environment map information to obtain the real-time map information. This embodiment can correct information errors of the offline map due to untimely updates, thereby improving the accuracy of the autonomous driving navigation map.

[0050] In an embodiment, after step S30, the preset first-level map and / or the preset second-level map can also be updated by using the real-time map information to obtain an updated preset first-level map and / or an updated preset second-level map. Specifically, the preset first-level map and / or the preset second-level map can also be updated according to a summary analysis result obtained by summarizing and analyzing the real-time map information collected within a certain time length to obtain an updated preset first-level map and / or an updated preset second-level map. This embodiment can update the offline map based on the online map, improve the real-time performance of the map, and reduce the cost required for manual annotation for generating the offline map.

[0051] The real-time map generation method provided by the above embodiment first acquires real-time position information of the autonomous vehicle, determines at least one target reachable key point at the real-time position of the autonomous vehicle according to the real-time position information through a preset first-level map, extracts map elements in a range indicated by the at least one target reachable key point from a preset second-level map to obtain offline target map information, and finally fuses the target map information with environment map information detected by the autonomous vehicle in real time to obtain real-time map information. The method reduces the running space occupied by real-time map generation through a hierarchical map mechanism.

[0052] Referring to Figure 2 Another embodiment of the real-time map generation method in the embodiment of the present application includes:

[0053] In step S201, environment map information and real-time position information detected by the autonomous vehicle in real time are acquired, and the environment map information includes real-time map elements.

[0054] It should be noted that the environment map information includes map elements in the environment at the real-time position of the autonomous vehicle. It can be understood that the map elements in the environment and the map elements in the preset second-level map include the same types, for example, the preset second-level map includes map elements of lane line types, and then the map elements in the environment can also detect map elements of lane line types, which is not limited here.

[0055] In step S202, at least one target reachable key point of the autonomous vehicle at the real-time position is determined through a preset first-level map based on the real-time position information, and target map elements in a range indicated by the at least one target reachable key point are extracted from a preset second-level map to obtain target map information.

[0056] In one embodiment, the above method of determining at least one target reachable key point of the autonomous vehicle at the real-time position through a preset first-level map based on the real-time position information includes: performing road reachability search on the real-time position of the autonomous vehicle through the preset first-level map based on the real-time position information to obtain all reachable key points of the autonomous vehicle at the real-time position, the preset first-level map including key point information at all lane branch points and key point information at preset lane spacings, and the key point information including key point position information; and performing key point screening of a preset reachable distance on all the reachable key points to obtain at least one target reachable key point of the autonomous vehicle at the real-time position.

[0057] The embodiment is a determination method of the target reachable key point. Based on the real-time position information, the road reachability of the real-time position of the autonomous vehicle is searched on the preset first-level map, so as to obtain all reachable key points of the autonomous vehicle at the real-time position. In these reachable key points, the key points that are too far away do not need to participate in subsequent calculation. Therefore, only the reachable key points within the preset reachable distance need to be screened, so as to obtain at least one target reachable key point of the autonomous vehicle at the real-time position. It should be noted that the preset first-level map in the embodiment includes key point information at all lane branches and key point information at preset lane intervals. The key points are arranged at all lane branches of the preset first-level map, so that the real-time map generated can contain lane change information, thereby improving the safety of autonomous driving navigation. The key points are arranged at the preset lane intervals of the preset first-level map, so that when there is no lane change in a long distance, that is, when the distance between the key points at the lane branches is too long (for example, for some highways, the distance between the key points at two road branches may be more than 20 km), the target map information extracted is too much, and the memory occupation is increased. As an example but not limitation, the preset lane interval can be set to 2 km. Then, on some roads without too long branch roads, a key point is arranged every 2 km, so that the distance between every two key points will not be too long, thereby ensuring that the target map information loaded into the memory is controlled within a certain range.

[0058] In an embodiment, the above-mentioned extracting the target map element within the range indicated by the at least one target reachable key point from the preset second-level map to obtain the target map information comprises: judging whether the preset range of the autonomous vehicle contains at least one target reachable key point; if the preset range of the autonomous vehicle does not contain at least one target reachable key point, extracting the map information within the range of the nearest reachable key point from the preset second-level map to obtain the target map information, wherein the nearest reachable key point is used to indicate the target reachable key point closest to the preset range; and if the preset range of the autonomous vehicle contains at least one target reachable key point, extracting the map information within the preset range from the preset second-level map to obtain the target map information.

[0059] In this embodiment, in order to further ensure that the target map information loaded into the memory is not too large, and at the same time ensure that the target map information loaded into the memory contains at least one target reachable key point, so as to facilitate the early generation of automatic driving decision. Therefore, when performing target map element extraction, first determine whether the preset range of the real-time position of the autonomous vehicle contains at least one target reachable key point, if yes, extract the map information in the preset range from the preset second-level map, that is, ensure that the target map information loaded into the memory contains at least one target reachable key point; and if the preset range of the real-time position of the autonomous vehicle does not contain at least one target reachable key point, then take the nearest reachable key point in all target reachable key points as the criterion, extract the map information in the range of the nearest reachable key point from the preset second-level map, thereby obtaining the target map information, wherein the nearest reachable key point refers to the target reachable key point closest to the preset range. In this way, it is not necessary to load the target map elements in the range of all target reachable key points into the memory, which ensures that the target map information loaded into the memory is not too large, and at the same time ensures that the target map information loaded into the memory contains at least one target reachable key point.

[0060] Step S203, comparing the real-time map elements and the target map elements to obtain a comparison result, the comparison result being used to indicate whether the target map information is changed;

[0061] In this step, by comparing the real-time map elements and the target map elements, it is determined whether the target map information is changed, that is, whether there is a difference between the preset second-level map and the real-time environment map. If the target map information is changed, it indicates that there is a difference between the preset second-level map and the real-time environment map, and also indicates that there is a difference between the offline map (i.e., the preset second-level map) and the online map (i.e., the real-time environment map), and a subsequent fusion step is needed to correct the deviation between the offline map and the online map. The present embodiment can accurately determine whether the map is changed through the comparison of the map elements.

[0062] In an embodiment, the real-time map element and the target map element include lane line elements; the step S203 includes: obtaining a confidence degree corresponding to the lane line elements in the real-time map element; comparing the lane line elements in the real-time map element and the target map element, and determining whether the lane line elements in the target map information change according to the confidence degree, to obtain a comparison result. It can be understood that the lane line elements are more stable elements in the map than other elements, and therefore, when comparing the real-time map element and the target map element, the comparison can be specifically performed on the lane line elements. If it is determined that the lane line elements in the target map information change, it can be determined that the target map information changes, and vice versa, so that whether to perform subsequent map information fusion is determined, so that the comparison efficiency is improved, and the real-time performance of the map generation is also improved. In the embodiment, the lane line elements in the real-time map element are obtained in real time, and have a corresponding confidence degree. Generally, the confidence degree depends on the accuracy of the algorithm / model used for real-time detection and the real-time road surface environment. For example, when the accuracy of the algorithm / model used is low, the confidence degree of the output lane line elements is also reduced. Similarly, when the real-time road surface environment is poor, such as in rainy weather or in the presence of an obstacle, the confidence degree of the output lane line elements is also reduced. Specifically, when the confidence degree is used to determine whether the lane line elements in the target map information change to obtain the comparison result, it is determined whether the confidence degree is greater than a preset confidence threshold. If yes, it is determined whether the lane line elements in the target map information change according to the lane line elements in the real-time map element. If the confidence degree is less than or equal to the preset confidence threshold, it is indicated that the reference value of the lane line elements in the real-time map element is low, and therefore, the comparison result that the lane line elements in the target map information do not change is directly determined according to the target map information.

[0063] In an embodiment, the step S203 includes: determining whether a deviation between the current frame lane line element and at least one historical frame lane line element is greater than a preset deviation threshold, the current frame lane line element being used to indicate the lane line elements in the real-time map element of the current frame, and the historical frame lane line element being used to indicate the lane line elements in the real-time map element of the historical frame; if the deviation between the current frame lane line element and the at least one historical frame lane line element is greater than the preset deviation threshold, the lane line elements in the real-time map element of the previous frame are determined as the lane line elements in the real-time map element of the current frame; and the lane line elements in the real-time map element of the current frame and the lane line elements in the target map element are compared to obtain a comparison result.

[0064] In this embodiment, when the real-time map elements and the target map elements are compared, the lane line elements in the real-time map elements can adopt the results of multi-frame accumulation, so that the accuracy of the lane line elements is improved. Specifically, first, it is judged whether the deviation between the lane line elements of the current frame and the lane line elements of the previous frame or the lane line elements of the previous several frames (i.e., at least one historical frame lane line element) is greater than a preset deviation threshold. If yes, the lane line elements in the real-time map elements of the previous frame (i.e., the last frame) are determined as the lane line elements in the real-time map elements of the current frame. If no, the current frame lane line elements do not need to be corrected, and the current frame lane line elements are directly determined as the lane line elements in the real-time map elements of the current frame. Based on this, when the comparison is performed, the determined lane line elements in the real-time map elements of the current frame are compared with the lane line elements in the target map elements, and the comparison result is obtained, so that when a frame is not accurately detected, a large deviation is avoided, and the accuracy of the real-time map generation is reduced.

[0065] In step S204, if the comparison result indicates that the target map information is changed, the environment map information and the target map information are fused by the real-time map elements and the target map elements to obtain real-time map information.

[0066] In this step, if the comparison result indicates that the target map is changed, it means that there is a difference between the offline map and the online detected map. Then, the environment map information and the target map information are difference fused according to the real-time map elements and the target map elements to obtain real-time map information. Specifically, the target map elements are difference element corrected according to the difference elements between the real-time map elements and the target map elements to obtain real-time map information. The real-time map information can be used for automatic driving real-time navigation, and has real-time and accuracy.

[0067] The real-time map generation method provided by the above embodiment first acquires real-time position information of an automatic driving vehicle, determines at least one target reachable key point from a preset first-level map according to the real-time position information, extracts map elements in a range indicated by the at least one target reachable key point from a preset second-level map to obtain offline target map information, and finally compares the target map information with map elements in automatic environment map information, and performs map fusion according to a comparison result to obtain real-time map information. The method reduces the running space occupied by real-time map generation through a hierarchical map mechanism, and improves the real-time and accuracy of automatic driving navigation.

[0068] Please refer to Figure 3 Another embodiment of the real-time map generation method in the embodiment of the application includes:

[0069] In the automatic driving process of the autonomous vehicle, real-time position information and real-time environment data collected by at least one environment perception device on the autonomous vehicle are acquired in real time, and an environment map element is generated from the real-time environment data;

[0070] In step S302, a lane center line element is obtained by performing lane center line detection on the environment map element by using a preset road center line detection model.

[0071] In step S303, environment map information is obtained by combining the environment map element and the real-time detected lane center line element.

[0072] The steps S301-S303 are a way of real-time detection of the environment map information. In this way, the lane center line is different from the environment map element. The environment map element is an actual existing map element in the environment, such as left and right lane lines, guardrails, and indicator lights. The lane center line is virtual data generated based on an algorithm / model and is not an actual existing map element in the environment. The lane center line can be used to indicate that the autonomous vehicle travels in the lane line. Therefore, this way distinguishes the different generation ways of the environment map element and the lane center line element. Specifically, real-time environment data collected by at least one environment perception device on the autonomous vehicle is acquired. The actual existing map element in the environment can be identified from the real-time environment data, and the environment map element is obtained. The lane center line element is obtained by performing lane center line detection on the environment map element by using a road center line detection model. Finally, the environment map information is obtained by combining the environment map element and the lane center line element. The environment map information includes the environment map element and the lane center line element.

[0073] In step S304, at least one target reachable key point of the autonomous vehicle at the real-time position is determined by using a preset first-level map based on the real-time position information, and target map elements in a target map element indication range of the at least one target reachable key point are extracted from a preset second-level map, to obtain target map information.

[0074] In step S305, the environment map information and the target map information are fused to obtain real-time map information.

[0075] In an embodiment, the real-time map element and the target map element comprise a lane center line element; the step S305 comprises: sampling the lane center line elements at the overlapping and non-overlapping positions of the environment map information and the target map information to obtain overlapping sampling results and non-overlapping sampling results; performing curve fitting of the lane center line elements on the overlapping sampling results and the non-overlapping sampling results, and performing lane center line smoothing update on the target map information through the fitting results to obtain the real-time map information. In this embodiment, when fusing the environment map and the target map information, in order to improve the smoothness of the lane center line fusion, a plurality of points are sampled at the overlapping and non-overlapping positions of the lane center lines of the two, to obtain the overlapping sampling results and the non-overlapping sampling results, and then the sampling points in the overlapping sampling results and the non-overlapping sampling results are refitted into a cubic curve, so that the lane center line in the target map information is updated, and the real-time map information containing the smooth lane center line is obtained, so that the map used for real-time navigation of automatic driving does not jump, and the stability is improved.

[0076] In an embodiment, after the step S305, the method further comprises: according to the real-time position information, performing road curvature degree calculation on the lane line elements outside the preset distance in the real-time map information to obtain a target road curvature degree; judging whether the target road curvature degree is less than a preset degree threshold; if the target road curvature degree is less than the preset degree threshold, performing road shape point extraction on the lane line elements outside the preset distance in the real-time map information, and removing the road indicating points outside the extracted road shape points in the real-time map information to obtain compressed real-time map information. In this embodiment, in order to further compress the real-time map information loaded into the memory, for some distant roads (outside the preset distance), if the road curvature degree is low (the target road curvature degree is less than the preset degree threshold), some road shape points can be reserved, and by removing the road indicating points outside the road shape points, the purpose of map compression is achieved. Specifically, in this embodiment, first, the road curvature degree calculation is performed on the lane line elements outside the preset distance in the real-time map information, and then it is judged whether the target road curvature degree obtained by calculation is less than the preset degree threshold. If yes, it indicates that the distant road is relatively flat, then the road shape points outside the preset distance are extracted, and the road indicating points outside the road shape points are removed, so that the compressed real-time map information is obtained, wherein the road indicating points refer to the road points.

[0077] The method for generating a real-time map provided by the above embodiment first acquires real-time position information of an autonomous vehicle, determines at least one target reachable key point from a preset first-level map according to the real-time position information, extracts map elements within a range indicated by the at least one target reachable key point from a preset second-level map to obtain offline target map information, and finally fuses the target map information with environment map information detected by the autonomous vehicle in real time to obtain real-time map information. The method reduces the running space occupied by real-time map generation through a hierarchical map mechanism.

[0078] Corresponding to the method embodiment, refer to Figure 4 FIG. 1 shows a schematic diagram of a device for generating a real-time map, which includes an acquisition module 40 configured to acquire environment map information detected by an autonomous vehicle in real time and real-time position information, an extraction module 42 configured to determine at least one target reachable key point of the autonomous vehicle at a real-time position thereof from a preset first-level map based on the real-time position information, extract target map elements within a range indicated by the at least one target reachable key point from a preset second-level map to obtain target map information, and a fusion module 44 configured to fuse the environment map information and the target map information to obtain real-time map information.

[0079] Optionally, the environment map information includes real-time map elements; the fusion module 44 includes a comparison unit configured to compare the real-time map elements and the target map elements to obtain a comparison result, the comparison result being used to indicate whether the target map information has changed, and a fusion unit configured to fuse the environment map information and the target map information through the real-time map elements and the target map elements to obtain real-time map information if the comparison result indicates that the target map information has changed.

[0080] Optionally, the real-time map elements and the target map elements include lane line elements; the comparison unit is further configured to acquire a confidence degree corresponding to a lane line element in the real-time map elements, compare the lane line elements in the real-time map elements and the target map elements, and determine whether a lane line element in the target map information has changed according to the confidence degree to obtain a comparison result.

[0081] Optionally, the comparing unit is further configured to: determine whether a deviation between the current frame lane line element and the at least one historical frame lane line element is greater than a preset deviation threshold, the current frame lane line element being used to indicate a lane line element in a real-time map element of a current frame, and the historical frame lane line element being used to indicate a lane line element in a real-time map element of a historical frame; if the deviation between the current frame lane line element and the at least one historical frame lane line element is greater than the preset deviation threshold, determine a lane line element in a real-time map element of a previous frame as a lane line element in a real-time map element of the current frame; and compare the lane line element in the real-time map element of the current frame and the lane line element in the target map element to obtain a comparison result.

[0082] Optionally, the real-time map element and the target map element include a lane center line element, and the fusion module 44 is further configured to: sample lane center line elements at overlapping and non-overlapping positions of the environment map information and the target map information to obtain overlapping position sampling results and non-overlapping position sampling results; perform curve fitting on the overlapping position sampling results and the non-overlapping position sampling results to obtain a fitting result; and perform lane center line smoothing update on the target map information based on the fitting result to obtain real-time map information.

[0083] Optionally, the device further includes: a calculation module configured to perform road curvature degree calculation on lane line elements outside a preset distance in the real-time map information based on the real-time position information to obtain a target road curvature degree; a judgment module configured to determine whether the target road curvature degree is less than a preset degree threshold; and a compression module configured to, if the target road curvature degree is less than the preset degree threshold, perform road shape point extraction on the lane line elements outside the preset distance in the real-time map information, and remove road indication points outside the extracted road shape points in the real-time map information to obtain compressed real-time map information.

[0084] Optionally, the acquisition module 40 is further configured to: acquire real-time position information and real-time environment data collected by at least one environment perception device on an autonomous vehicle during an autonomous driving process of the autonomous vehicle, and generate an environment map element based on the real-time environment data; perform lane center line detection on the environment map element based on a preset road center line detection model to obtain a real-time detected lane center line element; and combine the environment map element and the real-time detected lane center line element to obtain environment map information.

[0085] Optionally, the extraction module 42 is further configured to: based on the real-time position information, perform a road accessibility search on the real-time position of the autonomous vehicle by using a preset first-level map, to obtain all accessible key points of the autonomous vehicle at the real-time position, wherein the preset first-level map comprises key point information at all lane branches and key point information at preset lane intervals, and the key point information comprises key point position information; and perform key point screening on all accessible key points within a preset accessible distance, to obtain at least one target accessible key point of the autonomous vehicle at the real-time position.

[0086] Optionally, the extraction module 42 is further configured to: determine whether the at least one target accessible key point is included in a preset range of the autonomous vehicle; if the at least one target accessible key point is not included in the preset range of the autonomous vehicle, extract map information within a range of a nearest accessible key point from a preset second-level map, to obtain target map information, wherein the nearest accessible key point is used to indicate a target accessible key point closest to the preset range; and if the at least one target accessible key point is included in the preset range of the autonomous vehicle, extract map information within the preset range from the preset second-level map, to obtain the target map information.

[0087] The above real-time map generation apparatus first acquires real-time position information of an autonomous vehicle, determines at least one target accessible key point from a preset first-level map according to the real-time position information, extracts map elements within a range indicated by the at least one target accessible key point from a preset second-level map, to obtain offline target map information, and finally fuses the target map information with environment map information detected by the autonomous vehicle in real time, to obtain real-time map information. This method reduces the running space occupied by real-time map generation by using a hierarchical map mechanism.

[0088] The embodiment also provides an electronic device including a processor and a memory, wherein the memory stores machine-executable instructions capable of being executed by the processor, and the processor executes the machine-executable instructions to implement the above real-time map generation method. The electronic device can be a server or a terminal device.

[0089] Referring to Figure 5 The electronic device includes a processor 100 and a memory 101, wherein the memory 101 stores machine-executable instructions capable of being executed by the processor 100, and the processor 100 executes the machine-executable instructions to implement the above real-time map generation method.

[0090] Further, Figure 5 The electronic device also includes a bus 102 and a communication interface 103, and the processor 100, the communication interface 103 and the memory 101 are connected through the bus 102.

[0091] The memory 101 can include a high-speed random access memory (RAM), and can also include a non-volatile memory, such as at least one disk memory. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 103 (which can be wired or wireless), and the Internet, a wide area network, a local area network, a metropolitan area network, etc. can be used. The bus 102 can be an ISA bus, a PCI bus, or an EISA bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 Only one bidirectional arrow is used in the figure to represent the bus, but it does not mean that there is only one bus or only one type of bus.

[0092] The processor 100 can be an integrated circuit chip with signal processing capability. In the implementation process, each step of the above method can be completed by the integrated logic circuit of hardware in the processor 100 or the instruction in the form of software. The processor 100 described above can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. Each method, step and logic block disclosed in the embodiment of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. The steps of the method disclosed in combination with the embodiment of the present application can be directly embodied as a hardware decoding processor for execution, or a combination of hardware and software modules in the decoding processor for execution. The software module can be located in a random access memory, a flash memory, a read-only memory, a programmable read-only memory, an electrically erasable programmable memory, a register, etc. The storage medium in the art. The storage medium is located in the memory 101, and the processor 100 reads the information in the memory 101, and combines the hardware to complete the steps of the method of the above embodiment.

[0093] The embodiment also provides a computer readable storage medium, the computer readable storage medium stores computer executable instructions, when the computer executable instructions are called and executed by the processor, the computer executable instructions cause the processor to realize the above real-time map generation method.

[0094] The computer program product of the real-time map generation method, device, electronic equipment and storage medium provided by the embodiment of the application comprises a computer readable storage medium storing program codes, the program codes comprise instructions for executing the method described in the foregoing method embodiments, and specific implementation can be referred to the method embodiments, and details are not described herein.

[0095] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system and device described above can refer to the corresponding process in the foregoing method embodiments, and details are not described herein.

[0096] In addition, in the description of the embodiment of the application, unless otherwise explicitly specified and limited, the terms “mounting”, “connection”, “connecting” should be understood in a broad sense, for example, can be fixed connection, can also be detachable connection, or integrally connected; can be mechanical connection, can also be electrical connection; can be directly connected, can also be indirectly connected through an intermediate medium, can be the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the application can be understood according to specific circumstances.

[0097] If the functions are realized in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the application or parts of the technical solutions that essentially contribute to the prior art or parts of the technical solutions can be embodied in the form of software products. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the embodiments of the application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0098] In the description of the application, it should be noted that the terms “center”, “upper”, “lower”, “left”, “right”, “vertical”, “horizontal”, “inner”, “outer” and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, therefore cannot be understood as a limitation on the application. In addition, the terms “first”, “second”, “third” are only for the purpose of description, and cannot be understood as indicating or implying relative importance.

[0099] Finally, it should be noted that the above examples are merely specific embodiments of the present application, and are used to illustrate the technical solutions of the present application, but are not intended to limit the present application. The protection scope of the present application is not limited thereto. Although the present application has been described in detail with reference to the foregoing examples, those skilled in the art should understand that any person skilled in the art can still modify or easily think of changes to the technical solutions recorded in the foregoing examples, or make equivalent replacements to some of the technical features, within the technical range disclosed by the present application. These modifications, changes or replacements do not cause the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method of generating a real-time map, characterized by, The method for generating the real-time map comprises: obtaining environment map information and real-time position information detected by an autonomous vehicle in real time, wherein the environment map information comprises real-time map elements; based on the real-time position information, determining at least one target reachable key point of the autonomous vehicle at a real-time position thereof through a preset first-level map, and extracting target map elements within a target reachable key point indication range of the at least one target reachable key point from a preset second-level map to obtain target map information; fusing the environment map information and the target map information to obtain real-time map information; the real-time map elements and the target map elements comprise lane center line elements; the fusing of the environment map information and the target map information to obtain real-time map information comprises: sampling lane center line elements at overlapping and non-overlapping positions of the environment map information and the target map information to obtain overlapping position sampling results and non-overlapping position sampling results; performing curve fitting of lane center line elements on the overlapping position sampling results and the non-overlapping position sampling results, and performing lane center line smoothing update on the target map information through the fitting results to obtain real-time map information.

2. The method of claim 1, wherein, the environment map information comprises real-time map elements; the fusing of the environment map information and the target map information to obtain real-time map information comprises: comparing the real-time map elements and the target map elements to obtain a comparison result, wherein the comparison result is used to indicate whether the target map information has changed; if the comparison result indicates that the target map information has changed, fusing the environment map information and the target map information through the real-time map elements and the target map elements to obtain real-time map information.

3. The method of claim 2, wherein, the real-time map elements and the target map elements comprise lane line elements; the comparing of the real-time map elements and the target map elements to obtain a comparison result comprises: obtaining a confidence degree corresponding to a lane line element in the real-time map elements; comparing the lane line elements in the real-time map elements and the target map elements, and determining whether the lane line elements in the target map information have changed according to the confidence degree to obtain a comparison result.

4. The method of claim 2, wherein, the comparing of the real-time map elements and the target map elements to obtain a comparison result comprises: determining whether a deviation between a current frame lane line element and at least one historical frame lane line element is greater than a preset deviation threshold, wherein the current frame lane line element is used to indicate a lane line element in a real-time map element of a current frame, and the historical frame lane line element is used to indicate a lane line element in a real-time map element of a historical frame; if the deviation between the current frame lane line element and the at least one historical frame lane line element is greater than the preset deviation threshold, determining a lane line element in a real-time map element of a previous frame as a lane line element in a real-time map element of the current frame; comparing the lane line element in the real-time map element of the current frame and the lane line element in the target map element to obtain a comparison result.

5. The method of claim 1, wherein, After the environment map information and the target map information are fused to obtain real-time map information, the method further includes: According to the real-time position information, a road curvature degree of a lane line element outside a preset distance in the real-time map information is calculated to obtain a target road curvature degree; It is judged whether the target road curvature degree is less than a preset degree threshold; If the target road curvature degree is less than the preset degree threshold, a road shape point of the lane line element outside the preset distance in the real-time map information is extracted, and a road indication point outside the extracted road shape point in the real-time map information is removed to obtain compressed real-time map information.

6. The method of claim 1, wherein, The environment map information and the real-time position information detected by the automatic driving vehicle in real time are obtained, including: In the automatic driving process of the automatic driving vehicle, real-time position information and real-time environment data collected by at least one environment perception device on the automatic driving vehicle are obtained in real time, and an environment map element is generated through the real-time environment data; A lane center line element detected in real time is obtained by detecting the lane center line of the environment map element through a preset road center line detection model; The environment map information is obtained by combining the environment map element and the lane center line element detected in real time.

7. The method of claim 1, wherein, The at least one target reachable key point of the automatic driving vehicle at the real-time position is determined based on the real-time position information through a preset first-level map, including: Based on the real-time position information, a road reachability search is performed on the real-time position of the automatic driving vehicle through a preset first-level map to obtain all reachable key points of the automatic driving vehicle at the real-time position, the preset first-level map including key point information at all lane branch points and key point information at a preset lane distance, and the key point information including key point position information; The at least one target reachable key point of the automatic driving vehicle at the real-time position is obtained by performing key point screening of a preset reachable distance on all reachable key points.

8. The method of generating a real-time map according to any one of claims 1-7, wherein, The target map element in the target map information is extracted from the preset second-level map, including: It is judged whether the preset range of the automatic driving vehicle contains the at least one target reachable key point; If the preset range of the automatic driving vehicle does not contain the at least one target reachable key point, the map information in the range of the nearest reachable key point is extracted from the preset second-level map to obtain the target map information, and the nearest reachable key point is used to indicate the nearest target reachable key point from the preset range; If the preset range of the automatic driving vehicle contains the at least one target reachable key point, the map information in the preset range is extracted from the preset second-level map to obtain the target map information.

9. An apparatus for generating a real-time map, the apparatus comprising: The real-time map generation device includes: An acquisition module is configured to acquire environment map information and real-time position information detected by an automatic driving vehicle in real time, wherein the environment map information includes real-time map elements; The extraction module is configured to determine at least one target reachable key point at the real-time position of the autonomous vehicle based on the real-time position information and by using a preset first-level map, and extract target map elements within a range of the at least one target reachable key point from a preset second-level map to obtain target map information. The fusion module is configured to fuse the environment map information and the target map information to obtain real-time map information. The real-time map elements and the target map elements include lane center line elements, and the fusion module is further configured to: sample lane center line elements at overlapping and non-overlapping positions of the environment map information and the target map information to obtain overlapping position sampling results and non-overlapping position sampling results; perform curve fitting on the overlapping position sampling results and the non-overlapping position sampling results to obtain a lane center line, and perform lane center line smoothing update on the target map information based on the lane center line to obtain real-time map information.

10. An electronic device, comprising: A processor and a memory are included, the memory stores machine executable instructions that can be executed by the processor, and the processor executes the machine executable instructions to implement the real-time map generation method in any one of claims 1-8.

11. A computer readable storage medium, characterized in that, The computer readable storage medium stores computer executable instructions, and when the computer executable instructions are called and executed by a processor, the computer executable instructions cause the processor to implement the real-time map generation method in any one of claims 1-8.

Images