Initialization method and device for garage positioning, electronic equipment and storage medium

Abstract

The invention discloses a garage positioning initialization method and device, electronic equipment and a storage medium, and relates to the technical field of vehicle navigation. The method comprises the following steps: when a vehicle is in a driving state, determining a basement entrance connection point and determining the position of the connection point based on a pitch angle sudden change feature in three-dimensional position information and a confidence coefficient jump feature in confidence coefficient information; when the vehicle enters the target area again, performing multi-dimensional matching on the extracted real-time track data and the memory track data to generate a track matching degree; when the track matching degree exceeds a preset matching threshold value, pose initialization is executed based on the position of the connection point, and the initial pose of the vehicle is generated; based on the initial pose and the position of the fixed structure in the target area, the initial pose of the vehicle is optimized, and the optimized initial pose is determined. According to the invention, the risk of wrong recognition of the basement entrance is reduced, and the initial pose of the vehicle can be accurately determined.

Description

Technical Field

[0001] The present application relates to the field of vehicle navigation technology, and in particular to a method, device, electronic device and storage medium for initializing garage positioning. Background Art

[0002] At present, with the development of society and the advancement of technology, more and more people are beginning to use the valet parking function while driving their vehicles.

[0003] However, when faced with complex parking area environments with dense high-rise buildings, traditional positioning technology has low spatial position confidence due to the multi-path effect, making it difficult to accurately identify entrance channels in underground parking lots with multiple entrances. There is a risk of entrance recognition errors, and the parking space layouts of various garages are currently highly repetitive, resulting in inaccurate initial vehicle posture. Summary of the Invention

[0004] The embodiments of the present application provide a method, device, electronic device and storage medium for initializing garage positioning. The embodiments provided by the present application solve the technical problems in the prior art of the risk of entrance recognition errors and inaccurate initial posture of the vehicle. The embodiments provided by the present application reduce the risk of incorrect identification of the basement entrance and can accurately determine the initial posture of the vehicle.

[0005] In a first aspect of the embodiments of the present application, an embodiment of the present application provides a method for initializing garage positioning, the method for initializing garage positioning comprising:

[0006] When the vehicle is in a driving state, obtaining memory track data generated when the vehicle first enters a target area, the memory track data including three-dimensional position information, confidence information, and semantic label information;

[0007] Determine the basement entrance connection point and the connection point position based on the pitch angle mutation feature in the three-dimensional position information and the confidence jump feature in the confidence information;

[0008] When the vehicle enters the target area again, the extracted real-time trajectory data is multi-dimensionally matched with the memory trajectory data to generate a trajectory matching degree, wherein the multi-dimensional matching includes three-dimensional position information matching, confidence information matching, and semantic label information matching;

[0009] When the trajectory matching degree exceeds a preset matching threshold, performing posture initialization based on the connection point position to generate the initial posture of the vehicle;

[0010] Based on the initial posture and the position of the fixed structure in the target area, the initial posture of the vehicle is optimized to determine an optimized initial posture.

[0011] In a feasible implementation manner, determining the basement entrance connection point and the connection point position based on the pitch angle mutation feature in the three-dimensional position information and the confidence jump feature in the confidence information includes:

[0012] Determining whether there is a ramp in the target area based on a pitch angle mutation feature in the three-dimensional position information;

[0013] If so, determining the entry point and exit point of the ramp based on the angle change of the pitch angle mutation feature and a preset angle change threshold;

[0014] Determining a confidence state transition point position based on a confidence change amount of the confidence transition feature in the confidence information and a preset confidence change threshold;

[0015] Based on the position of the slope entry point, the position of the slope exit point and the position of the confidence state jump point, the basement entrance connection point and the connection point position are determined.

[0016] In a feasible implementation, the preset angle change threshold includes a first preset angle change threshold and a second preset angle change threshold, the first preset angle change threshold is greater than the second preset angle change threshold, and determining the entry point and exit point of the ramp based on the angle change amount of the pitch angle sudden change feature and the preset angle change threshold includes:

[0017] When the vehicle is at a first position, if the angle change is greater than the first preset angle change threshold, the first position is used as the entry point of the ramp;

[0018] When the vehicle is at the second position, if the angle change is less than the second preset angle change threshold, the second position is used as the exit point of the slope.

[0019] In a feasible implementation manner, determining the position of the confidence state transition point based on the confidence change amount of the confidence transition feature in the confidence information and a preset confidence change threshold includes:

[0020] When the vehicle is at the third position, if the confidence change is greater than the preset confidence change threshold, the third position is used as a confidence state transition point position.

[0021] In a feasible implementation manner, after determining whether there is a ramp in the target area based on the pitch angle mutation feature in the three-dimensional position information, the method further includes:

[0022] If the ramp does not exist in the target area, determining the position of the confidence state jump point based on the confidence change amount of the confidence jump feature in the confidence information and a preset confidence change threshold;

[0023] When it is determined that there is a parking space for the vehicle at the third position, the position of the confidence state jump point is determined to be the underground garage entrance connection point and the connection point position is determined.

[0024] In a feasible implementation, optimizing the initial posture of the vehicle based on the initial posture and the position of the fixed structure in the target area to determine the optimized initial posture includes:

[0025] Mapping the semantic label information corresponding to the initial pose to a preset global coordinate system to construct a local semantic map;

[0026] The position of the fixed structure in the local semantic map is matched with a pre-built map library, the initial posture of the vehicle is optimized, and an optimized initial posture is determined.

[0027] In a feasible implementation, matching the position of the fixed structure in the local semantic map to a pre-built map library, optimizing the initial posture of the vehicle, and determining the optimized initial posture includes:

[0028] extracting three-dimensional coordinate data of the fixed structure from the local semantic map;

[0029] performing difference calculation between the three-dimensional coordinate data of the fixed structure and the standard three-dimensional coordinate data of the pre-built structure in the pre-built map library to determine a minimum residual, and using the minimum residual as the deviation posture of the vehicle;

[0030] Based on the deviation posture, the initial posture of the vehicle is optimized to determine an optimized initial posture.

[0031] In a second aspect of the embodiments of the present application, the embodiments of the present application provide a device for initializing the positioning of the garage, comprising:

[0032] an acquisition module, configured to acquire, when the vehicle is in a driving state, memory track data generated when the vehicle first enters a target area, the memory track data including three-dimensional position information, confidence information, and semantic label information;

[0033] A determination module, configured to determine a connection point of an underground garage entrance and a position of the connection point based on a pitch angle mutation feature in the three-dimensional position information and a confidence jump feature in the confidence information;

[0034] a matching module, configured to perform multi-dimensional matching between the extracted real-time trajectory data and the stored trajectory data when the vehicle re-enters the target area, and generate a trajectory matching degree, wherein the multi-dimensional matching includes matching of three-dimensional position information, matching of confidence information, and matching of semantic label information;

[0035] A generation module, configured to perform posture initialization based on the position of the connection point to generate an initial posture of the vehicle when the trajectory matching degree exceeds a preset matching threshold;

[0036] An optimization module is used to optimize the initial posture of the vehicle based on the initial posture and the position of the fixed structure in the target area, and determine an optimized initial posture.

[0037] In a third aspect of the embodiments of the present application, the embodiments of the present application provide an electronic device comprising: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate via the bus, and the machine-readable instructions are executed by the processor to perform the steps of the initialization method for garage positioning as described above.

[0038] In a fourth aspect of the embodiments of the present application, the embodiments of the present application provide a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the initialization method for garage positioning as described above are executed.

[0039] The garage positioning initialization method, device, electronic device and storage medium provided in the embodiments of the present application, compared with the prior art, the embodiments provided by the present application obtain memory trajectory data generated when the vehicle first enters the target area when the vehicle is in the driving state, and determine the connection point of the underground garage entrance and the connection point position based on the pitch angle mutation characteristics in the three-dimensional position information and the confidence jump characteristics in the confidence information. When the vehicle enters the target area again, the extracted real-time trajectory data is multi-dimensionally matched with the memory trajectory data to generate a trajectory matching degree. When the trajectory matching degree exceeds a preset matching threshold, posture initialization is performed based on the connection point position to generate the vehicle's initial posture. Finally, based on the initial posture and the position of fixed structures in the target area, the initial posture of the vehicle is optimized to determine the optimized initial posture. The present application reduces the risk of incorrect identification of the underground garage entrance and can accurately determine the vehicle's initial posture. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 A flowchart of a method for initializing garage positioning provided by an embodiment of the present application is shown;

[0041] Figure 2A structural block diagram of a garage positioning initialization device provided by an embodiment of the present application is shown;

[0042] Figure 3 A schematic structural diagram of an electronic device provided in an embodiment of the present application is shown.

[0043] Figure 2 and Figure 3 The corresponding relationship between the reference numerals and the names of the drawings is as follows:

[0044] 200 garage positioning initialization device; 210 acquisition module; 220 determination module; 230 matching module; 240 generation module; 250 optimization module; 300 electronic device; 310 processor; 320 memory; 330 bus. DETAILED DESCRIPTION

[0045] In order to better understand the technical solutions provided by the embodiments of this specification, the technical solutions of the embodiments of this specification are described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of this specification and the specific features in the embodiments are detailed descriptions of the technical solutions of the embodiments of this specification, rather than limitations on the technical solutions of this specification. In the absence of conflict, the embodiments of this specification and the technical features in the embodiments can be combined with each other.

[0046] In this article, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or equipment comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or equipment. In the absence of further restrictions, the elements defined by the statement "comprising a ..." do not exclude the presence of other identical elements in the process, method, article or equipment comprising the elements. The term "two or more" includes two or more than two cases.

[0047] First, the applicable application scenarios of this application are introduced. The embodiments provided in this application are applicable to the field of vehicle navigation technology.

[0048] Currently, when faced with complex parking areas with dense high-rise buildings, traditional positioning technology has low spatial position confidence due to the multi-path effect, making it difficult to accurately identify entrance channels in underground parking lots with multiple entrances. There is a risk of entrance recognition errors, and the parking space layouts of various garages are currently highly repetitive, resulting in inaccurate initial vehicle posture.

[0049] Based on this, the embodiments of the present application provide a method, device, electronic device and storage medium for initializing garage positioning. The embodiments provided by the present application solve the technical problems in the prior art of the risk of entrance recognition errors and inaccurate initial posture of the vehicle. The embodiments provided by the present application reduce the risk of incorrect identification of the basement entrance and can accurately determine the initial posture of the vehicle.

[0050] Figure 1 FIG. 1 shows a flow chart of a method for initializing garage positioning provided by an embodiment of the present application. Figure 1 As shown, the initialization method for garage positioning includes the following steps:

[0051] S101. When a vehicle is in a driving state, obtain memory trajectory data generated when the vehicle first enters a target area, where the memory trajectory data includes three-dimensional position information, confidence information, and semantic label information.

[0052] In this step, if the embodiment provided by the present application wants to perform valet parking in the target area, it is first necessary to learn the process of valet parking, which can be specifically as follows: when it is determined that the vehicle is in a driving state, the above-mentioned vehicle is first controlled from the outside when it enters the target area for the first time, and the combined navigation (a positioning device) is controlled to output memory trajectory data containing three-dimensional position information, confidence information and semantic label information.

[0053] It is understood that the memory trace data in the embodiment provided in this application is as follows:

[0054]

[0055] Among them, mem path It is used to represent the memory trajectory data generated when the vehicle first enters the target area; t0 is used to represent the moment when the connection point POI is detected; n is used to represent the information recorded at n moments; pos g Used to represent the three-dimensional position information detected by the combined navigation; conf g Used to characterize the combined navigation given pos g confidence information; mod is used to represent semantic label information.

[0056] It should be noted that the semantic tag information mod in the embodiments provided in this application includes multiple types, such as:

[0057] First semantic label information, used to represent the semantic-free area;

[0058] The second semantic label information is used to characterize the gate area;

[0059] The third semantic label information is used to represent the stop line area;

[0060] The fourth semantic label information is used to characterize the sidewalk area;

[0061] The fifth semantic label information is used to represent the connection point area;

[0062] The specific formula is as follows:

[0063]

[0064] Here, the priority rules of semantic tags can be set as:

[0065] Connection point label priority > Gate label priority > Stop line label priority > Sidewalk label priority.

[0066] Among them, the target area in the embodiments provided in this application can be customized and used according to different application scenarios. The target area in the embodiments provided in this application can specifically be roads and road conditions with severe obstructions in cities or urban areas.

[0067] S102. Determine the connection point of the basement entrance and the position of the connection point based on the pitch angle mutation feature in the three-dimensional position information and the confidence jump feature in the confidence information.

[0068] In this step, in the embodiment provided by the present application, when it is detected that a vehicle is about to enter the entrance of the underground garage, the pitch angle mutation characteristics in the three-dimensional position information and the confidence jump characteristics in the confidence information are used to determine the connection point of the underground garage entrance and determine the position of the connection point, that is, the position of the underground garage entrance.

[0069] It can be understood that in the embodiment provided by the present application, the detection of the location of the basement entrance is divided into two situations, one is that there is a ramp at the above-mentioned basement entrance, and the other is that there is no ramp at the above-mentioned basement entrance. The two methods of determining the location of the basement entrance connection point are different, which is explained below with examples.

[0070] For example, if there is a ramp at the basement entrance, determining the basement entrance connection point and determining the location of the connection point include:

[0071] Based on the pitch angle mutation characteristics in the three-dimensional position information, determine whether there is a ramp in the target area; if so, determine the entry and exit points of the ramp based on the angle change of the pitch angle mutation characteristics and the preset angle change threshold; determine the position of the confidence state jump point based on the confidence change of the confidence jump characteristics in the confidence information and the preset confidence change threshold; based on the entry point position, exit point position and confidence state jump point position, determine the basement entrance connection point and the connection point position.

[0072] It should be noted that the preset angle change threshold includes a first preset angle change threshold and a second preset angle change threshold, the first preset angle change threshold is greater than the second preset angle change threshold, and the ramp entry point and the ramp exit point are determined based on the angle change amount of the pitch angle mutation characteristic and the preset angle change threshold, which can be:

[0073] When the vehicle is in a first position, if the angle change is greater than a first preset angle change threshold, the first position is used as the entry point of the ramp; when the vehicle is in a second position, if the angle change is less than a second preset angle change threshold, the second position is used as the exit point of the ramp.

[0074] It can be understood that the above judgment process is expressed by the following formula:

[0075]

[0076] POS T =pos g (t);

[0077] Among them, pitch is used to characterize the sudden change characteristics of the pitch angle; f(p) is used to characterize the degree of pitch change; 1 is used to characterize an uphill slope; -1 is used to characterize a downhill slope; pitch(t1) is used to characterize the angle change of the pitch at time t1, which is the moment when the uphill or downhill slope is detected; For characterization The angle change of pitch at a moment; TH p Used to represent the preset angle change threshold.

[0078] Exemplarily, when the vehicle is at the third position, if the confidence change is greater than a preset confidence change threshold, the third position is used as the confidence state transition point position.

[0079] It should be noted that the formula for characterizing the change in the confidence jump characteristic in the embodiment provided in this application is as follows:

[0080]

[0081] Among them, f(m) is used to characterize the confidence change of the confidence jump feature detected by the integrated navigation, 0 means no jump, and 1 means jump; cof g (t2) represents the confidence information output at time t2; cof g Indicates time before t2 Confidence of TH c Indicates the preset confidence change threshold.

[0082] In the above, the calculation formula for determining the entry point or exit point of a ramp to a flat slope is as follows:

[0083]

[0084] pos' T =pos g (t0);

[0085] Among them, TH p Used to represent the preset flat slope angle threshold; f'(p) indicates the out-of-slope sign.

[0086] In the above, after determining the entry and exit points, as well as the confidence state transition point, the connection point POI is determined and its location is determined according to the following formula:

[0087]

[0088] Among them, f'(p) represents the out-slope sign; f(p) represents the in-slope sign; norm(pos T -pos' T ) is used to represent the distance from the entry to the exit; pos g Used to represent the location of the connection point POI; POI is used to represent the connection point.

[0089] For example, if there is no ramp at the basement entrance, determining the basement entrance connection point and determining the location of the connection point includes:

[0090] If there is no ramp in the target area, the position of the confidence state jump point is determined based on the confidence change of the confidence jump feature in the confidence information and the preset confidence change threshold; when it is determined that there is a parking space for the vehicle at the third position, the position of the confidence state jump point is determined to be the connection point of the underground garage entrance and the connection point position is determined.

[0091] It should be noted that when it is determined that there is no ramp in the target area in the embodiment provided in this application, it is necessary to re-determine the position of the confidence state jump point based on the confidence change amount of the confidence jump feature in the confidence information and the preset confidence change threshold, and when it is determined that there is a parking space for the vehicle at the third position, the position of the confidence state jump point is determined to be the basement entrance connection point and the connection point position is determined.

[0092] In the above, the formula for characterizing the change in confidence jump characteristics is as follows:

[0093]

[0094] Among them, f(m) is used to characterize the confidence change of the confidence jump feature detected by the integrated navigation, 0 means no jump, and 1 means jump; cof g (t2) represents the confidence information output at time t2; cof g Indicates time before t2 Confidence information of TH c Indicates the preset confidence change threshold.

[0095] In the above, the confidence state transition point is determined to be the basement entrance connection point, and the formula for determining the connection point position is:

[0096]

[0097] Among them, pos g ,(f(m) is used to represent the third position where there is a parking space.

[0098] It should be noted that after determining the connection point and the location of the connection point at the basement entrance, the embodiment provided by this application needs to create a map based on the connection point location and the memory trajectory data, and store the above map file into two files to form outdoor navigation level map data and indoor road network data, which serve as prior data for navigation and positioning the next time the vehicle enters the target area.

[0099] The map file created in the embodiments provided in this application may include road network relationships, semantic road association information, and semantic global coordinate data, etc.

[0100] S103. When the vehicle re-enters the target area, multi-dimensional matching is performed on the extracted real-time trajectory data and the memory trajectory data to generate a trajectory matching degree, wherein the multi-dimensional matching includes three-dimensional position information matching, confidence information matching, and semantic label information matching.

[0101] In this step, after the vehicle has recorded the memory trajectory data generated when it first entered the target area and determined an underground garage entrance connection point and the location of the connection point, when the vehicle enters the target area again, that is, enters the same scene, it is necessary to perform multi-dimensional matching between the real-time trajectory data and the memory trajectory data, including three-dimensional position information matching, confidence information matching, and semantic label information matching, and determine whether the vehicle has entered the underground garage based on the trajectory matching degree.

[0102] It should be noted that the embodiments provided in this application will superimpose the matching results of multi-dimensional matching, including three-dimensional position information matching, confidence information matching, and semantic label information matching, and determine the final trajectory matching degree.

[0103] It is understood that the embodiment provided by the present application is to memorize each position in the real-time trajectory data after re-entering the target area until the vehicle reaches the connection point, and then match the number of each trajectory position in the above real-time trajectory data with the each trajectory position in the memorized trajectory data generated when the vehicle first enters the target area to determine the trajectory matching degree. The specific formula is as follows:

[0104]

[0105] Among them, match c Used to represent the result of number matching; 1 is used to represent a successful match; 0 is used to represent a failed match; ln(conter m ) is used to represent the number of memories; ln(conter g ) is used to characterize the number of cached items this time, and the difference in number caused by turning is not avoided. The embodiment provided in this application is willing to use ln to reduce the impact of this problem.

[0106] The trajectory matching process of the embodiment provided in this application is as follows: according to the memory trajectory data, the number of real-time trajectory data is interpolated to ensure that the number of two trajectories is exactly the same. Taking distance as an example, the weight of the middle value of the trajectory is set to 50, and then the weight is gradually reduced from the middle to both sides.

[0107] For example, the premise for matching the three-dimensional position information of the real-time trajectory data with the memory trajectory data is that the position deviation between the memory position in the memory trajectory data and the current position in the real-time trajectory data is less than a threshold value, and the matching formula is specifically:

[0108]

[0109] Among them, conter p Used to represent the number of positions that can be matched.

[0110] It should be noted that when forcing a position match between a point with a slope or a short 90° turn in the real-time trajectory data and the corresponding point in the memory trajectory data, it is necessary to calculate the sum of the position distances of the N points before and after the matching point; the calculation formula is as follows:

[0111]

[0112] Among them, dis p It is used to represent the average distance between the real-time trajectory of N values ​​before and after the key point with a slope or a short 90° turn and the cached memory trajectory; i is used to represent the key point number, pose g Used to represent real-time trajectory points, pose m Memory trace points used to characterize caches.

[0113] For example, the formula for matching the confidence information of the real-time trajectory data with the memory trajectory data is as follows:

[0114]

[0115] Among them, conter p Indicates the number of confidence matches, that is, the difference between the memory confidence and the real-time current confidence is less than the threshold for matching.

[0116] It should be noted that the prerequisite for matching is that the difference between the memory confidence and the real-time current confidence is less than the threshold.

[0117] It is understandable that in order to further constrain signal matching from being restricted by position, the embodiment provided in this application solves the areas with the highest and lowest confidence levels for both, specifically, it is necessary to solve the sum of the position distances of the N points before and after the two areas mentioned above; the calculation formula is as follows:

[0118]

[0119] Among them, dis c The average distance between the real-time trajectory of the N values ​​before and after the key point representing the highest and lowest areas and the cached memory trajectory; i is the average distance between the memory and the cache; i is used to represent the key point sequence number, pose g Used to represent real-time trajectory points, pose m Memory trace points used to characterize caches.

[0120] For example, the formula for matching semantic label information between real-time trajectory data and memory trajectory data is as follows:

[0121]

[0122] Among them, conter m It is used to represent the number of semantic matches that can be made, that is, the memory semantics is consistent with the real-time current semantics, which is considered to be matchable.

[0123] It should be noted that, in order to ensure that the constraint signal matching is not restricted by position, the embodiment provided in this application needs to solve the area with the highest semantic richness and solve the sum of the position distances of N points before and after the area; the calculation formula is as follows:

[0124]

[0125] Among them, dis m The average distance between the N values ​​before and after the key point for representing the area with the highest semantic richness and the cached memory trace; i is used to represent the key point sequence number, pose gUsed to represent real-time trajectory points, pose m Memory trace points used to characterize caches.

[0126] It is understandable that after determining the number of matches in each of the above dimensions, the final matching degree is calculated using the following formula:

[0127]

[0128] Among them, match D It is used to characterize the final trajectory matching degree. That is, under the condition that the number matching degree is normal, each matching distance is superimposed. Finally, when it is determined that the matching degree exceeds a certain preset matching threshold, the corresponding preset map is extracted.

[0129] S104: When the trajectory matching degree exceeds a preset matching threshold, posture initialization is performed based on the connection point position to generate an initial posture of the vehicle.

[0130] In this step, after determining that the trajectory matching degree exceeds the preset matching threshold, the embodiment provided by this application needs to use integrated navigation to solve the vehicle's indoor posture increment. The solution formula is as follows:

[0131]

[0132] in, Used to represent the pose increment.

[0133] Based on the posture increment, the initial posture of the vehicle is determined, and the solution formula is as follows:

[0134]

[0135] Among them, pose(t) is used to represent the initial pose of the vehicle.

[0136] S105 : Optimizing the initial posture of the vehicle based on the initial posture and the positions of fixed structures in the target area to determine an optimized initial posture.

[0137] In this step, the present application maps the semantic label information corresponding to the initial posture to a preset global coordinate system to construct a local semantic map; and matches the positions of fixed structures in the local semantic map with the pre-built map library, optimizes the initial posture of the vehicle, and determines the optimized initial posture.

[0138] It can be understood that the embodiment provided in the present application maps the semantic label information corresponding to the initial posture to a preset global coordinate system, and splices the semantics of multiple frames, that is, the posture of the same pillar is averagely weighted to obtain a local map, and the fixed structures in the local map are matched with the pre-built map library to determine the position of the fixed structures, and then determine the optimal position of the vehicle's initial posture, and then match the subsequent vehicle trajectory according to this optimal initial posture. The above steps complete the initialization of parking.

[0139] Among them, the fixed structure in the embodiment provided by this application can be any type of fixed reference object in the basement, and the specific type can be customized and used according to different application scenarios. The fixed structure in the embodiment provided by this application can be set as a column in the parking lot basement.

[0140] Exemplarily, the method for determining the optimized initial posture is specifically as follows: extracting three-dimensional coordinate data of fixed structures from a local semantic map; performing difference calculation between the three-dimensional coordinate data of the fixed structure and the standard three-dimensional coordinate data of pre-built structures in a pre-built map library to determine the minimum residual, and using the above minimum residual as the deviation posture of the vehicle; based on the deviation posture, optimizing the initial posture of the vehicle to determine the optimized initial posture.

[0141] Among them, the pre-built map library in the embodiment provided by this application may include road network relationships, semantic road association information, and semantic global coordinate data, etc.

[0142] In the above, the three-dimensional coordinate data of the fixed structure in the embodiments provided in this application can be extracted using a lidar point cloud clustering algorithm, or can be determined by recognizing visual images through a convolutional neural network.

[0143] The initialization method for garage positioning provided by the embodiment of the present application, compared with the prior art, obtains the memory trajectory data generated when the vehicle first enters the target area when the vehicle is in the driving state, and determines the connection point of the underground garage entrance and the connection point position based on the pitch angle mutation feature in the three-dimensional position information and the confidence jump feature in the confidence information. When the vehicle enters the target area again, the extracted real-time trajectory data is multi-dimensionally matched with the memory trajectory data to generate a trajectory matching degree. When the trajectory matching degree exceeds a preset matching threshold, the posture initialization is performed based on the connection point position to generate the vehicle's initial The initial posture of the vehicle is finally optimized based on the initial posture and the position of the fixed structure in the target area to determine the optimized initial posture. This application reduces the risk of incorrect recognition of the basement entrance and can accurately determine the initial posture of the vehicle, thereby improving the accuracy of posture positioning. This application improves the accuracy of extracting the parking map, i.e., the local semantic map, by memorizing the memory trajectory data generated when the vehicle first enters the target area, and by matching the memory trajectory data generated when the vehicle first enters the target area with the real-time trajectory data in multiple dimensions such as voice tag information, thereby reducing the complexity of initializing the posture.

[0144] Figure 2 The structure frame of a garage positioning initialization device provided by an embodiment of the present application is shown. Figure 2 As shown, the initialization device 200 for garage positioning includes:

[0145] The acquisition module 210 is used to acquire the memory track data generated when the vehicle first enters the target area when the vehicle is in the driving state. The memory track data includes three-dimensional position information, confidence information and semantic label information.

[0146] The determination module 220 is used to determine the connection point of the basement entrance and the position of the connection point based on the pitch angle mutation feature in the three-dimensional position information and the confidence jump feature in the confidence information.

[0147] The matching module 230 is used to perform multi-dimensional matching between the extracted real-time trajectory data and the memory trajectory data when the vehicle re-enters the target area to generate a trajectory matching degree, wherein the multi-dimensional matching includes three-dimensional position information matching, confidence information matching, and semantic label information matching.

[0148] The generation module 240 is used to perform posture initialization based on the connection point position to generate the initial posture of the vehicle when the trajectory matching degree exceeds a preset matching threshold.

[0149] The optimization module 250 is used to optimize the initial posture of the vehicle based on the initial posture and the positions of fixed structures in the target area, and determine an optimized initial posture.

[0150] Exemplarily, the determination module 220 is specifically configured to:

[0151] Based on the pitch angle mutation characteristics in the three-dimensional position information, determine whether there is a ramp in the target area.

[0152] If so, the entry and exit points of the ramp are determined based on the angle change of the pitch angle mutation characteristic and a preset angle change threshold.

[0153] The position of the confidence state transition point is determined based on the confidence change amount of the confidence jump feature in the confidence information and a preset confidence change threshold.

[0154] Based on the entry point location, exit point location and confidence state jump point location, determine the basement entrance connection point and the connection point location.

[0155] Exemplarily, the preset angle change threshold includes a first preset angle change threshold and a second preset angle change threshold, the first preset angle change threshold being greater than the second preset angle change threshold, and determining the entry point and exit point of the ramp based on the angle change amount of the pitch angle mutation characteristic and the preset angle change threshold includes:

[0156] When the vehicle is at a first position, if the angle change is greater than a first preset angle change threshold, the first position is used as the entry point of the slope.

[0157] When the vehicle is at the second position, if the angle change is less than a second preset angle change threshold, the second position is used as the exit point of the slope.

[0158] Exemplarily, determining the position of the confidence state transition point based on the confidence change amount of the confidence transition feature in the confidence information and a preset confidence change threshold includes:

[0159] When the vehicle is at the third position, if the confidence change is greater than a preset confidence change threshold, the third position is used as the confidence state jump point position.

[0160] Exemplarily, if there is no ramp in the target area, the position of the confidence state transition point is determined based on the confidence change amount of the confidence transition feature in the confidence information and a preset confidence change threshold;

[0161] When it is determined that there is a parking space for the vehicle at the third position, the position of the confidence state jump point is determined to be the underground garage entrance connection point and the connection point position is determined.

[0162] Exemplarily, the optimization module 250 is specifically configured to:

[0163] The semantic label information corresponding to the initial pose is mapped to the preset global coordinate system to construct a local semantic map.

[0164] The positions of fixed structures in the local semantic map are matched with the pre-built map library, the initial posture of the vehicle is optimized, and the optimized initial posture is determined.

[0165] Exemplarily, the positions of fixed structures in the local semantic map are matched with a pre-built map library, and the initial posture of the vehicle is optimized. Determining the optimized initial posture includes:

[0166] Extract the 3D coordinate data of fixed structures from the local semantic map.

[0167] The three-dimensional coordinate data of the fixed structure is calculated with the standard three-dimensional coordinate data of the pre-built structure in the pre-built map library to determine the minimum residual, and the above minimum residual is used as the deviation posture of the vehicle.

[0168] Based on the deviation posture, the initial posture of the vehicle is optimized to determine the optimized initial posture.

[0169] The garage positioning initialization device 200 provided in the embodiment of the present application, compared with the prior art, obtains the memory trajectory data generated when the vehicle first enters the target area when the vehicle is in the driving state, and determines the garage entrance connection point and the connection point position based on the pitch angle mutation feature in the three-dimensional position information and the confidence jump feature in the confidence information. When the vehicle enters the target area again, the extracted real-time trajectory data is multi-dimensionally matched with the memory trajectory data to generate a trajectory matching degree. When the trajectory matching degree exceeds the preset matching threshold, the posture initialization is performed based on the connection point position to generate the vehicle The initial posture of the vehicle is optimized based on the initial posture and the position of the fixed structure in the target area to determine the optimized initial posture. This application reduces the risk of incorrect recognition of the basement entrance and can accurately determine the initial posture of the vehicle, thereby improving the accuracy of posture positioning. This application improves the accuracy of extracting the parking map, i.e., the local semantic map, by memorizing the memory trajectory data generated when the vehicle first enters the target area, and by matching the memory trajectory data generated when the vehicle first enters the target area with the real-time trajectory data in multiple dimensions such as voice tag information, thereby reducing the complexity of initializing the posture.

[0170] Figure 3 FIG. 1 shows a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 3 As shown, the electronic device 300 includes a processor 310 , a memory 320 , and a bus 330 .

[0171] The memory 320 stores machine-readable instructions executable by the processor 310. When the electronic device 300 is running, the processor 310 communicates with the memory 320 via the bus 330. When the machine-readable instructions are executed by the processor 310, the above-mentioned Figure 1 The steps of the method for initializing garage positioning in the illustrated method embodiment are specifically implemented in accordance with the method embodiment and will not be described in detail here.

[0172] The embodiment of the present application also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the computer program can execute the above-mentioned Figure 1 The steps of the method for initializing garage positioning in the illustrated method embodiment are specifically implemented in accordance with the method embodiment and will not be described in detail here.

[0173] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0174] It should be noted that, in the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0175] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-readable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-readable program code.

[0176] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0177] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0178] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0179] An embodiment of the present application further provides a computer program product, which includes computer software instructions. When the computer software instructions are executed on a processing device, the processing device executes the process of the garage positioning initialization method.

[0180] A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that a computer can store or a data storage device such as a server or data center that includes one or more available media integrated. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state drive (SSD)).

[0181] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0182] In the several embodiments provided in this application, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0183] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0184] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0185] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0186] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

[0187] Although the preferred embodiments of this specification have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concepts. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of this specification.

[0188] Obviously, those skilled in the art may make various changes and modifications to this specification without departing from the spirit and scope of this specification. Thus, if such changes and modifications fall within the scope of the claims of this specification and their equivalents, this specification is intended to include such changes and modifications.

Claims

1. A method for initializing garage positioning, characterized in that: The initialization method for garage positioning includes: When the vehicle is in a driving state, obtaining memory track data generated when the vehicle first enters a target area, the memory track data including three-dimensional position information, confidence information, and semantic label information; Determine the basement entrance connection point and the connection point position based on the pitch angle mutation feature in the three-dimensional position information and the confidence jump feature in the confidence information; When the vehicle enters the target area again, the extracted real-time trajectory data is multi-dimensionally matched with the memory trajectory data to generate a trajectory matching degree, wherein the multi-dimensional matching includes three-dimensional position information matching, confidence information matching, and semantic label information matching; When the trajectory matching degree exceeds a preset matching threshold, performing posture initialization based on the connection point position to generate the initial posture of the vehicle; Based on the initial posture and the position of the fixed structure in the target area, the initial posture of the vehicle is optimized to determine an optimized initial posture.

2. The method for initializing garage positioning according to claim 1, characterized in that: The determining of the basement entrance connection point and the location of the connection point based on the pitch angle mutation feature in the three-dimensional position information and the confidence jump feature in the confidence information includes: Determining whether there is a ramp in the target area based on a pitch angle mutation feature in the three-dimensional position information; If so, determining the entry point and exit point of the ramp based on the angle change of the pitch angle mutation feature and a preset angle change threshold; Determining a confidence state transition point position based on a confidence change amount of the confidence transition feature in the confidence information and a preset confidence change threshold; Based on the position of the slope entry point, the position of the slope exit point and the position of the confidence state jump point, the basement entrance connection point and the connection point position are determined.

3. The method for initializing garage positioning according to claim 2, characterized in that: The preset angle change threshold includes a first preset angle change threshold and a second preset angle change threshold, the first preset angle change threshold being greater than the second preset angle change threshold, and determining the entry point and exit point of the ramp based on the angle change amount of the pitch angle sudden change feature and the preset angle change threshold includes: When the vehicle is at a first position, if the angle change is greater than the first preset angle change threshold, the first position is used as the entry point of the ramp; When the vehicle is at the second position, if the angle change is less than the second preset angle change threshold, the second position is used as the exit point of the slope.

4. The method for initializing garage positioning according to claim 2, characterized in that: The determining of the confidence state transition point position based on the confidence change amount of the confidence transition feature in the confidence information and a preset confidence change threshold comprises: When the vehicle is at the third position, if the confidence change is greater than the preset confidence change threshold, the third position is used as a confidence state transition point position.

5. The method for initializing garage positioning according to claim 4, characterized in that: After determining whether there is a ramp in the target area based on the pitch angle mutation feature in the three-dimensional position information, the method further includes: If the ramp does not exist in the target area, determining the position of the confidence state jump point based on the confidence change amount of the confidence jump feature in the confidence information and a preset confidence change threshold; When it is determined that there is a parking space for the vehicle at the third position, the position of the confidence state jump point is determined to be the underground garage entrance connection point and the connection point position is determined.

6. The method for initializing garage positioning according to claim 1, characterized in that: Optimizing the initial posture of the vehicle based on the initial posture and the position of the fixed structure in the target area to determine the optimized initial posture includes: Mapping the semantic label information corresponding to the initial pose to a preset global coordinate system to construct a local semantic map; The position of the fixed structure in the local semantic map is matched with a pre-built map library, the initial posture of the vehicle is optimized, and an optimized initial posture is determined.

7. The method for initializing garage positioning according to claim 6, characterized in that: Matching the position of the fixed structure in the local semantic map to a pre-built map library, optimizing the initial posture of the vehicle, and determining the optimized initial posture includes: extracting three-dimensional coordinate data of the fixed structure from the local semantic map; performing difference calculation between the three-dimensional coordinate data of the fixed structure and the standard three-dimensional coordinate data of the pre-built structure in the pre-built map library to determine a minimum residual, and using the minimum residual as the deviation posture of the vehicle; Based on the deviation posture, the initial posture of the vehicle is optimized to determine an optimized initial posture.

8. A garage positioning initialization device, characterized in that: The initialization device for garage positioning includes: an acquisition module, configured to acquire, when the vehicle is in a driving state, memory track data generated when the vehicle first enters a target area, the memory track data including three-dimensional position information, confidence information, and semantic label information; A determination module, configured to determine a connection point of an underground garage entrance and a position of the connection point based on a pitch angle mutation feature in the three-dimensional position information and a confidence jump feature in the confidence information; a matching module, configured to perform multi-dimensional matching between the extracted real-time trajectory data and the stored trajectory data when the vehicle re-enters the target area, and generate a trajectory matching degree, wherein the multi-dimensional matching includes matching of three-dimensional position information, matching of confidence information, and matching of semantic label information; A generation module, configured to perform posture initialization based on the position of the connection point to generate an initial posture of the vehicle when the trajectory matching degree exceeds a preset matching threshold; An optimization module is used to optimize the initial posture of the vehicle based on the initial posture and the position of the fixed structure in the target area, and determine an optimized initial posture.

9. An electronic device, characterized in that: include: A processor, a memory, and a bus, wherein the memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor and the memory communicate via the bus, and the machine-readable instructions are executed by the processor to execute the steps of the garage positioning initialization method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for initializing garage positioning according to any one of claims 1 to 7 are executed.

Images