A method for vehicle positioning by updating Gaussian distribution probability

Abstract

The embodiment of the present application discloses a method for locating a vehicle by updating the probability of Gaussian distribution, which can improve the accuracy of vehicle positioning. The method includes: initializing a Gaussian distribution function; iteratively updating the Gaussian distribution function according to the real position data obtained from the vehicle and the position data estimated by using the Gaussian distribution function; when the Gaussian distribution function meets the preset convergence condition, the area under the curve in the position area centered on each candidate position point of the vehicle is calculated by the Gaussian distribution function to obtain the position hit probability corresponding to each candidate position point; with the goal of maximizing the sum of the position hit probabilities, n target candidate position points of the vehicle are determined from each candidate position point; and the position positioning result of the vehicle is determined according to the n target candidate position points.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method for locating a vehicle position by updating Gaussian distribution probability. Background Art

[0002] Currently, vehicles mainly use Global Positioning System (GPS) technology for vehicle positioning to support the application of intelligent vehicle technologies such as vehicle collision avoidance and assisted driving.

[0003] However, when using GPS technology to locate a vehicle, it is easily interfered by the external environment, resulting in low accuracy in vehicle positioning. For example, when a vehicle is driving in a city, the urban environment with tall buildings will cause great interference to the GPS signal, resulting in large errors in the positioning results. Summary of the invention

[0004] The purpose of the embodiment of the present application is to provide a method for locating a vehicle by updating the Gaussian distribution probability, which can improve the accuracy of vehicle positioning.

[0005] To solve the above technical problems, in a first aspect, an embodiment of the present application provides a method for locating a vehicle position by updating a Gaussian distribution probability, the method comprising:

[0006] Initializing a Gaussian distribution function, where the Gaussian distribution function is used to characterize the position distribution probability of the vehicle;

[0007] Iteratively updating the Gaussian distribution function according to real position data acquired for the vehicle and position data estimated using the Gaussian distribution function;

[0008] When the Gaussian distribution function satisfies a preset convergence condition, the area under the curve in the position region centered on each candidate position point of the vehicle is calculated by using the Gaussian distribution function to obtain the position hit probability corresponding to each candidate position point;

[0009] Taking the maximization of the sum of the position hit probabilities as the goal, n target candidate position points of the vehicle are determined from the candidate position points, where n is the number of target candidate position points selected in advance;

[0010] The position positioning result of the vehicle is determined according to the n target candidate position points.

[0011] As a possible implementation, with the goal of maximizing the sum of the position hit probabilities, n target candidate position points of the vehicle are determined from the candidate position points, including:

[0012] Dividing each of the candidate position points into a plurality of position point groups, each of the position point groups including n candidate position points;

[0013] For each of the position point groups, the sum of the position hit probabilities corresponding to the position point group is updated according to the distances between the candidate position points in the position point group, so that the sum of the position hit probabilities corresponding to the position point group and the distances between the candidate position points in the position point group are in a positive correlation relationship;

[0014] The n candidate position points included in the position point group with the largest sum of updated position hit probabilities are determined as the n target candidate position points.

[0015] As a possible implementation manner, updating the sum of the position hit probabilities corresponding to the position point group according to the distances between the candidate position points in the position point group includes:

[0016] Updating the position hit probability corresponding to each candidate position point in the position point group by using the first formula or the second formula;

[0017] The first formula is expressed as follows:

[0018]

[0019] The second formula is expressed as follows:

[0020]

[0021] in, represents the updated position hit probability corresponding to the i-th candidate position point in the position point group; represents the position hit probability before updating corresponding to the i-th candidate position point; For return and 0, whichever is greater. It is used to sum the return values ​​of max() corresponding to every two candidate position points in the position point group, Represents the i-th candidate position point in the position point group and the jth candidate location point The distance between Represents the i-th candidate position point in the position point group The center of the candidate location point grouped with the location point The distance between them; n represents the number of candidate position points in the position point group; , Indicates the constant coefficient of the setting.

[0022] As a possible implementation, iteratively updating the Gaussian distribution function according to the real position data acquired for the vehicle and the position data estimated using the Gaussian distribution function includes:

[0023] Using the currently obtained Gaussian distribution function, determining a plurality of candidate position points of the vehicle and position hit probabilities corresponding to each of the plurality of candidate position points;

[0024] Sorting the plurality of candidate position points in descending order of the position hit probability to obtain a candidate position point sequence;

[0025] Executing a matching process to sequentially match each candidate position point in the candidate position point sequence with the real position data;

[0026] When the current candidate position point in the candidate position point sequence successfully matches the real position data, the matching process is terminated, and according to the sorting position of the current candidate position point, it is determined whether the currently obtained Gaussian distribution function satisfies the preset convergence condition, and the preset convergence condition includes: the candidate position point ranked first in the candidate position point sequence successfully matches the real position data;

[0027] When the currently obtained Gaussian distribution function does not satisfy the preset convergence condition, the currently obtained Gaussian distribution function is updated according to the real position data, and it is further determined whether the updated Gaussian distribution function satisfies the preset convergence condition.

[0028] As a possible implementation, determining the position positioning result of the vehicle based on the n target candidate position points includes: determining the target candidate position point with the highest position hit probability among the n target candidate position points as the position positioning result of the vehicle.

[0029] As a possible implementation, the Gaussian distribution function is expressed as follows:

[0030]

[0031] in, represents the Gaussian distribution function, Indicates the vehicle position, represents the mean of the vehicle positions, Represents the covariance matrix The norm of represents pi, and exp() represents the exponential function;

[0032] When the vehicle position is represented by two-dimensional position coordinates, the covariance matrix It is expressed as follows:

[0033]

[0034] in, and Represents the standard deviation of the vehicle position in the x-axis and y-axis directions, Represents the covariance coefficient between the vehicle position variable in the x-axis direction and the variable in the y-axis direction.

[0035] As a possible implementation manner, determining the position positioning result of the vehicle according to the n target candidate position points includes:

[0036] Taking the target candidate position point with the highest position hit probability as the center point of the linear search line, searching from the center point along the linear search line to both sides of the center point within a set distance range, and taking each of the searched target candidate position points among the n target candidate position points as each first candidate position point;

[0037] Taking the target candidate position point with the highest position hit probability as the center point of the spiral search line, searching along the spiral search line within a set radius from the center point, and taking each of the searched target candidate position points among the n target candidate position points as each second candidate position point;

[0038] Determine a weight for each of the first candidate position points according to the position hit probability, and calculate a weighted sum of each of the first candidate position points to obtain first positioning data;

[0039] Determine a weight for each of the second candidate position points according to the position hit probability, and calculate a weighted sum of each of the second candidate position points to obtain second positioning data;

[0040] The weights of the first positioning data and the second positioning data are determined according to the road condition where the vehicle is located, and a weighted sum of the first positioning data and the second positioning data is calculated to obtain a position positioning result of the vehicle.

[0041] According to a second aspect of an embodiment of the present application, there is provided a device for locating a vehicle position by updating a Gaussian distribution probability, the device comprising:

[0042] An initialization module, used for initializing a Gaussian distribution function, wherein the Gaussian distribution function is used for characterizing a position distribution probability of a vehicle;

[0043] An updating module, configured to iteratively update the Gaussian distribution function according to real position data acquired for the vehicle and position data estimated using the Gaussian distribution function;

[0044] A calculation module, configured to calculate the area under the curve in a position region centered on each candidate position point of the vehicle by using the Gaussian distribution function when the Gaussian distribution function satisfies a preset convergence condition, so as to obtain a position hit probability corresponding to each candidate position point;

[0045] A screening module, for determining n target candidate position points of the vehicle from the candidate position points with the goal of maximizing the sum of the position hit probabilities, wherein n is a preset number of target candidate position points to be selected;

[0046] A positioning module is used to determine the position positioning result of the vehicle according to the n target candidate position points.

[0047] According to a third aspect of an embodiment of the present application, a computer program product is provided, including a computer program / instruction, which, when executed by a processor, implements the steps of the method for locating a vehicle by updating the Gaussian distribution probability as described in the first aspect.

[0048] In a fourth aspect of an embodiment of the present application, a computer-readable storage medium is provided, on which a computer program / instruction is stored. When the computer program / instruction is executed by a processor, the steps of the method for locating a vehicle by updating the Gaussian distribution probability as described in the first aspect are implemented.

[0049] In a fifth aspect of an embodiment of the present application, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the method for locating a vehicle by updating the probability of a Gaussian distribution as described in the first aspect are implemented.

[0050] It can be seen from the above technical solution that the embodiment of the present application perceives the vehicle position as a Gaussian distribution, and initializes a Gaussian distribution function, and iteratively updates the Gaussian distribution function according to the acquired real position data, so that the vehicle position positioning method becomes a self-learning and adaptive method, that is, it has the ability to learn and adapt to various position errors caused by environmental factors, and selects target candidate position points with the goal of maximizing the sum of position hit probabilities, so that each selected target candidate position point can be closer to and have a greater probability of directly hitting the real position of the vehicle, thereby achieving improved vehicle positioning accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0052] Figure 1 A flowchart of a method for locating a vehicle position by updating a Gaussian distribution probability provided in an embodiment of the present application;

[0053] Figure 2 A flowchart of a method for determining a target candidate location point provided in an embodiment of the present application;

[0054] Figure 3 A schematic diagram of the structure of a device for locating a vehicle position by updating a Gaussian distribution probability provided in an embodiment of the present application;

[0055] Figure 4 A schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0056] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0057] First, see Figure 1 As shown, it is an implementation flow chart of a method for locating a vehicle position by updating a Gaussian distribution probability provided by an embodiment of the present application. The method may include the following steps:

[0058] Step S101: Initialize a Gaussian distribution function, where the Gaussian distribution function is used to characterize the position distribution probability of a vehicle.

[0059] In a specific implementation, the Gaussian distribution function may be initialized according to experience or based on a standard normal distribution, thereby obtaining an initial probability distribution for describing all possible positions of the vehicle at future moments.

[0060] Exemplarily, the Gaussian distribution function It can be expressed as follows:

[0061]

[0062] in, Indicates the vehicle position, represents the mean of the vehicle positions, Represents the covariance matrix The norm of represents pi, and exp() represents the exponential function.

[0063] Optionally, the vehicle position is represented by a two-dimensional position coordinate , then the covariance matrix It can be expressed as follows:

[0064]

[0065] in, and Represents the standard deviation of the vehicle position in the x-axis and y-axis directions, Represents the covariance coefficient between the vehicle position variable in the x-axis direction and the variable in the y-axis direction.

[0066] Step S102: iteratively updating the Gaussian distribution function according to the real position data acquired for the vehicle and the position data estimated using the Gaussian distribution function.

[0067] In specific implementation, the location data returned by different positioning methods when applicable can be used as the real location data of the vehicle. For example, the location data returned by a Bluetooth beacon in a specific area (such as a parking lot), the location data returned by a GPS in an open environment, and the location data returned by a visual positioning device under good lighting conditions can be used as the real location data of the vehicle.

[0068] After obtaining the real position data, the real position data is compared with the position data estimated using the Gaussian distribution function. If the estimated position data is inconsistent with the real position data, the Gaussian distribution function is updated using the real position data (such as updating the mean and covariance of the Gaussian distribution function) so that the Gaussian distribution function learns and adapts to the current position error. Among them, the update of the Gaussian distribution function can be implemented based on algorithms such as Maximum Likelihood Estimation (MLE) or Maximum A Posteriori (MAP). For example, each element in the mean and covariance of the Gaussian distribution function can be given a priori of a standard normal distribution, and the update of the mean and covariance of the Gaussian distribution function is regarded as a regularized parameter optimization problem, and then the problem is solved based on the real position data and MAP.

[0069] Step S103: When the Gaussian distribution function satisfies a preset convergence condition, the area under the curve within the position region centered on each candidate position point of the vehicle is calculated using the Gaussian distribution function to obtain the position hit probability corresponding to each candidate position point.

[0070] In a specific implementation, when the number of updates of the Gaussian distribution function reaches a preset number threshold, or when the accuracy of the position data estimated using the Gaussian distribution function for multiple consecutive times reaches a preset accuracy threshold, it can be determined that the Gaussian distribution function meets the preset convergence condition, and then the Gaussian distribution function can be used to locate the vehicle position.

[0071] When using the Gaussian distribution function to locate the vehicle, the circle center area centered on the candidate position point can be used as the position area corresponding to the candidate position point according to a preset radius (which can be determined based on the actual acceptable accuracy loss), and the area under the curve of the Gaussian distribution function is calculated for the position area to obtain the position hit probability corresponding to the candidate position point, that is, the probability that the actual position of the vehicle is located in the position area where the candidate position point is located.

[0072] Step S104: with the goal of maximizing the sum of the position hit probabilities, n target candidate position points of the vehicle are determined from the various candidate position points, wherein n is a preset number of target candidate position points to be selected.

[0073] In specific implementation, by maximizing the sum of position hit probabilities, n target candidate position points closest to the vehicle's actual position can be found, that is, the vehicle's actual position may be among these target candidate position points or may be close to these target candidate position points.

[0074] Step S105: Determine the position positioning result of the vehicle according to the n target candidate position points.

[0075] In a specific implementation, if n is equal to 1, the one target candidate position point found is directly used as the position positioning result of the vehicle; if n is greater than 1, the vehicle position is further estimated based on the multiple target candidate position points found, such as finding the overlapping area of ​​the position areas of the multiple target candidate position points, and using the overlapping area (or the center of the overlapping area) as the position positioning result of the vehicle.

[0076] It can be seen from the above technical solution that the embodiment of the present application perceives the vehicle position as a Gaussian distribution, and initializes a Gaussian distribution function, and iteratively updates the Gaussian distribution function according to the acquired real position data, so that the vehicle position positioning method becomes a self-learning and adaptive method, that is, it has the ability to learn and adapt to various position errors caused by environmental factors, and selects target candidate position points with the goal of maximizing the sum of position hit probabilities, so that each selected target candidate position point can be closer to and have a greater probability of directly hitting the real position of the vehicle, thereby achieving improved vehicle positioning accuracy.

[0077] In one embodiment, see Figure 2 The target candidate location point determination method is shown in the flowchart of the implementation of the target candidate location point determination method, the target candidate location point determination method includes:

[0078] Step S201: Divide the candidate location points into a plurality of location point groups, each of the location point groups including n candidate location points.

[0079] In a specific implementation, the same candidate location point may be divided into different location point groups, but it is necessary to ensure that the n candidate location points included in different location point groups are not completely the same.

[0080] Step S202: For each of the position point groups, the sum of the position hit probabilities corresponding to the position point group is updated according to the distances between the candidate position points in the position point group, so that the sum of the position hit probabilities corresponding to the position point group is positively correlated with the distances between the candidate position points in the position point group.

[0081] In a specific implementation, the average value of the distance between each candidate position point in the position point group can be calculated, and the sum of the sum of the position hit probabilities corresponding to the position point group and the average value can be used as the sum of the updated position hit probabilities. In this way, the sum of the position hit probabilities corresponding to the position point group will increase as the distance between the candidate position points in the position point group increases, so that it conforms to the positive correlation relationship.

[0082] Optionally, updating the sum of the position hit probabilities corresponding to the position point group according to the distances between the candidate position points in the position point group includes:

[0083] Updating the position hit probability corresponding to each candidate position point in the position point group by using the first formula or the second formula;

[0084] The first formula is expressed as follows:

[0085]

[0086] The second formula is expressed as follows:

[0087]

[0088] in, represents the updated position hit probability corresponding to the i-th candidate position point in the position point group; represents the position hit probability before updating corresponding to the i-th candidate position point; For return and 0, whichever is greater. It is used to sum the return values ​​of max() corresponding to every two candidate position points in the position point group, Represents the i-th candidate position point in the position point group and the jth candidate location point The distance between Represents the i-th candidate position point in the position point group The center of the candidate location point grouped with the location point The distance between them; n represents the number of candidate position points in the position point group; , Represents the set constant coefficient, which can be used to control the extent to which the location hit probability changes with the distance between candidate location points.

[0089] In a specific implementation, the position hit probabilities corresponding to the candidate position points in the position point group are updated by using the first formula, so that the candidate position points that are farther away from other candidate position points have a greater position hit probability, and the candidate position points that are closer to other candidate position points have a smaller position hit probability.

[0090] By updating the location hit probabilities corresponding to the candidate location points in the location point group through the second formula, the candidate location points that are farther away from the center of the candidate location points in the location point group can have a greater location hit probability, and the candidate location points that are closer to the center of the candidate location points can have a smaller location hit probability.

[0091] Step S203: n candidate position points included in the position point group with the largest sum of updated position hit probabilities are determined as the n target candidate position points.

[0092] In this embodiment, considering that when n target candidate position points are selected with the goal of maximizing the sum of position hit probabilities, there may be a problem of excessive overlap among the n selected target candidate position points, which makes the n selected target candidate position points likely to be densely distributed on one side of the actual position of the vehicle, rather than being more scattered around the actual position of the vehicle, thereby affecting the final positioning result. For example, a large positioning error will occur when the center of the n candidate position points is used as the positioning result.

[0093] The embodiment of the present application updates the sum of the position hit probabilities corresponding to the position point groups according to the distances between the candidate position points, which can avoid the above-mentioned excessive overlap problem among the selected n target candidate position points, thereby ensuring the vehicle positioning accuracy.

[0094] Optionally, iteratively updating the Gaussian distribution function according to real position data acquired for the vehicle and position data estimated using the Gaussian distribution function comprises:

[0095] Using the currently obtained Gaussian distribution function, determining a plurality of candidate position points of the vehicle and position hit probabilities corresponding to each of the plurality of candidate position points;

[0096] Sorting the plurality of candidate position points in descending order of the position hit probability to obtain a candidate position point sequence;

[0097] Executing a matching process to sequentially match each candidate position point in the candidate position point sequence with the real position data;

[0098] When the current candidate position point in the candidate position point sequence successfully matches the real position data, the matching process is terminated, and according to the sorting position of the current candidate position point, it is determined whether the currently obtained Gaussian distribution function satisfies the preset convergence condition, and the preset convergence condition includes: the candidate position point ranked first in the candidate position point sequence successfully matches the real position data;

[0099] When the currently obtained Gaussian distribution function does not satisfy the preset convergence condition, the currently obtained Gaussian distribution function is updated according to the real position data, and it is further determined whether the updated Gaussian distribution function satisfies the preset convergence condition.

[0100] In the specific implementation, each candidate location point is matched with the real location data in order from large to small in terms of location hit probability. If the distance between the candidate location point and the real location data is less than a set value (or the real location data belongs to the location area of ​​the candidate location point), it is confirmed that the candidate location point and the real location data are matched successfully.

[0101] After finding the candidate position point that successfully matches the real position data, it is first determined whether the candidate position point is the candidate position point with the highest position hit probability (ie ranked first) in the candidate position point sequence.

[0102] If the candidate position point is the candidate position point with the highest position hit probability in the candidate position point sequence, it is determined that the Gaussian distribution function meets the preset convergence condition, and then the Gaussian distribution function is used to locate the vehicle, and the Gaussian distribution function can continue to be updated to ensure the position error adaptability of the Gaussian distribution function.

[0103] If the candidate position point is not the candidate position point with the highest position hit probability in the candidate position point sequence, it is determined that the Gaussian distribution function does not meet the preset convergence condition, the Gaussian distribution function is not used to locate the vehicle, and the Gaussian distribution function is updated and the matching process is executed to continue to determine whether the updated Gaussian distribution function meets the preset convergence condition.

[0104] In this embodiment, the preset convergence condition is that the candidate position point ranked first in the candidate position point sequence is successfully matched with the actual position data, so as to hit the actual position of the vehicle for the first time and iteratively update the Gaussian distribution function, thereby ensuring that the distance between the candidate position point with the highest position hit probability and the actual position of the vehicle is not too large, so that the distribution of the n target candidate position points (whose distribution will usually converge to the target candidate position point with the highest position hit probability) will not deviate too much from the actual position of the vehicle, thereby ensuring the accuracy of vehicle positioning.

[0105] Optionally, after successfully matching the first candidate position point in the candidate position point sequence with the actual position data as the preset convergence condition, and iteratively updating the Gaussian distribution function with the goal of hitting the actual position of the vehicle for the first time, in the vehicle positioning stage, the target candidate position point with the highest position hit probability among the n target candidate position points can be directly determined as the vehicle positioning result, thereby improving the positioning efficiency while ensuring the positioning accuracy.

[0106] Optionally, in one embodiment, determining the position positioning result of the vehicle according to the n target candidate position points includes:

[0107] Taking the target candidate position point with the highest position hit probability as the center point of the linear search line, searching from the center point along the linear search line to both sides of the center point within a set distance range, and taking each of the searched target candidate position points among the n target candidate position points as each first candidate position point;

[0108] Taking the target candidate position point with the highest position hit probability as the center point of the spiral search line, searching along the spiral search line within a set radius from the center point, and taking each of the searched target candidate position points among the n target candidate position points as each second candidate position point;

[0109] Determine a weight for each of the first candidate position points according to the position hit probability, and calculate a weighted sum of each of the first candidate position points to obtain first positioning data;

[0110] Determine a weight for each of the second candidate position points according to the position hit probability, and calculate a weighted sum of each of the second candidate position points to obtain second positioning data;

[0111] The weights of the first positioning data and the second positioning data are determined according to the road condition where the vehicle is located, and a weighted sum of the first positioning data and the second positioning data is calculated to obtain a position positioning result of the vehicle.

[0112] In this embodiment, it is considered that the vehicle's motion forms are generally two types: rotational motion and translational motion; therefore, the n target candidate position points are further screened using the spiral search line and the straight search line adapted to the two motion forms to obtain the first candidate position points (which include the target candidate position points with the highest probability of hitting the position) and the second candidate position points (which include the target candidate position points with the highest probability of hitting the position).

[0113] Then, a larger weight is set for the first candidate position point and the second candidate position point with a larger position hit probability, thereby obtaining the first positioning data corresponding to the translation movement and the second positioning data corresponding to the rotation movement in the form of a weighted sum.

[0114] Subsequently, based on the road condition where the vehicle is located, the weights of the first positioning data and the second positioning data are determined. The mapping relationship between the road condition and the weight can be determined based on experience or by machine learning.

[0115] For example, considering that the vehicle mainly performs translational movement when it is on a straight road section, the target candidate position point with the highest position hit probability and the target candidate position points on both sides (i.e., the above-mentioned first candidate position points) are closer and have a higher probability of hitting the real position data of the vehicle. For example, when the vehicle moves horizontally (or vertically) along a straight road section, based on the horizontal search line (or vertical search line) corresponding to the straight road section, the target candidate position point with the highest position hit probability and the target candidate position points on its left and right sides (or upper and lower sides) and within the set distance range will be used as the first candidate position points to calculate the first positioning data, so a larger weight can be set for the first positioning data, thereby further improving the accuracy of the vehicle's position positioning results on the straight road section.

[0116] For another example, considering that the vehicle mainly performs rotational motion when it is on a multi-bend road section, the target candidate position point with the highest probability of position hit and the target candidate position points spirally distributed within a set radius around it (i.e., the above-mentioned second candidate position points) are closer and have a higher probability of hitting the real position data of the vehicle. For example, when the vehicle is rotating along a bend in a multi-bend road section, based on the spiral search line (such as the Archimedean spiral) corresponding to the multi-bend road section, the target candidate position point with the highest probability of position hit and the target candidate position points spirally distributed within a set radius around it will be used as the second candidate position points to calculate the second positioning data, so a larger weight can be set for the second positioning data, thereby further improving the accuracy of the vehicle's position positioning results on the multi-bend road section.

[0117] It should be noted that, for the method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the embodiments of the present application are not limited by the described order of actions, because according to the embodiments of the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present application.

[0118] The present application also provides a device for locating a vehicle position by updating the Gaussian distribution probability. Figure 3 As shown, the device comprises:

[0119] An initialization module, used for initializing a Gaussian distribution function, wherein the Gaussian distribution function is used for characterizing a position distribution probability of a vehicle;

[0120] An updating module, configured to iteratively update the Gaussian distribution function according to real position data acquired for the vehicle and position data estimated using the Gaussian distribution function;

[0121] A calculation module, configured to calculate the area under the curve in a position region centered on each candidate position point of the vehicle by using the Gaussian distribution function when the Gaussian distribution function satisfies a preset convergence condition, so as to obtain a position hit probability corresponding to each candidate position point;

[0122] A screening module, for determining n target candidate position points of the vehicle from the candidate position points with the goal of maximizing the sum of the position hit probabilities, wherein n is a preset number of target candidate position points to be selected;

[0123] A positioning module is used to determine the position positioning result of the vehicle according to the n target candidate position points.

[0124] Optionally, the screening module includes:

[0125] A first screening submodule, configured to divide the candidate position points into a plurality of position point groups, each of the position point groups including n candidate position points;

[0126] A second screening submodule is used to update, for each of the position point groups, the sum of the position hit probabilities corresponding to the position point group according to the distances between the candidate position points in the position point group, so that the sum of the position hit probabilities corresponding to the position point group and the distances between the candidate position points in the position point group are in a positive correlation relationship;

[0127] The third screening submodule is used to determine the n candidate position points contained in the position point group with the largest sum of updated position hit probabilities as the n target candidate position points.

[0128] Optionally, the second screening submodule includes:

[0129] A fourth screening submodule, configured to update the position hit probability corresponding to each candidate position point in the position point group by using the first formula or the second formula;

[0130] The first formula is expressed as follows:

[0131]

[0132] The second formula is expressed as follows:

[0133]

[0134] in, represents the updated position hit probability corresponding to the i-th candidate position point in the position point group; represents the position hit probability before updating corresponding to the i-th candidate position point; For return and 0, whichever is greater. It is used to sum the return values ​​of max() corresponding to every two candidate position points in the position point group, Represents the i-th candidate position point in the position point group and the jth candidate location point The distance between Represents the i-th candidate position point in the position point group The center of the candidate location point grouped with the location point The distance between them; n represents the number of candidate position points in the position point group; , Indicates the constant coefficient of the setting.

[0135] Optionally, the update module includes:

[0136] A first updating submodule, configured to determine a plurality of candidate position points of the vehicle and position hit probabilities corresponding to each of the plurality of candidate position points using the currently obtained Gaussian distribution function;

[0137] A second updating submodule is used to sort the multiple candidate position points in descending order of the position hit probability to obtain a candidate position point sequence;

[0138] A third updating submodule is used to execute a matching process to sequentially match each candidate position point in the candidate position point sequence with the real position data;

[0139] The fourth updating submodule is used to terminate the matching process if the current candidate position point in the candidate position point sequence successfully matches the real position data, and determine whether the currently obtained Gaussian distribution function satisfies the preset convergence condition according to the sorting position of the current candidate position point, and the preset convergence condition includes: the candidate position point ranked first in the candidate position point sequence successfully matches the real position data;

[0140] The fifth updating submodule is used to update the currently obtained Gaussian distribution function according to the real position data when the currently obtained Gaussian distribution function does not meet the preset convergence condition, and continue to determine whether the updated Gaussian distribution function meets the preset convergence condition.

[0141] Optionally, the positioning module includes:

[0142] The first positioning submodule is used to determine the target candidate position point with the highest position hit probability among the n target candidate position points as the position positioning result of the vehicle.

[0143] Optionally, the Gaussian distribution function is expressed as follows:

[0144]

[0145] in, represents the Gaussian distribution function, Indicates the vehicle position, represents the mean of the vehicle positions, Represents the covariance matrix The norm of represents pi, and exp() represents the exponential function;

[0146] When the vehicle position is represented by two-dimensional position coordinates, the covariance matrix It is expressed as follows:

[0147]

[0148] in, and Represents the standard deviation of the vehicle position in the x-axis and y-axis directions, Represents the covariance coefficient between the vehicle position variable in the x-axis direction and the variable in the y-axis direction.

[0149] Optionally, the positioning module includes:

[0150] The second positioning submodule is used to use the target candidate position point with the highest position hit probability as the center point of the linear search line, search from the center point along the linear search line to both sides of the center point within a set distance range, and use each searched target candidate position point among the n target candidate position points as each first candidate position point;

[0151] A third positioning submodule is used to use the target candidate position point with the highest position hit probability as the center point of the spiral search line, search along the spiral search line within a set radius from the center point, and use each searched target candidate position point among the n target candidate position points as each second candidate position point;

[0152] a fourth positioning submodule, configured to determine a weight for each of the first candidate position points according to the position hit probability, and calculate a weighted sum of each of the first candidate position points to obtain first positioning data;

[0153] a fifth positioning submodule, configured to determine a weight for each of the second candidate position points according to the position hit probability, and calculate a weighted sum of each of the second candidate position points to obtain second positioning data;

[0154] The sixth positioning submodule is used to determine the weights of the first positioning data and the second positioning data according to the road conditions of the vehicle, and calculate the weighted sum of the first positioning data and the second positioning data to obtain the position positioning result of the vehicle.

[0155] It can be seen from the above technical solution that the embodiment of the present application perceives the vehicle position as a Gaussian distribution, and initializes a Gaussian distribution function, and iteratively updates the Gaussian distribution function according to the acquired real position data, so that the vehicle position positioning method becomes a self-learning and adaptive method, that is, it has the ability to learn and adapt to various position errors caused by environmental factors, and selects target candidate position points with the goal of maximizing the sum of position hit probabilities, so that each selected target candidate position point can be closer to and have a greater probability of directly hitting the real position of the vehicle, thereby achieving improved vehicle positioning accuracy.

[0156] The present application also provides an electronic device, referring to Figure 4 , Figure 4 Schematic diagram of an electronic device proposed in an embodiment of the present application. Figure 4 As shown, the electronic device 100 includes: a memory 110 and a processor 120. The memory 110 and the processor 120 are connected via a bus communication. A computer program is stored in the memory 110. The computer program can be run on the processor 120 to implement the steps in the method for locating the vehicle position by updating the Gaussian distribution probability disclosed in the embodiment of the present application.

[0157] The embodiment of the present application also provides a computer-readable storage medium having a computer program / instruction stored thereon. When the computer program / instruction is executed by a processor, the method for locating a vehicle by updating the Gaussian distribution probability as disclosed in the embodiment of the present application is implemented.

[0158] The embodiment of the present application also provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the method for locating a vehicle position by updating the Gaussian distribution probability as disclosed in the embodiment of the present application.

[0159] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0160] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, devices 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-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0161] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of the methods, systems, devices, storage media, and 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 processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal 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.

[0162] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0163] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0164] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used 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 "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including 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 terminal device. In the absence of further restrictions, the elements defined by the sentence "including one..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0165] The above is a detailed introduction to a method for locating a vehicle position by updating the probability of Gaussian distribution provided by the present application. This article uses specific examples to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for general technical personnel in this field, according to the ideas of the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A method for locating a vehicle position by updating a Gaussian distribution probability, characterized in that: The method comprises: Initializing a Gaussian distribution function, where the Gaussian distribution function is used to characterize the position distribution probability of the vehicle; Iteratively updating the Gaussian distribution function according to the real position data acquired for the vehicle and the position data estimated using the Gaussian distribution function; When the Gaussian distribution function satisfies a preset convergence condition, the area under the curve in the position region centered on each candidate position point of the vehicle is calculated by using the Gaussian distribution function to obtain the position hit probability corresponding to each candidate position point; Taking the maximization of the sum of the position hit probabilities as the goal, n target candidate position points of the vehicle are determined from the candidate position points, where n is the number of target candidate position points selected in advance; Determine the position positioning result of the vehicle according to the n target candidate position points; With the goal of maximizing the sum of the position hit probabilities, n target candidate position points of the vehicle are determined from the candidate position points, including: Dividing each of the candidate position points into a plurality of position point groups, each of the position point groups including n candidate position points; For each of the position point groups, the sum of the position hit probabilities corresponding to the position point group is updated according to the distances between the candidate position points in the position point group, so that the sum of the position hit probabilities corresponding to the position point group and the distances between the candidate position points in the position point group are in a positive correlation relationship; The n candidate position points included in the position point group with the largest sum of updated position hit probabilities are determined as the n target candidate position points.

2. The method according to claim 1, characterized in that The method of updating the sum of the position hit probabilities corresponding to the position point grouping according to the distances between the candidate position points in the position point grouping includes: Updating the position hit probability corresponding to each candidate position point in the position point group by using the first formula or the second formula; The first formula is expressed as follows: The second formula is expressed as follows: in, represents the updated position hit probability corresponding to the i-th candidate position point in the position point group; represents the position hit probability before updating corresponding to the i-th candidate position point; For return and 0, whichever is greater. It is used to sum the return values ​​of max() corresponding to every two candidate position points in the position point group, Represents the i-th candidate position point in the position point group and the jth candidate location point The distance between Represents the i-th candidate position point in the position point group The center of the candidate location point grouped with the location point The distance between them; n represents the number of candidate position points in the position point group; , Indicates the constant coefficient of the setting.

3. The method according to claim 1, characterized in that Iteratively updating the Gaussian distribution function according to the real position data acquired for the vehicle and the position data estimated using the Gaussian distribution function, including: Using the currently obtained Gaussian distribution function, determining a plurality of candidate position points of the vehicle and position hit probabilities corresponding to each of the plurality of candidate position points; Sorting the plurality of candidate position points in descending order of the position hit probability to obtain a candidate position point sequence; Executing a matching process to sequentially match each candidate position point in the candidate position point sequence with the real position data; When the current candidate position point in the candidate position point sequence successfully matches the real position data, the matching process is terminated, and according to the sorting position of the current candidate position point, it is determined whether the currently obtained Gaussian distribution function satisfies the preset convergence condition, and the preset convergence condition includes: the candidate position point ranked first in the candidate position point sequence successfully matches the real position data; When the currently obtained Gaussian distribution function does not satisfy the preset convergence condition, the currently obtained Gaussian distribution function is updated according to the real position data, and it is further determined whether the updated Gaussian distribution function satisfies the preset convergence condition.

4. The method according to claim 3, characterized in that Determining a position positioning result of the vehicle according to the n target candidate position points includes: determining a target candidate position point with the highest position hit probability among the n target candidate position points as the position positioning result of the vehicle.

5. The method according to claim 1, characterized in that The Gaussian distribution function is expressed as follows: in, represents the Gaussian distribution function, Indicates the vehicle position, represents the mean of the vehicle positions, Represents the covariance matrix The norm of represents pi, and exp() represents the exponential function; When the vehicle position is represented by two-dimensional position coordinates, the covariance matrix It is expressed as follows: in, and Represents the standard deviation of the vehicle position in the x-axis and y-axis directions, Represents the covariance coefficient between the vehicle position variable in the x-axis direction and the variable in the y-axis direction.

6. The method according to any one of claims 1 to 5, characterized in that: Determining the position positioning result of the vehicle according to the n target candidate position points includes: Taking the target candidate position point with the highest position hit probability as the center point of the linear search line, searching from the center point along the linear search line to both sides of the center point within a set distance range, and taking each of the searched target candidate position points among the n target candidate position points as each first candidate position point; Taking the target candidate position point with the highest position hit probability as the center point of the spiral search line, searching along the spiral search line within a set radius from the center point, and taking each of the searched target candidate position points among the n target candidate position points as each second candidate position point; Determine a weight for each of the first candidate position points according to the position hit probability, and calculate a weighted sum of each of the first candidate position points to obtain first positioning data; Determine a weight for each of the second candidate position points according to the position hit probability, and calculate a weighted sum of each of the second candidate position points to obtain second positioning data; The weights of the first positioning data and the second positioning data are determined according to the road condition where the vehicle is located, and a weighted sum of the first positioning data and the second positioning data is calculated to obtain a position positioning result of the vehicle.

7. A device for locating a vehicle position by updating Gaussian distribution probability, characterized in that: The device comprises: An initialization module, used for initializing a Gaussian distribution function, wherein the Gaussian distribution function is used for characterizing a position distribution probability of a vehicle; An updating module, configured to iteratively update the Gaussian distribution function according to real position data acquired for the vehicle and position data estimated using the Gaussian distribution function; A calculation module, configured to calculate the area under the curve in a position region centered on each candidate position point of the vehicle by using the Gaussian distribution function when the Gaussian distribution function satisfies a preset convergence condition, so as to obtain a position hit probability corresponding to each candidate position point; A screening module, for determining n target candidate position points of the vehicle from the candidate position points with the goal of maximizing the sum of the position hit probabilities, wherein n is a preset number of target candidate position points to be selected; A positioning module, used to determine the position positioning result of the vehicle according to the n target candidate position points; The screening module comprises: A first screening submodule, configured to divide the candidate position points into a plurality of position point groups, each of the position point groups including n candidate position points; A second screening submodule is used to update, for each of the position point groups, the sum of the position hit probabilities corresponding to the position point group according to the distances between the candidate position points in the position point group, so that the sum of the position hit probabilities corresponding to the position point group and the distances between the candidate position points in the position point group are in a positive correlation relationship; The third screening submodule is used to determine the n candidate position points contained in the position point group with the largest sum of updated position hit probabilities as the n target candidate position points.

8. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instruction is executed by a processor, the method for locating a vehicle position by updating a Gaussian distribution probability as claimed in any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instruction is executed by a processor, the method for locating a vehicle position by updating a Gaussian distribution probability as claimed in any one of claims 1 to 6 is implemented.

Images