A method, apparatus, and computer device for optimizing a positioning algorithm.

Abstract

This invention provides a method, apparatus, and computer device for optimizing a positioning algorithm. The method includes: acquiring an initial positioning algorithm and several influencing parameters related to the initial positioning algorithm; constructing several optimization modules and determining the priority of each optimization module, the initial value of each influencing parameter, and its value range; acquiring the initial positioning result obtained based on the initial positioning algorithm and determining the optimization module with the highest priority as the current module; traversing the influencing parameters of the current module and determining the target adjustment value of the influencing parameters of the current module, with the target positioning algorithm based on the target adjustment value showing the greatest improvement in positioning accuracy; continuing to use the optimization module with the next lowest priority as the current module and performing the above optimization operation again until the target adjustment value of the influencing parameters of all optimization modules is determined. This scheme can automatically optimize positioning algorithms with different observation qualities, thereby adapting to different chips, improving positioning accuracy, and reducing repetitive work for R&D personnel.

Description

Technical Field

[0001] This invention relates to the field of positioning technology, and more particularly to a method, apparatus, and computer equipment for optimizing a positioning algorithm. Background Technology

[0002] Global Navigation Satellite System (GNSS) provides global, all-weather, high-precision positioning and navigation. Among these, satellite navigation employing Real-Time Kinematic (RTK) carrier phase differential technology offers particularly high positioning accuracy and is widely used in surveying, mapping, autonomous driving navigation, and various other fields. Algorithms based on GNSS technology for positioning are called GNSS positioning algorithms. Inertial navigation uses inertial sensors (such as accelerometers and gyroscopes) to measure the acceleration and angular velocity of a moving object, then calculates its position, velocity, and attitude. Algorithms based on inertial navigation for positioning are called inertial navigation positioning algorithms. Other commonly used positioning algorithms include radar positioning algorithms. Furthermore, a fusion positioning algorithm combines at least two of these algorithms, including GNSS, inertial navigation, and radar positioning algorithms.

[0003] However, the inventors discovered through research that the same positioning algorithm cannot perform at its best when applied to different hardware. For example, a GNSS positioning algorithm cannot guarantee the best positioning performance when adapted to different chips and antennas.

[0004] Therefore, there is an urgent need for a method to automatically optimize the parameters of the positioning algorithm for different hardware implementations, so as to ensure that the positioning algorithm can perform its best positioning performance when applied to different hardware. Summary of the Invention

[0005] The purpose of this invention is to provide a method, apparatus, and computer equipment for optimizing positioning algorithms, which can automatically optimize positioning algorithms with different observation qualities, thereby adapting to different chips, improving positioning accuracy, and reducing repetitive work for R&D personnel.

[0006] The technical solution provided by this invention is as follows:

[0007] Firstly, this application provides a method for optimizing a positioning algorithm, comprising the following steps:

[0008] Step 1: Obtain the initial positioning algorithm and obtain several influencing parameters that are related to the initial positioning algorithm;

[0009] Step 2: Construct several optimization modules and determine the priority of each optimization module, the initial value and value range of each influencing parameter, wherein each optimization module contains several associated influencing parameters;

[0010] Step 3: Obtain the initial positioning result obtained according to the initial positioning algorithm, and determine the optimization module with the highest priority as the current module;

[0011] Step 4: Traverse the influence parameters of the current module and determine the target adjustment value of the influence parameters of the current module. When the influence parameters of the current module are set to the target adjustment value, obtain the target positioning algorithm. The target positioning algorithm improves the positioning accuracy by the most compared with the initial positioning result.

[0012] Step 5: Continue to use the next highest priority tuning module as the current module, and execute Step 4 again until the target adjustment values ​​of the influence parameters of all tuning modules are determined.

[0013] This solution acquires several influencing parameters related to the initial positioning algorithm and constructs several optimization modules. Each optimization module contains several associated influencing parameters. During optimization, the highest-priority optimization module is selected as the current module. By iterating through the influencing parameters of the current module, the target adjustment value of the influencing parameters can be determined. When the influencing parameters of the current module are set to the target adjustment value, the positioning accuracy of the target positioning algorithm is improved the most compared to the initial positioning result. This target adjustment value is then used as the basis for the optimization of the next optimization module, and so on, until the target adjustment values ​​of the influencing parameters of all optimization modules are determined, thus achieving the optimization of the positioning algorithm. This solution enables automated optimization of positioning algorithms with different observation qualities, thereby adapting to different chips, improving positioning accuracy, and reducing repetitive work for R&D personnel.

[0014] In some implementations, step 4 further includes:

[0015] For each of the aforementioned optimization modules, the number of adjustments made to the influencing parameters is obtained;

[0016] The influencing parameters are adjusted according to the number of adjustments described.

[0017] The number of adjustments required varies for different influencing parameters.

[0018] In some embodiments, the method further includes:

[0019] The related influence parameters are adjusted uniformly, while the unrelated influence parameters are adjusted individually.

[0020] In some implementations, the influencing parameter includes a target parameter, and the process further includes the following steps prior to step 2:

[0021] Obtain the difference between the first antenna information and the second antenna information, and adjust the target parameters based on the difference;

[0022] Wherein, the first antenna information is the antenna information corresponding to the initial positioning algorithm before adjustment, the second antenna information is the antenna information corresponding to the initial positioning algorithm after adjustment, and the antenna information is the observation information collected by the corresponding antenna in an open scene.

[0023] In some implementations, the positioning accuracy for each adjustment is determined based on the following steps: for each adjustment, obtaining intermediate positioning results under multiple different application scenarios, and calculating the positioning error of the intermediate positioning results;

[0024] Obtain multiple statistical indicators of the positioning error;

[0025] The intermediate positioning result is scored based on the multiple statistical indicators, and the score is used to represent the positioning accuracy of the intermediate positioning result.

[0026] In some implementations, the statistic includes at least one of root mean square error, CEP68, CEP95, CEP99.7, and maximum error.

[0027] In some implementations, traversing the influence parameters of the current module includes: traversing the influence parameters of the current module based on a preset traversal algorithm, wherein the preset traversal algorithm includes at least a binary search method.

[0028] In some implementations, the method further includes: the order of the priority is adjustable.

[0029] Secondly, this application provides a device for optimizing a positioning algorithm, comprising:

[0030] The acquisition module is used to acquire the initial positioning algorithm and acquire several influencing parameters that are related to the initial positioning algorithm;

[0031] A construction module is used to construct several tuning modules and determine the priority of each tuning module, the initial value and the value range of each influence parameter, wherein each tuning module contains several associated influence parameters;

[0032] A preprocessing module is used to obtain the initial positioning result obtained according to the initial positioning algorithm, and to determine the optimization module with the highest priority as the current module;

[0033] The first processing module is used to traverse the influence parameters of the current module and determine the target adjustment value of the influence parameters of the current module. When the influence parameters of the current module are set to the target adjustment value, a target positioning algorithm is obtained. The target positioning result obtained based on the target positioning algorithm has the greatest improvement in positioning accuracy compared with the initial positioning result.

[0034] The second processing module is used to continue to use the next-highest priority tuning module as the current module and execute the processing operation of the first processing module until the target adjustment value of the influence parameters of all the tuning modules is determined.

[0035] Thirdly, this application provides a computer device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the optimization method for the positioning algorithm described in the first aspect.

[0036] The positioning algorithm optimization method, apparatus, and computer equipment provided by this invention enable automated optimization of positioning algorithms with different observation qualities, thereby adapting to different chips, improving positioning accuracy, and reducing repetitive work for R&D personnel. Attached Figure Description

[0037] The preferred embodiments will now be described in a clear and easy-to-understand manner, with reference to the accompanying drawings, to further explain the above-mentioned characteristics, technical features, advantages, and implementation methods of this solution.

[0038] Figure 1 This is a schematic diagram of the overall process of one embodiment of the present invention;

[0039] Figure 2 This is a schematic diagram of the system framework of an embodiment of the present invention. Detailed Implementation

[0040] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the specific implementation methods of the present invention will be described below with reference to the accompanying drawings. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings and other implementation methods can be obtained based on these drawings without any creative effort.

[0041] To keep the drawings concise, only the parts relevant to the invention are shown schematically in each figure, and they do not represent the actual structure of the product. Furthermore, for ease of understanding, in some figures, only one of components with the same structure or function is shown schematically, or only one is labeled. In this document, "one" can mean not only "only one" but also "more than one".

[0042] This invention constructs an optimization model that can automatically optimize data of different observation qualities, ultimately generating a complete GNSS positioning algorithm adapted to that observation quality data. This algorithm can be directly used for engineering implementation, accelerating the algorithm optimization process, improving algorithm performance, and reducing repetitive work for R&D personnel. The following is a detailed description of this solution with reference to the accompanying drawings:

[0043] In one embodiment, refer to the appendix to the specification. Figure 1 This application provides a method for optimizing a positioning algorithm, including the following steps:

[0044] S1. Obtain the initial positioning algorithm and obtain several influencing parameters that are related to the initial positioning algorithm.

[0045] The positioning algorithm can refer to one of several positioning algorithms, such as GNSS positioning algorithm, inertial navigation positioning algorithm, radar positioning algorithm, or fusion positioning algorithm.

[0046] The initial positioning algorithm refers to the positioning algorithm before parameter adjustments are made.

[0047] Influencing parameters refer to the parameters whose values ​​need to be adjusted in the localization algorithm. Different influencing parameters may be related, and these relationships can be pre-built.

[0048] Among them, GNSS positioning algorithms can obtain corresponding positioning results using raw observations broadcast by satellites. However, due to the performance differences of different chips in GNSS receivers and the performance differences of antennas, the quality of the raw observations obtained by the algorithm varies greatly. The same GNSS positioning algorithm cannot adapt to raw observations of different observation qualities, or in other words, the same GNSS positioning algorithm cannot perform its best positioning performance when adapted to different chips.

[0049] The following section will use the GNSS positioning algorithm as a specific example to illustrate the optimization method for positioning algorithms.

[0050] GNSS positioning is the entire process of calculating the three-dimensional coordinates, velocity, and time of any point on the Earth's surface (or near the Earth) using satellite signals. GNSS positioning algorithms include steps such as signal acquisition, code / phase tracking, data decoding, observation extraction, positioning calculation, and accuracy assessment.

[0051] The relationship between GNSS positioning algorithms and several influencing parameters means that GNSS positioning can only be achieved by determining these influencing parameters, and adjusting these influencing parameters can effectively improve the positioning accuracy of GNSS.

[0052] In an optional embodiment, the influencing parameters may include at least one of the following: PVT weighting parameters, PVT star selection weighting parameters, RTK side floating-point deselection star weighting parameters, RTK partial ambiguity fixed correlation parameters, etc.

[0053] S2. Construct several optimization modules and determine the priority of each optimization module, the initial value and the value range of each influence parameter, wherein each optimization module contains several associated influence parameters.

[0054] Among them, the optimization module is a set of modules obtained by dividing the initial positioning algorithm; for example, the GNSS positioning algorithm can be divided into multiple optimization modules according to the steps it includes.

[0055] The priority of each optimization module can be preset based on the impact of different optimization modules on the accuracy of the positioning results of the positioning algorithm. Of course, the priority of each optimization module can also be set based on other logic.

[0056] It should be noted that the priority order of the various tuning modules is not fixed and can be adjusted based on hardware information, tuning goals, etc.

[0057] In addition, the range of values ​​for each influencing parameter may vary due to differences in parameter attributes, scenarios, and hardware. Therefore, before optimization, it is necessary to determine the initial value and range of each influencing parameter, as well as the magnitude of each adjustment.

[0058] For example, several optimization modules are set up in the GNSS positioning algorithm, and the influencing parameters that are related to the GNSS positioning algorithm are stored in each optimization module. Since there may be correlations between the various influencing parameters, to improve optimization accuracy, influencing parameters with correlations are placed in the same optimization module as much as possible. During optimization, each optimization module is optimized sequentially. To ensure orderly optimization, the priority of each optimization module is determined first, and each optimization module is optimized in order of priority. Influencing parameters that have a greater impact on GNSS positioning accuracy are placed in higher priority optimization modules.

[0059] S3. Obtain the initial positioning result obtained according to the initial positioning algorithm, and determine the optimization module with the highest priority as the current module.

[0060] Before adjusting the influencing parameters, in order to understand the changes in positioning accuracy before and after the adjustment, we can first obtain the positioning results of the initial positioning algorithm before the adjustment, that is, the initial positioning results.

[0061] S4. Traverse the influence parameters of the current module and determine the target adjustment value of the influence parameters of the current module. When the influence parameters of the current module are set to the target adjustment value, obtain the target positioning algorithm. The target positioning result obtained based on the target positioning algorithm has the greatest improvement in positioning accuracy compared with the initial positioning result.

[0062] During optimization, each optimization module contains multiple influencing parameters. This solution uses a traversal approach to iterate through the possible values ​​of each influencing parameter of the current optimization module, and adjusts each influencing parameter multiple times until the target adjustment value of each influencing parameter is obtained, thereby completing the optimization of the current module.

[0063] Each time an influencing parameter is adjusted, the positioning result obtained by the adjusted positioning algorithm is acquired, and the difference between this result and the initial positioning result is compared to determine whether the adjustment improved the positioning accuracy of the algorithm and the extent of the improvement. Finally, the value of the influencing parameter that provides the greatest improvement in positioning accuracy is taken as the target adjustment value. After setting all influencing parameters of the current module to their target adjustment values, the adjustment of the current module is complete.

[0064] Optionally, when iterating through the various influencing parameters of the current tuning module, individual influencing parameters can be adjusted sequentially, or multiple influencing parameters can be adjusted simultaneously; this application does not impose any restrictions. The adjustment range of each influencing parameter can be adjusted according to the specific parameter; this application does not impose any restrictions.

[0065] Optionally, there can be only one target adjustment value or there can be multiple target adjustment values, and this application does not impose any restrictions. When there are multiple target adjustment values, the values ​​of the top N influencing parameters whose positioning accuracy improvement is higher than a certain threshold are selected as target adjustment values ​​and used as the basis for the next optimization module, so that the top N target adjustment values ​​of positioning accuracy improvement can be obtained in each optimization.

[0066] S5, continue to use the next highest priority tuning module as the current module, and execute step 4 again until the target adjustment value of the influence parameters of all tuning modules is determined.

[0067] Based on the same tuning method, each tuning module is tuned in descending order of priority, and finally the target localization algorithm with all tuning modules adjusted is obtained, thus completing the tuning process of the entire localization algorithm.

[0068] In this embodiment, there are multiple parameters affecting the positioning accuracy of the positioning algorithm. These parameters are first compiled, and several optimization modules are constructed. Each optimization module contains at least one influencing parameter. The modules are independent of each other, while related parameters are stored within the same module whenever possible. Since each parameter has a different impact on positioning accuracy, this application pre-sets the priority order of each optimization module based on experience, and presets the initial values ​​and ranges of each influencing parameter. During optimization, each optimization module is optimized sequentially according to its priority. When optimizing each module, the process iterates through the initial values ​​and ranges of each influencing parameter, as well as the preset adjustment magnitude for each parameter, obtaining a positioning result for each adjustment. The optimization process considers the correlation between different influencing parameters and the priority order of adjusting different optimization modules. Lower-priority optimization modules are optimized based on the adjustment results of higher-priority modules, thus effectively handling the ordered optimization of multi-module, multi-parameter positioning algorithms. The improved positioning accuracy after optimization also reflects the optimized positioning algorithm, which is better adapted to the current hardware.

[0069] therefore, Figure 1 The optimization method for the positioning algorithm provided in the software can automatically optimize the parameters of the positioning algorithm for different hardware, reduce repetitive work for R&D personnel, and ensure that the positioning algorithm can perform its best positioning performance when applied to different hardware.

[0070] In an optional embodiment, the influence parameters of the current module can be traversed based on a preset traversal algorithm, which includes at least a binary search method. This improves the efficiency of traversing the influence parameters, enabling efficient automated tuning.

[0071] In an optional embodiment, based on the foregoing embodiments, step 4 further includes: obtaining the number of adjustments to the influencing parameters for each tuning module; and adjusting the influencing parameters according to the number of adjustments.

[0072] Since the impact of the parameters contained in each tuning module on positioning accuracy varies, in order to improve the tuning speed when tuning each tuning module, we can first determine the number of times the parameters need to be adjusted, and then adjust the possible values ​​of each parameter according to the number of adjustments.

[0073] In an optional embodiment, the number of adjustments to each influencing parameter is determined based on its degree of impact on positioning accuracy; for example, parameters with a higher degree of influence require more adjustments. By adjusting parameters with a higher degree of influence multiple times, the target adjustment values ​​for these parameters can be better determined; for parameters with a lower degree of influence, the number of adjustments can be reduced to improve the adjustment efficiency of the positioning algorithm.

[0074] In an optional embodiment, the optimization method of the positioning algorithm may further include: uniformly adjusting the associated influencing parameters and individually adjusting the unassociated influencing parameters.

[0075] Furthermore, since each optimization module includes several related influencing parameters, these parameters not only individually affect positioning accuracy but may also collectively have an additional impact on positioning accuracy. Therefore, this solution adjusts the possible values ​​of related influencing parameters uniformly to improve optimization accuracy. Uniform adjustment means adjusting multiple related influencing parameters simultaneously, with each related influencing parameter forming a set of possible adjustment values. However, when adjusting the possible values ​​of unrelated influencing parameters, these parameters are adjusted individually.

[0076] Furthermore, the priority order of this application is not fixed. During optimization, the optimization method of the positioning method provided in this application may also include: adjusting the priority order according to the degree of optimization of the adjustment data collected by the GNSS positioning algorithm after each adjustment.

[0077] In an optional embodiment, after optimizing the positioning method, the solution may further include: storing the target adjustment values ​​of the influence parameters contained in each optimization module into a binary file, obtaining the binary file during GNSS positioning system initialization, and performing real-time GNSS algorithm calculation based on the binary file.

[0078] In an optional embodiment, based on the foregoing embodiment, the influencing parameters include target parameters, and before step 2, the method further includes: obtaining the difference between the first antenna information and the second antenna information, and adjusting the target parameters based on the difference; wherein, the first antenna information is the initial positioning algorithm before adjustment, such as the antenna information corresponding to the initial GNSS positioning algorithm, and the second antenna information is the antenna information corresponding to the initial positioning algorithm after adjustment, and the antenna information is the observation information collected by the corresponding antenna in an open scene.

[0079] When optimizing the GNSS positioning algorithm, this solution can also be optimized through other optimization methods. For example, based on the difference between the two antennas (the average carrier noise power density ratio CN0 and the number of satellites in an open scene), CN0 can be refreshed to change the base value. In this case, it may not be necessary to adjust other optimization modules, so subsequent steps can be skipped. Alternatively, after executing this embodiment, the priority of subsequent steps and the impact parameters of the adjustments may differ from those in the aforementioned embodiment.

[0080] In an optional embodiment, based on the foregoing embodiments, the positioning accuracy at each adjustment can be determined based on the following steps: for each adjustment, obtain intermediate positioning results under multiple different application scenarios and calculate the positioning error of the intermediate positioning results; obtain multiple statistical indicators of the positioning error; determine a score for the intermediate positioning results based on the multiple statistical indicators, the score being used to represent the positioning accuracy of the intermediate positioning results.

[0081] For vehicle-mounted GNSS positioning algorithms, application scenarios can include open spaces, driving on general elevated roads, driving under difficult elevated roads, driving on low-rise buildings, driving under trees, and so on.

[0082] It should be noted that when applied to different devices (such as vehicles, aircraft, etc.) and different positioning algorithms, multiple application scenarios can be defined, that is, application scenarios include but are not limited to the examples above.

[0083] In an optional embodiment, the statistical metric may include at least one of RMSE (root mean square error), CEP68 (circular error probability of 68%), CEP95 (circular error probability of 95%), CEP99.7 (circular error probability of 99.7%), and MAX (maximum error).

[0084] It should be noted that statistical indicators include, but are not limited to, the examples above, and may also include other indicators for error assessment; this application does not impose any restrictions.

[0085] In this embodiment, for each adjustment, the positioning result can be scored based on multiple statistical indicators of positioning results from multiple application scenarios (referred to as "intermediate positioning results") to represent the positioning accuracy of the current positioning result. By comparing the differences between the scores of positioning results obtained from different adjustments, the improvement in positioning accuracy between different adjustments can be determined.

[0086] In a specific example, data from the same application scenario within a certain time period are grouped together. For each epoch of data, the GNSS algorithm calculates a positioning result. This result differs from the true value by a certain value called the accuracy error (diff) for that epoch (taking horizontal accuracy error as an example). Since each epoch has a diff, after a period of time, a set of diff = [diff1, diff2, diff3, ...] will be obtained. Each diff set can be evaluated using multiple statistical indicators. Specifically, the evaluation standard can be based on the weighted scores of multiple statistical indicators. These statistical indicators can include the following indicators in the horizontal and / or vertical directions: RMSE, CEP68, CPE95, CEP99.7, MAX, etc. Multiple sets of data can be used to evaluate the merits of a positioning algorithm, demonstrating its universality.

[0087] In one embodiment, refer to the appendix to the specification. Figure 2 This application also provides a device for optimizing a positioning algorithm, including an acquisition module 10, a construction module 20, a preprocessing module 30, a first processing module 40, and a second processing module 50. The acquisition module 10 is used to acquire an initial positioning algorithm and acquire several influencing parameters that are related to the initial positioning algorithm; the construction module 20 is used to construct several tuning modules and determine the priority of each tuning module, the initial value of each influencing parameter, and the value range, wherein each tuning module contains several related influencing parameters; the preprocessing module 30 is used to acquire the initial positioning result obtained according to the initial positioning algorithm and determine the tuning module with the highest priority as the current module; the first processing module 40 is used to traverse the influencing parameters of the current module and determine the target adjustment value of the influencing parameters of the current module, wherein when the influencing parameter of the current module is set to the target adjustment value, the target positioning algorithm is acquired, wherein the target positioning result obtained based on the target positioning algorithm has the greatest improvement in positioning accuracy compared to the initial positioning result; the second processing module 50 is used to continue to use the tuning module with the next highest priority as the current module and execute the processing operation of the first processing module 40 until the target adjustment value of the influencing parameters of all tuning modules is determined.

[0088] The technical concept of the positioning algorithm optimization device provided in this embodiment is the same as that of the positioning algorithm optimization method provided in the foregoing embodiments, and will not be described again in this embodiment. In addition, the acquisition module 10, construction module 20, preprocessing module 30, first processing module 40 and second processing module 50 in this application can be individual virtual modules, or they can be freely combined into a new module. For example, the first processing module 40 and the second processing module 50 can be combined into a processing module. This application does not impose any restrictions.

[0089] In one embodiment, based on the foregoing embodiments, this application also provides a computer device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the tuning method of the positioning method of the foregoing embodiments.

[0090] Specifically, the computer device can be used to optimize positioning chips, such as GNSS chips, and the device stores the obtained optimal positioning algorithm into the corresponding GNSS chip.

[0091] The optimization method of the positioning method of this application can be implemented using computer-executable program code. Therefore, these code snippets can be stored in a storage device for execution by a computing device, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, this invention is not limited to any particular hardware and software combination.

[0092] It should be noted that the above embodiments can be freely combined as needed. The above description is only a preferred embodiment of the present invention. It should be pointed out that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method of tuning a positioning algorithm, characterized by, The method comprises the following steps: Step 1, obtaining an initial positioning algorithm and a plurality of influence parameters associated with the initial positioning algorithm; Step 2, constructing a plurality of tuning modules, and determining the priority of each tuning module, the initial value and the value range of each influence parameter, wherein each tuning module comprises a plurality of associated influence parameters; Step 3, obtaining an initial positioning result obtained according to the initial positioning algorithm, and determining the tuning module with the highest priority as the current module; Step 4, traversing the influence parameters of the current module, and determining the target adjustment value of the influence parameters of the current module, wherein when the influence parameters of the current module are set to the target adjustment value, a target positioning algorithm is obtained, and the positioning accuracy of the target positioning result obtained based on the target positioning algorithm is improved the most compared with the initial positioning result; Step 5, continuing to take the tuning module with the next priority as the current module, and executing the step 4 again until the target adjustment value of the influence parameters of all the tuning modules is determined; Wherein, the priority of each tuning module is set in advance based on the influence of different tuning modules on the accuracy of the positioning result of the positioning algorithm.

2. The method of claim 1, wherein, The step 4 further comprises: For each tuning module, the number of adjustments of the influence parameters is obtained; According to the number of adjustments, the influence parameters are adjusted; Wherein, the number of adjustments of different influence parameters is different.

3. The method of claim 1, wherein, The method further comprises: The associated influence parameters are adjusted uniformly, and the unassociated influence parameters are adjusted individually.

4. The method of claim 1 to 3, wherein The influence parameters include target parameters, and the step 2 further comprises: Obtaining the difference between the first antenna information and the second antenna information, and adjusting the target parameters based on the difference; Wherein, the first antenna information is the antenna information corresponding to the initial positioning algorithm before adjustment, the second antenna information is the antenna information corresponding to the initial positioning algorithm after adjustment, and the antenna information is the observation information collected by the corresponding antenna in the empty field scenario.

5. The method of claim 1 to 3, wherein The positioning accuracy at each adjustment is determined based on the following steps: For each adjustment, a plurality of intermediate positioning results in different application scenarios are obtained, and the positioning error of the intermediate positioning result is calculated; A plurality of statistical indicators of the positioning error are obtained; Based on the plurality of statistical indicators, the score of the intermediate positioning result is determined, and the score is used to represent the positioning accuracy of the intermediate positioning result.

6. The method of claim 5, wherein, The statistical indicators include at least one of the root mean square error, CEP68, CEP95, CEP99.7 and maximum error, the CEP68 represents the circular error probability of 68%, the CEP95 represents the circular error probability of 95%, and the CEP99.7 represents the circular error probability of 99.7%.

7. The method of claim 1 to 3, wherein The traversal of the influence parameters of the current module comprises: traversing the influence parameters of the current module based on a preset traversal algorithm, and the preset traversal algorithm at least comprises dichotomy.

8. The method of claim 1 to 3, wherein, The method further comprises that the order of the priority is adjustable.

9. An apparatus for tuning a positioning algorithm, characterized in that The method comprises the following steps: An acquisition module is configured to acquire an initial positioning algorithm and a plurality of influence parameters associated with the initial positioning algorithm; A construction module is configured to construct a plurality of tuning modules, determine a priority of each of the tuning modules, an initial value and a value range of each of the influence parameters, wherein each of the tuning modules comprises a plurality of associated influence parameters; A preprocessing module is configured to acquire an initial positioning result obtained according to the initial positioning algorithm, and determine the tuning module with the highest priority as a current module; A first processing module is configured to traverse the influence parameters of the current module, determine a target adjustment value of the influence parameters of the current module, and acquire a target positioning algorithm when the influence parameters of the current module are set to the target adjustment value, wherein a positioning result obtained based on the target positioning algorithm has the largest improvement in positioning accuracy compared with the initial positioning result; A second processing module is configured to continue to determine the tuning module with the next priority as the current module, and perform the processing operation of the first processing module until the target adjustment value of the influence parameters of all the tuning modules is determined. The priority of each of the tuning modules is preset based on the influence of different tuning modules on the accuracy of the positioning result of the positioning algorithm.

10. A computer device comprising a memory, a processor, and a computer program stored on the memory, wherein the computer program comprises instructions that, when executed by the processor, cause the processor to perform the method of any one of claims 1-9. The processor executes the computer program to implement the steps of the tuning method of the positioning algorithm of any one of claims 1-8.

Images