An RTK high-precision positioning method based on an improved adaptive genetic algorithm star selection strategy

Abstract

This invention discloses a high-precision RTK positioning method based on an improved adaptive genetic algorithm-based satellite selection strategy, belonging to the field of satellite navigation and positioning technology. A subset of satellites with a specified number of satellites is selected to minimize the loss of the Position Accuracy Factor (PDOP) and satellite geometric volume. The satellite selection problem is transformed into a combinatorial optimization problem. A fitness function is designed with the reciprocal of PDOP and geometric volume as its core, and adaptive crossover and mutation probabilities are introduced. The genetic operation parameters are dynamically adjusted according to the population evolutionary state to balance global search and local convergence capabilities. Iterative evolution is performed under the condition of satisfying subset size constraints. When the convergence condition is met, the subset of satellites with locally optimal PDOP values ​​is output for subsequent RTK ambiguity resolution and positioning. This invention can quickly converge to a near-globally optimal satellite geometry configuration in polynomial time while also considering satellite quality, significantly reducing computational complexity and improving the real-time performance and reliability of RTK positioning.

Description

Technical Field

[0001] This invention belongs to the field of satellite navigation and positioning technology, specifically relating to a high-precision RTK positioning method based on an improved adaptive genetic algorithm-based satellite selection strategy. Background Technology

[0002] When multi-constellation GNSS receivers operate in complex environments, they simultaneously receive signals from multiple navigation systems such as GPS, GLONASS, Galileo, and BeiDou, resulting in a significant increase in the number of visible satellites. In real-time dynamic differential techniques, if a traditional full-constellation strategy is used to jointly solve for all visible satellites, the computational complexity increases non-linearly with the number of satellites, leading to a series of problems such as high real-time processing pressure, increased power consumption, and increased positioning delay. Regarding the satellite subset selection problem, traditional techniques mainly include the following methods: Method 1: Elevation Angle Method. This method typically sorts all visible satellites by elevation angle from largest to smallest, and then prioritizes selecting a few satellites with the highest elevation angles for positioning calculation. The advantage of this method is its minimal computational cost and simplicity of implementation. Method 2: Signal-to-Noise Ratio (SNR) Truncation Method. This method sorts satellite signals in descending order based on their SNR and selects the top N satellites. This method is computationally simple and easy to implement. Method 3: Methods based on traversal or partial traversal, such as recursive search and branch-and-bound methods, can guarantee finding a globally optimal or suboptimal solution.

[0003] Method 1 ignores easily overlooked spatial geometric distributions, potentially increasing the Position Precision Factor (PDOP) and affecting positioning accuracy. Method 2 also ignores geometric distributions, easily leading to increased PDOP, and is highly susceptible to environmental fluctuations and unstable satellite selection. The computational cost of Method 3 increases exponentially with M and N, making it only suitable for cases where M is small and unsuitable for real-time systems. Current methods for RTK real-time positioning have many limitations. Summary of the Invention

[0004] This invention provides a high-precision RTK positioning method based on an improved adaptive genetic algorithm-based satellite selection strategy. It addresses the problem of large computational load for full constellation solution using real-time dynamic differential technology when dealing with multiple constellation GNSS receivers in complex environments with a large number of visible satellites from multiple navigation systems such as GPS, GLONASS, Galileo, and BeiDou.

[0005] This invention is achieved through the following technical solution: A method for RTK localization based on an improved adaptive genetic algorithm star selection strategy, the method comprising the following steps: Step 1: In the current epoch, acquire the common-view satellites and preprocess them; Step 2: Based on the common-view satellites from Step 1, set the size of the target satellite subset to be... The number of satellites can be dynamically adjusted according to the receiver's processing capabilities and positioning accuracy requirements until the target number of satellites is reached. Step 3: Based on the number of target satellites in Step 2, use PDOP and satellite geometry to evaluate the merits of satellite subsets, making fitness inversely proportional to PDOP and directly proportional to geometry. Step 4: Based on the objectives of the satellite subsets in Step 3, simulate biological gene recombination by exchanging some genes of the parent individuals to generate new offspring in order to explore new solution spaces; Step 5: Introduce new information into the new solution space from Step 4 by randomly changing individual genes to prevent the population from getting trapped in local optima; Step 6: Implement an elite retention strategy on the new offspring obtained in Step 4 to prevent the loss of the optimal solution during the evolution process; Step 7: Iterate through steps 3 to 6 until the convergence condition is met; Step 8: Select the optimal satellite subset from Step 7. The corresponding raw observation data is input into the RTK positioning engine to realize the RTK positioning strategy.

[0006] Furthermore, step 1 specifically involves assuming the original set of visible satellites is... The quantity is Preprocessing involves removing elements with an elevation angle lower than a preset cutoff angle. Satellites with a carrier-to-noise ratio below a preset threshold (C / ) are excluded. The satellites were used to obtain the effective satellite set:

[0007] in This refers to the number of valid satellites.

[0008] Furthermore, step 2 specifically includes initializing the genetic algorithm parameters, i.e., the population size. ; Satellite selection uses a satellite numbering mechanism, the first... The first generation of the population individual Defined as a Dimensional vector:

[0009] in Satellite number; When initializing the population, it is randomly generated. Individuals constitute the initial population:

[0010] Individuals are identified by a reference satellite. Set the initial angle window:

[0011] in In the positive direction The search begins with an initial tolerance of ±5°, randomly selecting a satellite within that range. If no usable satellite is found within this tolerance range, the tolerance is increased by a factor of 2 each time. The search then proceeds from the angle of the currently selected satellite. The search begins at a certain degree and continues until the reference satellite azimuth range is reached. After each successful screening, the window radius is halved, and the next round of screening begins, and so on until the target number of satellites is reached.

[0012] Furthermore, step 3 specifically involves, for an individual Define its corresponding satellite subset index set as Then the observation matrix It is composed of the direction cosines of each satellite in this subset; let the satellites be... The elevation angle in the station-centered coordinate system is azimuth angle is Then its direction cosine vector is : Observation matrix Transpose and original Multiply the results and then take the inverse of the result to obtain the matrix. And then The square root of the trace of the matrix is ​​the PDOP value; fitness function Defined as the reciprocal of the PDOP value and the normalized geometric volume, and introducing a fitness function to adjust the parameters. Preventing division by zero:

[0013] Among them, a higher fitness value indicates a smaller PDOP (Potentially Targeted Opportunity) for the corresponding satellite subset and a better geometric configuration; utilizing A tournament selection process is conducted until the mating pool reaches the required size.

[0014] Furthermore, step 4 specifically involves employing multi-point crossover and combining it with adaptive crossover probability to dynamically adjust the crossover strength based on the population's evolutionary state; randomly pairing individuals in the mating pool, assuming the two parent individuals in a pair are... and With adaptive crossover probability Decide whether to perform the crossover operation if the random number is greater than 1. Then, the parent generation is directly copied to the child generation; when performing crossover, a multi-point crossover operator is used; first, a random mask vector of the same length as the chromosome is generated. Each bit of the mask is independently and randomly generated as either 0 or 1; then the two child generations... and It is generated by the following formula:

[0015] in, This represents element-wise multiplication, where 1 represents a vector of all 1s; that is, the positions where the mask is 1, and the child vectors. Inheritance from the father genes, offspring Inheritance from the father The genes; positions with a mask of 0 are swapped.

[0016] Furthermore, step 5 specifically involves employing a satellite quality score variation strategy; simultaneously, the variation probability utilizes an adaptive mechanism; for individual... With adaptive mutation probability Decide whether to perform mutation; when performing mutation, the satellite with the worst satellite quality score in the current individual will be replaced with the satellite with the best satellite quality score that is not present.

[0017] Furthermore, step 6 specifically involves the satellite subset being in the... After generational evolution is complete, the population with the highest fitness will be selected. Each individual replaces the least fit individual in the next generation of the population. Individuals, among whom For population size 5% to 10% and at least 1.

[0018] Furthermore, step 7 specifically involves setting the convergence condition to any of the following conditions: the current satellite subset's evolutionary generation. , The maximum number of generations; the fitness of the best individual in the population is continuous. The iteration remains unchanged; once the convergence condition is met, the iteration stops, and the individual with the highest fitness from the current population is selected as the optimal solution. ; Obtain the optimal satellite subset That is, PDOP with local optimum One satellite.

[0019] Furthermore, step 8, RTK positioning, specifically involves constructing a double-difference observation equation to eliminate receiver and satellite clock bias; using the LAMBDA method or its improved algorithm to perform integer ambiguity resolution; if the ambiguity is successfully fixed, outputting the fixed solution coordinates; if the fixing fails, outputting the floating-point solution coordinates, and the option to recalculate using a spare subset or use the full constellation solution as a backup.

[0020] A method for RTK positioning based on an improved adaptive genetic algorithm-based star selection strategy, as described above, is applied to high-precision navigation and positioning in complex environments.

[0021] The beneficial effects of this invention are: This invention provides a method for rapidly and robustly selecting satellite subsets with locally optimal PDOP values ​​in complex environments, ensuring both excellent geometric configuration and real-time performance while also considering observation quality.

[0022] This invention can quickly converge to a near-global optimal satellite geometry configuration in polynomial time while taking into account satellite quality, significantly reducing computational complexity and improving the real-time performance and reliability of RTK positioning.

[0023] This invention fully utilizes the global search and local convergence balancing capabilities of an improved adaptive genetic algorithm. Based on combinatorial optimization and satellite geometry optimization objectives, it establishes a fitness function with the inverse of PDOP and geometric volume as its core, and an adaptive genetic operation parameter model that best matches the algorithm, starting from the perspective of satellite subset selection. This forms a real-time dynamic differential positioning method based on the improved adaptive genetic algorithm's satellite selection strategy. It overcomes the limitation of large computational load in RTK full constellation solution in complex environments of multi-constellation GNSS, shortens the RTK positioning solution time, significantly reduces computational complexity, improves operational efficiency in application fields such as UAV swarms, unmanned vehicle collaboration, and surface / underwater unmanned systems, enhances the real-time performance and reliability of RTK positioning, reduces equipment computing power consumption and operation and management costs, and constructs an efficient high-precision RTK positioning service for complex environments. This is of great significance for promoting the development of my country's high-precision satellite navigation application industry. Attached Figure Description

[0024] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0025] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods are omitted so as not to obscure the description of this application with unnecessary detail.

[0026] It should be understood that, when used in this specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0027] It should also be understood that the terminology used in this application specification is for the purpose of describing particular embodiments only and is not intended to limit the application. As used in this application specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0028] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0029] Many specific details are set forth in the following description in order to provide a full understanding of this application. However, this application may also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the spirit of this application. Therefore, this application is not limited to the specific embodiments disclosed below.

[0030] Implementation Method 1 A subset of satellites with a specified number of satellites is selected to minimize the Position Dilution of Precision (PDOP) and satellite geometric volume loss. First, an adaptive window recursive selection method is used to transform the satellite selection problem into a combinatorial optimization problem. Then, a fitness function is designed with the reciprocal of PDOP and geometric volume as its core, and adaptive crossover and mutation probabilities are introduced. Genetic operation parameters are dynamically adjusted according to the population evolutionary state to balance global search and local convergence capabilities. Tournament selection, multi-point crossover, and satellite score mutation are then employed for iterative evolution under the condition of satisfying subset size constraints. When the convergence condition is met, the subset of satellites with locally optimal PDOP values ​​is output for subsequent RTK ambiguity resolution and positioning.

[0031] This embodiment provides a method for RTK localization based on an improved adaptive genetic algorithm star selection strategy, such as... Figure 1 As shown, the method includes the following steps: Step 1: In the current epoch, acquire the common-view satellites and preprocess them; Step 2: Based on the common-view satellites from Step 1, set the size of the target satellite subset to be... The number of satellites can be dynamically adjusted according to the receiver's processing capabilities and positioning accuracy requirements until the target number of satellites is reached. Step 2's initialization approach balances the needs of both genetic algorithms and RTK localization. It achieves coordinated optimization of signal quality and geometric distribution, providing a high-quality, highly diverse initial population for RTK localization and accelerating convergence to the optimal solution. Step 3: Based on the number of target satellites in Step 2, use PDOP and satellite geometry to evaluate the merits of satellite subsets, making fitness inversely proportional to PDOP and directly proportional to geometry. Step 4: Based on the objectives of the satellite subsets in Step 3, simulate biological gene recombination by exchanging some genes of the parent individuals to generate new offspring in order to explore new solution spaces; Step 5: Introduce new information into the new solution space from Step 4 by randomly changing individual genes to prevent the population from getting trapped in local optima; Step 5, satellite quality-oriented mutation, embeds domain knowledge into the evolutionary process, ensuring the search direction remains focused on high signal-to-noise ratio and high elevation angle regions. The mutation operation no longer involves random walks but instead performs biased perturbations along the quality gradient. This maintains population diversity while enabling refined mining of the high-quality solution space, improving convergence efficiency and solution quality. Step 6: Implement an elite retention strategy on the new offspring obtained in Step 4 to prevent the loss of the optimal solution during the evolution process; Step 7: Iterate through steps 3 to 6 until the convergence condition is met; Step 8: Select the optimal satellite subset from Step 7. The corresponding raw observation data (pseudorange, carrier phase, ephemeris) are input into the RTK positioning engine to realize the positioning of the RTK satellite selection strategy.

[0032] Further, step 1 specifically involves, at the current epoch, the receiver acquiring and tracking all visible satellites, obtaining the broadcast ephemeris, satellite number (PRN), elevation angle (E), azimuth angle (A), carrier-to-noise ratio (C / N0), and pseudorange / carrier phase observations for each satellite. Let the original set of visible satellites be... The quantity is To remove low-quality satellites, the following preprocessing steps are performed: satellites with elevation angles lower than a preset cutoff angle are removed. Satellites with a carrier-to-noise ratio below a preset threshold (C / ) are excluded. The satellites were used to obtain the effective satellite set:

[0033] in, This refers to the number of valid satellites.

[0034] Furthermore, step 2 specifically includes setting the target satellite subset size. The value is dynamically adjusted between 8 and 16 based on the receiver's processing power or positioning accuracy requirements. Initialize the genetic algorithm parameters: population size is set to... Maximum number of generations Initial crossover probability Initial value of mutation probability =0.05; Fitness function adjustment parameter ,and , Adaptive adjustment coefficient This is used to adjust the degree of adaptability of crossover and mutation probabilities; convergence criteria parameters. fitness change threshold ; Satellite selection uses a satellite numbering mechanism, such as the satellite numbering system. The first generation of the population individual Defined as a Dimensional vector:

[0035] in, Satellite number; When initializing the population, it is randomly generated. Individuals constitute the initial population:

[0036] Individuals are identified by a reference satellite. Set the initial angle window:

[0037] Where D = 60°, In the positive direction The search is performed with an initial tolerance of ±5°, i.e., in the positive direction. The search begins with an initial tolerance of ±5°, and any satellite within that tolerance is selected and placed into the current instance. If no usable satellite is found within that tolerance, the tolerance is increased by a factor of 2 each time until the tolerance reaches its maximum value. Until the target number of satellites is reached; after each successful round of screening, the window radius is halved, and the next round of screening begins.

[0038] The selection criteria for reference satellites are as follows:

[0039] in, For single-star comprehensive quality score, For carrier-to-noise ratio scoring, Scoring for elevation angle For multipath evaluation, the smaller the error, the better. The stability score reflects the continuity and reliability of satellite observations. Multipath effect value:

[0040] in, The current elevation angle of the satellite. =25 is the reference elevation angle. As a reference carrier-to-noise ratio and a constraint on the selection of reference satellites, the reference satellites are chosen from a set of satellites. The one with the highest score that meets the restrictions.

[0041] Furthermore, step 3 specifically involves, for an individual Define its corresponding satellite subset index set as Then the observation matrix It is composed of the direction cosines of each satellite in this subset; let the satellites be... The elevation angle in the station-centered coordinate system is azimuth angle is Then its direction cosine vector is ,Right now:

[0042] in, To obtain the trace of a matrix, which is the value of the main diagonal of the matrix; fitness function Defined as the reciprocal of the PDOP value and the normalized geometric volume, and introducing a fitness function to adjust the parameters. Preventing division by zero:

[0043] Among them, the larger the fitness value, the smaller the PDOP of the satellite subset corresponding to that individual, and the better the geometric configuration; this invention utilizes Tournament selection is performed, which involves randomly selecting from the population each time. Individuals, this time Select the individuals with the highest fitness and place them in the mating pool. Repeat this process until the mating pool reaches the required size. .

[0044] Furthermore, step 4 specifically involves the following: This invention employs multi-point crossover, combined with adaptive crossover probability, to dynamically adjust the crossover strength based on the population's evolutionary state; individuals in the mating pool are randomly paired, assuming the two parent individuals in the pairing are... and With adaptive crossover probability Decide whether to perform the crossover operation if the random number is greater than 1. Then directly copy from parent to child, this time using a random number. Crossover is performed based on probability; if crossover is required, the parent generation is directly copied to the child generation; adaptive crossover probability. The calculation formula is:

[0045] in, The first The maximum, minimum, and average fitness of the population. For a very small positive number =10 -6 This is used to prevent the denominator from being zero. The degree of influence of probability; when performing crossover, a multi-point crossover operator is used; first, a random mask vector of the same length as the chromosome is generated. Each bit of the mask is independently and randomly generated as either 0 or 1 (with a 50% probability of 0 or 1); then the two child generations... and It is generated by the following formula:

[0046] in, This represents element-wise multiplication, where 1 represents a vector of all 1s; that is, the positions where the mask is 1, and the child vectors. Inheritance from the father genes, offspring Inheritance from the father The genes; positions with a mask of 0 are swapped.

[0047] Furthermore, step 5 specifically involves the present invention employing a satellite quality score mutation strategy. This strategy naturally maintains subset size constraints without requiring additional repair and effectively preserves diversity. Simultaneously, the mutation probability uses an adaptive mechanism, giving individuals with lower fitness a higher probability of mutation, thereby improving inferior solutions. For individuals... With adaptive mutation probability Decide whether to perform mutation; adaptive mutation probability. The calculation formula is:

[0048] Specifically, during mutation, the individual with the worst satellite quality score is replaced with the individual with the best satellite quality score that is not present in the dataset. Other symbols have the same meaning as before. If mutation is performed, extract from the satellite set. A randomly selected satellite from the top ten satellites and the current individual The satellite with the worst rating will be replaced.

[0049] Furthermore, step 6 specifically involves the satellite subset being in the... After generational evolution is complete, the current population size will be... The most adaptable Individuals replace the next generation's population size. The worst adaptability Individuals, among whom For population size 5% to 10% and at least 1; that is, in the first After generational evolution is complete, the generation with the highest fitness in the current population will be selected. Individuals are directly preserved to the next generation. Use this An elite individual replaces the least fittest member of the next generation. Each individual element ensures the monotonic convergence of the algorithm.

[0050] Furthermore, step 7 specifically involves setting the convergence condition to any of the following conditions: the current satellite subset's evolutionary generation. , The maximum number of generations; the fitness of the best individual in the population is continuous. No change has occurred over time, that is ,in The preset minimum threshold, i.e. Once the convergence condition is met, stop iterating and select the individual with the highest fitness from the current population as the optimal solution. ; Obtain the optimal satellite subset That is, PDOP with local optimum One satellite.

[0051] Furthermore, step 8, RTK positioning, specifically involves constructing a double-difference observation equation to eliminate receiver and satellite clock bias; using the LAMBDA method or its improved algorithm to perform integer ambiguity resolution; if the ambiguity is successfully fixed, outputting the fixed solution coordinates; if the fixing fails, outputting the floating-point solution coordinates, and the option to recalculate using a spare subset or use the full constellation solution as a backup.

[0052] Implementation Method 2 This embodiment provides a system for RTK positioning based on an improved adaptive genetic algorithm star selection strategy. The system is described in Embodiment 1 and includes: Acquisition and Preprocessing Module: In the current epoch, acquire common-view satellites and preprocess them; Processing and Adjustment Module: Based on the common-view satellites acquired and preprocessed by the module, the size of the target satellite subset is set to [value missing]. The number of satellites can be dynamically adjusted according to the receiver's processing capabilities and positioning accuracy requirements until the target number of satellites is reached. This initialization approach balances the needs of both genetic algorithms and RTK localization. It achieves coordinated optimization of signal quality and geometric distribution, providing a high-quality, highly diverse initial population for RTK localization and accelerating convergence to the optimal solution. The satellite subset evaluation module evaluates the quality of satellite subsets based on the number of target satellites in the processing and adjustment module. It uses PDOP and satellite geometry to evaluate the quality of satellite subsets, making fitness inversely proportional to PDOP and directly proportional to geometry. A new solution space exploration module: The evaluation of the merits and demerits of satellite subsets is based on the evaluation of the merits and demerits of satellite subsets. It simulates biological gene recombination and generates new offspring by exchanging some genes of parent individuals in order to explore new solution spaces. Optimal solution calculation module: The new solution space of the new solution space exploration module introduces new information by randomly changing individual genes to prevent the population from getting trapped in local optima; Satellite quality-oriented mutation embeds domain knowledge into the evolutionary process, ensuring that the search direction always focuses on regions with high signal-to-noise ratio and high elevation angle. The mutation operation no longer involves random walks but instead performs biased perturbations along the quality gradient. This maintains population diversity while enabling fine-grained mining of the high-quality solution space, improving convergence efficiency and solution quality. An elite preservation strategy is adopted to prevent the loss of the optimal solution during the evolution process; Iterative execution module: Iteratively executes the target evaluation module for the superiority and inferiority of satellite subsets, the new solution space exploration module, and the optimal solution calculation module until the convergence condition is met; The RTK positioning module: selects the optimal subset of satellites from the iterative running module. The corresponding raw observation data (pseudorange, carrier phase, ephemeris) are input into the RTK positioning engine to realize the positioning of the RTK satellite selection strategy.

[0053] Implementation Method 3 This embodiment provides a method for RTK positioning based on an improved adaptive genetic algorithm-based star selection strategy, as described in Embodiment 1, which is applied to high-precision navigation and positioning in complex environments such as urban canyons and mountainous areas.

Claims

1. A method for RTK localization based on an improved adaptive genetic algorithm star selection strategy, characterized in that, The method includes the following steps: Step 1: In the current epoch, acquire the common-view satellites and preprocess them; Step 2: Based on the common-view satellites from Step 1, set the size of the target satellite subset to be... The number of satellites can be dynamically adjusted according to the receiver's processing capabilities and positioning accuracy requirements until the target number of satellites is reached. Step 3: Based on the number of target satellites in Step 2, use PDOP and satellite geometry to evaluate the merits of satellite subsets, making fitness inversely proportional to PDOP and directly proportional to geometry. Step 3 specifically involves, for an individual Define its corresponding satellite subset index set as Then the observation matrix It is composed of the direction cosines of each satellite in this subset; let the satellites be... The elevation angle in the station-centered coordinate system is azimuth angle is Then its direction cosine vector is : Observation matrix Transpose and original Multiply the results and then take the inverse of the result to obtain the matrix. And then The square root of the trace of the matrix is ​​the PDOP value; fitness function Defined as the reciprocal of the PDOP value and the normalized geometric volume, and introducing a fitness function to adjust the parameters. Preventing division by zero: Among them, a higher fitness value indicates a smaller PDOP (Potentially Targeted Opportunity) of the satellite subset corresponding to that individual, and a better geometric configuration; utilizing The tournament selection process continues until the mating pool reaches the required size. Step 4: Based on the objectives of the satellite subsets in Step 3, simulate biological gene recombination by exchanging some genes of the parent individuals to generate new offspring in order to explore new solution spaces; Step 5: Introduce new information into the new solution space from Step 4 by randomly changing individual genes to prevent the population from getting trapped in local optima; Step 6: Implement an elite retention strategy on the new offspring obtained in Step 4 to prevent the loss of the optimal solution during the evolution process; Step 7: Iterate through steps 3 to 6 until the convergence condition is met; Step 8: Select the optimal satellite subset from Step 7. The corresponding raw observation data is input into the RTK positioning engine to realize the RTK positioning strategy.

2. The method according to claim 1, characterized in that, Step 1 specifically involves assuming the original set of visible satellites is... The quantity is Preprocessing involves removing elements with an elevation angle lower than a preset cutoff angle. Satellites with a carrier-to-noise ratio below a preset threshold (C / ) are excluded. The satellites were used to obtain the effective satellite set: in This refers to the number of valid satellites.

3. The method according to claim 1, characterized in that, Step 2 specifically includes initializing the genetic algorithm parameters, i.e., the population size. Satellite selection adopts a satellite numbering mechanism, the first... The first generation of the population individual Defined as a Dimensional vector: in Satellite number; When initializing the population, it is randomly generated. Individuals constitute the initial population: Individuals are identified by a reference satellite; Set the initial angle window: in In the positive direction The search begins with an initial tolerance of ±5°, and a satellite is randomly selected within that range. If no usable satellite is found within that tolerance range, the tolerance is increased by a factor of 2 each time. And from the perspective of the currently selected satellite The search begins at a certain degree and continues until the reference satellite azimuth range is reached. After each successful screening, the window radius is halved, and the next round of screening begins, and so on until the target number of satellites is reached.

4. The method according to claim 1, characterized in that, Step 4 specifically involves employing multi-point crossover and combining it with adaptive crossover probability to dynamically adjust the crossover strength based on the population's evolutionary state; randomly pairing individuals in the mating pool, assuming the two parent individuals in a pair are... and With adaptive crossover probability Decide whether to perform the crossover operation if the random number is greater than 1. Then, the parent generation is directly copied to the child generation; when performing crossover, a multi-point crossover operator is used; first, a random mask vector of the same length as the chromosome is generated. Each bit of the mask is independently and randomly generated as either 0 or 1; then the two child generations... and It is generated by the following formula: in, This represents element-wise multiplication, where 1 represents a vector of all 1s; that is, the positions where the mask is 1, and the child vectors. Inheritance from the father genes, offspring Inheritance from the father The genes; positions with a mask of 0 are swapped.

5. The method according to claim 1, characterized in that, Step 5 specifically involves employing a satellite quality score variation strategy; simultaneously, an adaptive mechanism is used for the variation probability; for individual... With adaptive mutation probability Decide whether to perform mutation; when performing mutation, the satellite with the worst satellite quality score in the current individual will be replaced with the satellite with the best satellite quality score that is not present.

6. The method according to claim 1, characterized in that, Specifically, step 6 involves the satellite subset being in the... After generational evolution is complete, the population with the highest fitness will be selected. Each individual replaces the least fit individual in the next generation of the population. Individuals, among whom For population size 5% to 10% and at least 1.

7. The method according to claim 6, characterized in that, Step 7 specifically involves setting the convergence condition to any of the following conditions: the number of generations of the current satellite subset. , The maximum number of generations; the fitness of the best individual in the population is continuous. The iteration remains unchanged; once the convergence condition is met, the iteration stops, and the individual with the highest fitness from the current population is selected as the optimal solution. ; Obtain the optimal satellite subset That is, PDOP with local optimum One satellite.

8. The method according to claim 1, characterized in that, Step 8, RTK positioning, specifically involves constructing a double-difference observation equation to eliminate receiver and satellite clock errors; using the LAMBDA method or its improved algorithm to resolve integer ambiguities; if the ambiguity is successfully fixed, outputting the fixed solution coordinates; if the fixing fails, outputting the floating-point solution coordinates, and the option to recalculate using a spare subset or use the full constellation solution as a backup.

9. A method for RTK positioning based on an improved adaptive genetic algorithm-based star selection strategy as described in any one of claims 1-8, which is applied to high-precision navigation and positioning in complex environments.

Images