A system for providing artificial intelligence-assisted positioning services for vehicles

Abstract

The present invention relates to a system (1) for developing an intelligent positioning module by combining global navigation satellite systems technology with next generation cellular-based positioning technologies.

Description

[0001] DESCRIPTION

[0002] A SYSTEM FOR PROVIDING ARTIFICIAL INTELLIGENCE-ASSISTED

[0003] POSITIONING SERVICES FOR VEHICLES

[0004] Technical Field

[0005] The present invention relates to a system for developing an intelligent positioning module by combining global navigation satellite systems technology with next generation cellular-based positioning technologies.

[0006] Background of the Invention

[0007] Today, the standard techniques commonly used for outdoor positioning are built on technologies based on global navigation satellite systems (GNSS) (for example, global positioning system (GPS), and real-time kinematics (RTK)). However, the high latency problems and scalability limitations due to the unavailability of GNSS service in confined spaces (for example, tunnels and parking garages), create the need for a more comprehensive approach that provides low latency, high capacity and accuracy.

[0008] For this reason, considering the studies and deficiencies included in the current technique, it is understood that there is a need for a system for creating a service architecture that will run the mobile network service with maximum efficiency by taking into account the expectations of meeting the network density and vehicle positioning requests for mobile vehicles; providing uninterrupted positioning services needed by vehicles in different situations (urban, intercity, etc.) by means of artificial intelligence and multi-source information flow; and enabling operators and service providers to have the capacity to improve service continuity for smart and connected vehicles to a maximum level by virtue of this information. The Chinese patent document no. CN113596727, an application included in the state of the art, discloses mobile phone positioning and navigation system and method applied to mine. The said invention relates to a mobile phone positioning and navigation system and method applied to a mine and belongs to the technical field of mine positioning. According to the method, the UWB / ZigBee accurate positioning technology is adopted, a high-integration-level positioning module is designed, this module is integrated into a mining mobile phone, and two-way communication with the mining mobile phone is achieved through low-power- consumption Bluetooth. According to the method, a UWB / ZigBee accurate positioning system needs to be installed in an underground coal mine, and whole mine roadway coverage of wireless positioning signals is achieved. The positioning module is integrated in a shell of the mining mobile phone, obtains position coordinate information based on a UWB / ZigBee wireless positioning technology, and sends this information to the mining mobile phone through Bluetooth. The mining mobile phone downloads and updates a mine GIS roadway map through WIFI or a 4G / 5G wireless network. The positioning function of the mining mobile phone is achieved according to the position coordinate information of the positioning module, the position coordinate information of a target object is input, and the navigation function of the mining mobile phone is achieved through a shortest path algorithm.

[0009] The Chinese patent document no. CN116338751, another application included in the state of the art, discloses a multi-satellite system and ground network conduction fusion high-precision positioning module. The said invention describes a multisatellite system and ground network communication fusion high-precision positioning module. This module carries a satellite communication unit, a satellite navigation unit and a ground network communication unit, navigation enhancement information transmitted by multiple ways from a low-orbit satellite communication link, a low-orbit navigation signal link, a high-orbit communication satellite communication link and a 4G / 5G mobile network transmission link can be received at the same time, so that high-precision positioning is achieved, and the defects of a traditional high-precision positioning mode are effectively overcome.

[0010] The Korean patent document no. KR20210076765, another application included in the state of the art, discloses a method for correcting the monitoring position of the quality of mobile telecommunication network. The said invention relates to a method for correcting a measuring position for quality data of a mobile communication network. This method corrects the measuring position for quality data of the mobile communication network such as the RSRP, the SINR, or the RSSI based on actual GPS position data of an adjacent road and enables to precisely monitor the quality. The relevant method comprises the steps of (a) an analysis server which stores GPS location data performs measurements by moving along a road with a vehicle, and matching these measurements with GPS location data; (b) matching each measuring point with the nearest road; (c) if there is a crossroad on the road, determining whether a certain number of measuring points exist on another road; (d) moving the measuring point with the error to the original road by determining that there is no mapping error if there are a certain number of measuring points on the other road, or that there is a mapping error if there is not. The movement to the original road in the step fourth step is conducted by using the azimuth and distance calculated in the first step. The steps (b) to (d) are conducted targeting the quality data having the measuring positions with a predetermined interval or more.

[0011] The United States patent document no. US2024192385, another application included in the state of the art, discloses a method and device for supporting positioning integrity in wireless communication system. The said invention relates to a fifth generation (5G) or sixth generation (6G) communication system for supporting higher data rates. A method performed by a user equipment (UE) in a wireless communication system is provided. The method may comprise the steps of transmitting, to a location server, capability information of the UE related to global navigation satellite system (GNSS) positioning integrity (PI); receiving information about one or more key performance indicators (KPIs) from the location server; and transmitting, to the location server, resultant information about the GNSS PI, based on the one or more KPIs.

[0012] The Chinese patent document no. CN117665884, another application included in the state of the art, discloses an indoor and outdoor seamless positioning and 5G communication integrated terminal. The said invention describes an indoor and outdoor seamless positioning and 5G communication integrated terminal and belongs to the technical field of positioning. The system comprises a wireless communication module, an indoor and outdoor seamless navigation module, a main control unit module and a GNSS antenna. One end of the main control unit module is connected with the indoor and outdoor seamless navigation module for acquiring original data and the other end is connected with the wireless communication module. The invention enables the spatio-temporal information to be transmitted to the cloud server through the 5G signal in real time, and the position information of the user to be displayed in the map. In addition, the terminal is small in size, comprises a battery which is high in reliability and suitable for being used in a low- temperature environment; a low-power-consumption management strategy is integrated, and the endurance time is prolonged to the maximum extent.

[0013] Summary of the Invention

[0014] An object of the present invention is to realize a system developed with the aim of developing an intelligent positioning module by combining GNSS technology with next generation cellular-based positioning technologies.

[0015] Another object of the present invention is to realize a system developed with the aim of enabling artificial intelligence models to be adapted to the positioning module in order to create more specific core network positioning functions (for example, location management function (LMF) and gateway mobile location center (GMLC)) that enable the analysis of the most efficient positioning technique, the maximization of positioning accuracy and the achievement of targeted key performance indicators.

[0016] A further object of the present invention is to realize a system developed with the aim of enabling precise positioning services to be selected with the collective learning method of different positioning algorithms; performing traffic optimization by developing forecasting models for vehicles with historical traffic and location data; and providing uninterrupted positioning services by using existing telecommunication services (2G, 3G, 4G) in case 5G connection cannot be established or is not available.

[0017] A further object of the present invention is to realize a system developed with the aim of creating a service architecture that will run the mobile network service with maximum efficiency by taking into account the expectations of meeting the network density and vehicle positioning requests for mobile vehicles; providing uninterrupted positioning services needed by vehicles in different situations (urban, intercity, etc.) by means of artificial intelligence and multi-source information flow; and enabling operators and service providers to have the capacity to improve service continuity for smart and connected vehicles to a maximum level by virtue of this information.

[0018] A further object of the present invention is to realize a system developed with the aim of contributing to revenue growth by increasing the efficiency of positioning services by means of innovative methods in the form of traffic prediction and road planning integration; increasing customer satisfaction and service quality by providing more accurate and faster positioning by means of the HYPOS and HYPRED modules; gaining advantages in terms of sustainability with reduced carbon dioxide emissions while reducing the time spent on the road and fuel consumption by navigational improvements; and also gaining advantages in emergency rescue operations. Detailed Description of the Invention

[0019] “A System for Providing Artificial Intelligence- Assisted Positioning Services for Vehicles” realized to fulfd the objectives of the present invention is shown in the figure attached, in which:

[0020] Figure 1 is a schematic view of the inventive system.

[0021] The components illustrated in the figure are individually numbered, where the numbers refer to the following:

[0022] 1. System

[0023] 2. Database

[0024] 2.1. Road Information Database

[0025] 2.2. Network Information Database

[0026] 2.3. Current User Location Database

[0027] 2.4. Historical User Location Database

[0028] 2.5. Data Anonymization Module

[0029] 3. Core Network

[0030] 4. Gateway Mobile Location Center

[0031] 5. Location Management Function

[0032] 6. HYPRED Module

[0033] 6.1. Traffic Prediction Module

[0034] 6.2. Road Matching Module

[0035] 7. HYPOS Module

[0036] 7.1. Traditional Positioning Methods Module

[0037] 7.2. Artificial Intelligence-Assisted Positioning Module

[0038] A. Vehicle The inventive system (1) developed with the aim of developing an intelligent positioning module by combining global navigation satellite systems technology with next generation cellular-based positioning technologies comprises at least one database (2) which is configured to keep a record of data in the form of road maps, geographical conditions, building locations and road status information via the road information database (2.1) running thereon; to keep a record of network information comprising cell identities, signal power values, neighbour cell information, coverage maps, timestamps and location information, non-3GPP access networks (WiFi, UWB, LoraWAN etc.) data and sensor data (BLE, UBP, TBS etc.) via the network information database (2.2) running thereon; to keep a record of the current location of users via the current user location database (2.3) running thereon; to keep a record of the past locations of users via the historical user location database (2.4) running thereon; to keep data in the form of “hash”, therefore, to protect data security; and to provide user privacy by replacing subscriber information with a special “hash” via the data anonymization module (2.5) running thereon; at least one core network (3) which is configured to enable data exchange; to perform centralized management; and to enable all modules and databases to operate efficiently; at least one Gateway Mobile Location Center (4) which is configured to be located inside the core network (3); to store location information by collecting it in a central location; to provide traffic predictions and road recommendations; and to query subscribers on different Radio Access Technology (RAT); at least one Location Management Function (5) which is configured to be located inside the core network (3); to collect and process location information; to run positioning algorithms; and to enable 3G SAS and 4G E-SMLCs to use this data with the relevant APIs; at least one HYPRED module (6) configured to be a module integrated into the Gateway Mobile Location Center (4); to determine estimated positions by predicting traffic and network load and using historical information, and to provide 2G, 3G, 4G and 5G positioning services when needed; to enable the best travel option to be recommended by means of the integration of traffic prediction results with route planning, taking into account network conditions and route status; to predict future vehicle (A) traffic and the mobile network load related to it by using historical vehicle (A) traffic data via the traffic prediction module (6.1) running thereon; to enable the network to be used efficiently by predicting traffic density and network usage beforehand; to match the current road status, road speed limits and network conditions and to optimize the routes of vehicles (A) with the data obtained from the traffic prediction module via the road matching module (6.2) running thereon; at least one HYPOS module (7) which is configured to be a positioning algorithm module integrated into the Location Management Function (5); to be a machine learning algorithm module comprising different models that can perform positioning with higher efficiency and accuracy by feeding each other instead of going through a single model, since different positioning techniques such as time of arrival difference, carrier phase, A-GNSS, and RTK have specific advantages and disadvantages; to enable the most effective positioning method to be determined by using historical data in the form of a road map and available network resources in the current location of the vehicles (A); to update the architecture of the ensemble models with the aim of improving accuracy by using the outputs of the forecasting models provided by the HYPRED module (6); to determine the position of vehicles (A) by using traditional positioning techniques in the form of angle of arrival and time difference of arrival via the traditional positioning methods module (7.1) running thereon; to integrate different models by using machine learning algorithms via the artificial intelligence-assisted positioning module (7.2) running thereon; to provide positioning with higher accuracy and efficiency by using data obtained through traditional methods; to determine the most effective positioning method by using historical data in the form of road maps and available network resources; and to continuously update these methods.

[0039] The database (2) included in the inventive system (1) is configured to establish connection with the HYPRED module (6) and the HYPOS module (7). The road information database (2.1) running on the database (2) is configured to keep a record of data in the form of road maps, geographical conditions, building locations and road status information. The network information database (2.2) running on the database (2) is configured to keep a record of network information comprising cell identities, signal power values, neighbour cell information, coverage maps, timestamps and location information, non-3GPP access networks (WiFi, UWB, LoraWAN etc.) data and sensor data (BLE, UBP, TBS etc.). The current user location database (2.3) running on the database (2) is configured to keep a record of the current location of users. The historical user location database (2.4) running on the database (2) is configured to keep a record of past locations of users, to store this data to be used in the HYPRED module (6) and the HYPOS module (7), and to keep data in the form of “hash”, therefore, to protect data security. The data anonymization module (2.5) running on the database (2) is configured to provide user privacy by replacing subscriber information with a special “hash”.

[0040] The core network (3) included in the inventive system (1) is configured to enable data exchange; to perform centralized management; and to enable all modules and databases to operate efficiently by working in integration with Gateway Mobile Location Center (GMLC) (4) and Location Management Function (LMF) (5).

[0041] The Gateway Mobile Location Center (4) included in the inventive system (1) is configured to be inside the core network (3). The Gateway Mobile Location Center (4) is configured to store location information by collecting it in a central location; to provide traffic predictions and road recommendations through the HYPRED module (6); and to query subscribers on different RATs. The Location Management Function (5) included in the inventive system (1) is configured to be inside the core network (3). The Location Management Function (5) is configured to collect and process location information; to run positioning algorithms through HYPOS module (7); and to enable 3G SAS and 4G E-SMLCs to use this data with the relevant APIs.

[0042] The HYPRED module (6) included in the inventive system (1) is configured to be a module integrated into the Gateway Mobile Location Center (GMLC) (4). The HYPRED module (6) is configured to determine estimated positions by predicting traffic and network load and using historical information, and to provide 2G, 3G, 4G and 5G positioning services when needed. The HYPRED module (6) is configured to enable the best travel option to be recommended by means of the integration of traffic prediction results with route planning, taking into account network conditions and route status. The HYPRED module (6) is configured to enable network and route data to be used optimally. The HYPRED module (6) is configured to store network information in the form of database, timestamps, cell identities in the GMLC (4) and location provided by the LMF (5), as well as external information in the form of traffic data and road maps. The HYPRED module (6) is configured to perform positioning over different Radio Access Technologies (RAT) and to provide position improvement by superposition of different positions. The traffic prediction module (6.1) running on the HYPRED module (6) is configured to predict future vehicle (A) traffic and the mobile network load related to it by using historical vehicle (A) traffic data; and to enable the network to be used efficiently by predicting traffic density and network usage beforehand. The road matching module (6.2) running on the HYPRED module (6) is configured to match the current road status, road speed limits and network conditions and to optimize the routes of vehicles (A) with the data obtained from the traffic prediction module.

[0043] The HYPOS module (7) included in the inventive system (1) is configured to be a positioning algorithm module integrated into the Location Management Function (LMF) (5). The HYPOS module (7) is configured to be a machine learning algorithm module comprising different models that can perform positioning with higher efficiency and accuracy by feeding each other instead of going through a single model, since different positioning techniques (angle of arrival (AoA), time of arrival difference (ToA), carrier phase, A-GNSS (assisted-GNSS), RTK (realtime kinematics)). The HYPOS module (7) is configured to enable the most effective positioning method to be determined by using historical data in the form of a road map and available network resources in the current location of the vehicles (A). The HYPOS module (7) is configured to update the architecture of the ensemble models with the aim of improving accuracy by using the outputs of the forecasting models provided by the HYPRED module (6). The traditional positioning methods module (7.1) running on the HYPOS module (7) is configured to determine the position of vehicles (A) by using traditional positioning techniques in the form of angle of arrival (AoA) and time difference of arrival (ToA). The artificial intelligence-assisted positioning module (7.2) running on the HYPOS module (7) is configured to integrate different models by using machine learning algorithms; to provide positioning with higher accuracy and efficiency by using data obtained through traditional methods; to determine the most effective positioning method by using historical data in the form of road maps and available network resources; and to continuously update these methods.

[0044] Industrial Application of the Invention

[0045] In the inventive system (1), the HYPOS module (7) and the HYPRED module (6) continuously receive data from data storage units and network functions. Traditional methods and artificial intelligence-assisted models use these data to determine the location of vehicles. The artificial intelligence models that will be used herein recommend the most suitable route through the traffic prediction module and the road recommendation module by analyzing the network and road status. The data flow in the artificial intelligence model takes place in three main stages as pre-processing and data cleaning, feeding data into the model and postprocessing. In the first stage, pre-processing and data cleaning, raw data coming from the HYPRED module (6) - such as network information, traffic data and external road maps - are first filtered and standardized. In this stage, missing, inconsistent or noisy data is detected and corrected, therefore, errors that might adversely affect the performance of the model are minimized. Then, the cleaned data is fed to the machine learning algorithms in the HYPOS module (7). In this stage, each of the ensemble models analyzes different positioning techniques and evaluates the advantages and disadvantages of these techniques. As the data flows through the layers of the model, it is continuously updated and optimized with forecasting data coming from HYPRED. Finally, in the post-processing phase, the predictions and results produced by the model are analyzed before being transmitted to decision support systems and the positioning service. In this stage, the results are evaluated in terms of accuracy and reliability, and additional corrections are made where necessary. The HYPOS module (7) determines the most efficient positioning method by continuously updating the obtained location data. The HYPRED module (6), on the other hand, enables the network and routes to be used in the best possible way by continuously optimizing traffic predictions and road recommendations. Therefore, by regularly feeding the system (1) with fresh data, the output of the prediction algorithm can be continuously improved. Here, the driving test can also be fed with Minimize Drive Test (MDT) information that can be obtained from terminals, as well as location and network measurement information that can be obtained from drones and similar aerial vehicles (UAVs). All modules enable data to be processed securely and quickly by working in integration with the core network (3) and data storage units. This integration enables users to access the most accurate and up-to-date location information and follow the most efficient routes.

[0046] Within these basic concepts; it is possible to develop various embodiments of the inventive “A System (1) for Providing Artificial Intelligence-Assisted Positioning Services for Vehicles”; the invention cannot be limited to examples disclosed herein and it is essentially according to claims.

Claims

CLAIMS1. A system ( 1 ) developed with the aim of developing an intelligent positioning module by combining global navigation satellite systems technology with next generation cellular-based positioning technologies; comprising at least one database (2) which is configured to keep a record of data in the form of road maps, geographical conditions, building locations and road status information via the road information database (2.1) running thereon; to keep a record of network information comprising cell identities, signal power values, neighbour cell information, coverage maps, timestamps and location information, non-3GPP access networks (WiFi, UWB, LoraWAN etc.) data and sensor data (BLE, UBP, TBS etc.) via the network information database (2.2) running thereon; to keep a record of the current location of users via the current user location database (2.3) running thereon; to keep a record of the past locations of users via the historical user location database (2.4) running thereon; to keep data in the form of “hash”, therefore, to protect data security; and to provide user privacy by replacing subscriber information with a special “hash” via the data anonymization module (2.5) running thereon; at least one core network (3) which is configured to enable data exchange; to perform centralized management; and to enable all modules and databases to operate efficiently; at least one Gateway Mobile Location Center (4) which is configured to be located inside the core network (3); to store location information by collecting it in a central location; to provide traffic predictions and road recommendations; and to query subscribers on different Radio Access Technology (RAT); at least one Location Management Function (5) which is configured to be located inside the core network (3); to collect and process location information; to run positioning algorithms; and to enable 3G SAS and 4G E-SMLCs to use this data with the relevant APIs; and characterized byat least one HYPRED module (6) configured to be a module integrated into the Gateway Mobile Location Center (4); to determine estimated positions by predicting traffic and network load and using historical information, and to provide 2G, 3G, 4G and 5G positioning services when needed; to enable the best travel option to be recommended by means of the integration of traffic prediction results with route planning, taking into account network conditions and route status; to predict future vehicle (A) traffic and the mobile network load related to it by using historical vehicle (A) traffic data via the traffic prediction module (6.1) running thereon; to enable the network to be used efficiently by predicting traffic density and network usage beforehand; to match the current road status, road speed limits and network conditions and to optimize the routes of vehicles (A) with the data obtained from the traffic prediction module via the road matching module (6.2) running thereon; at least one HYPOS module (7) which is configured to be a positioning algorithm module integrated into the Location Management Function (5); to be a machine learning algorithm module comprising different models that can perform positioning with higher efficiency and accuracy by feeding each other instead of going through a single model, since different positioning techniques such as time of arrival difference, carrier phase, A-GNSS, and RTK have specific advantages and disadvantages; to enable the most effective positioning method to be determined by using historical data in the form of a road map and available network resources in the current location of the vehicles (A); to update the architecture of the ensemble models with the aim of improving accuracy by using the outputs of the forecasting models provided by the HYPRED module (6); to determine the position of vehicles (A) by using traditional positioning techniques in the form of angle of arrival and time difference of arrival via the traditional positioning methods module (7.1) running thereon; to integrate different models by using machine learning algorithms via the artificial intelligence-assisted positioning module (7.2) running thereon; toprovide positioning with higher accuracy and efficiency by using data obtained through traditional methods; to determine the most effective positioning method by using historical data in the form of road maps and available network resources; and to continuously update these methods.

2. A system (1) according to Claim 1 ; characterized by the database (2) which is configured to establish connection with the HYPRED module (6) and the HYPOS module (7).

3. A system (1) according to Claim 1 or 2; characterized by the database (2) which is configured to keep a record of data in the form of road maps, geographical conditions, building locations and road status information through the road information database (2.1) running thereon.

4. A system (1) according to Claim 3; characterized by the database (2) which is configured to keep a record of network information comprising cell identities, signal power values, neighbour cell information, coverage maps, timestamps and location information, non-3GPP access networks (WiFi, UWB, LoraWAN etc.) data and sensor data (BLE, UBP, TBS etc.) through the network information database (2.2) running thereon.

5. A system (1) according to any one of the preceding claims; characterized by the database (2) which is configured to keep a record of the current location of users through the current user location database (2.3) running thereon.

6. A system (1) according to any one of the preceding claims; characterized by the database (2) which is configured to keep a record of past locations of users, to store this data to be used in the HYPRED module (6) and the HYPOS module (7), and to keep data in the form of “hash”, therefore, to protect data security through the historical user location database (2.4) running thereon.

7. A system (1) according to any one of the preceding claims; characterized by the database (2) which is configured to provide user privacy by replacing subscriber information with a special “hash” through the data anonymization module (2.5) running thereon.

8. A system (1) according to any one of the preceding claims; characterized by the core network (3) which is configured to enable data exchange; to perform centralized management; and to enable all modules and databases to operate efficiently by working in integration with Gateway Mobile Location Center (4) and Location Management Function (5).

9. A system (1) according to any one of the preceding claims; characterized by the Gateway Mobile Location Center (4) which is configured to be inside the core network (3).

10. A system (1) according to any one of the preceding claims; characterized by the Gateway Mobile Location Center (4) which is configured to store location information by collecting it in a central location; to provide traffic predictions and road recommendations through the HYPRED module (6); and to query subscribers on different RATs.

11. A system (1) according to any one of the preceding claims; characterized by the Location Management Function (5) which is configured to be inside the core network (3).

12. A system (1) according to any one of the preceding claims; characterized by the Location Management Function (5) which is configured to collect and process location information; to run positioning algorithms through HYPOS module (7); and to enable 3G SAS and 4G E-SMLCs to use this data with the relevant APIs.

13. A system (1) according to any one of the preceding claims; characterized by the HYPRED module (6) which is configured to be a module integrated into the Gateway Mobile Location Center (4).

14. A system (1) according to any one of the preceding claims; characterized by the HYPRED module (6) which is configured to determine estimated positions by predicting traffic and network load and using historical information, and to provide 2G, 3G, 4G and 5G positioning services when needed.

15. A system (1) according to any one of the preceding claims; characterized by the HYPRED module (6) which is configured to enable the best travel option to be recommended by means of the integration of traffic prediction results with route planning, taking into account network conditions and route status.

16. A system (1) according to any one of the preceding claims; characterized by the HYPRED module (6) which is configured to enable network and route data to be used optimally.

17. A system (1) according to any one of the preceding claims; characterized by the HYPRED module (6) which is configured to store network information in the form of database, timestamps, cell identities in the GMLC (4) and location provided by the LMF (5), as well as external information in the form of traffic data and road maps.

18. A system (1) according to any one of the preceding claims; characterized by the HYPRED module (6) which is configured to perform positioning over different Radio Access Technologies and to provide position improvement by superposition of different positions.

19. A system (1) according to any one of the preceding claims; characterized by the HYPRED module (6) which is configured to predict future vehicle (A) trafficand the mobile network load related to it by using historical vehicle (A) traffic data; and to enable the network to be used efficiently by predicting traffic density and network usage beforehand through the traffic prediction module (6.1) running thereon.

20. A system (1) according to any one of the preceding claims; characterized by the HYPRED module (6) which is configured to match the current road status, road speed limits and network conditions and to optimize the routes of vehicles (A) with the data obtained from the traffic prediction module through the road matching module (6.2) running thereon.

21. A system (1) according to any one of the preceding claims; characterized by the HYPOS module (7) which is configured to be a positioning algorithm module integrated into the Location Management Function (5).

22. A system (1) according to any one of the preceding claims; characterized by the HYPOS module (7) which is configured to be a machine learning algorithm module comprising different models that can perform positioning with higher efficiency and accuracy by feeding each other instead of going through a single model, since different positioning techniques such as time of arrival difference, carrier phase, A-GNSS, and RTK have specific advantages and disadvantages23. A system (1) according to any one of the preceding claims; characterized by the HYPOS module (7) which is configured to enable the most effective positioning method to be determined by using historical data in the form of a road map and available network resources in the current location of the vehicles (A).

24. A system (1) according to any one of the preceding claims; characterized by the HYPOS module (7) which is configured to update the architecture of the ensemble models with the aim of improving accuracy by using the outputs of the forecasting models provided by the HYPRED module (6).

25. A system (1) according to any one of the preceding claims; characterized by the HYPOS module (7) which is configured to determine the position of vehicles (A) by using traditional positioning techniques in the form of angle of arrival and time difference of arrival through the traditional positioning methods module (7.1) running thereon.

26. A system (1) according to any one of the preceding claims; characterized by the HYPOS module (7) which is configured to integrate different models by using machine learning algorithms through the artificial intelligence-assisted positioning module (7.2) running thereon; to provide positioning with higher accuracy and efficiency by using data obtained through traditional methods; to determine the most effective positioning method by using historical data in the form of road maps and available network resources; and to continuously update these methods.

Images