Bluetooth and UWB fusion positioning method and system based on Kalman filtering

Abstract

The invention belongs to the technical field of mobile wireless positioning, and particularly relates to a Bluetooth and UWB fusion positioning method and system based on Kalman filtering, and the method comprises the steps: taking a UWB distance measurement value and a distance converted by a Bluetooth signal distance conversion model as an observation vector through combining the low power consumption characteristic of a Bluetooth signal and the high precision advantage of UWB; according to the real-time position of a detected target, a Kalman filtering algorithm (Kalman filtering algorithm) is adopted to carry out data fusion processing, so that real-time positioning of an observation target is realized. According to the algorithm, a positioning service area is divided into a high-precision positioning area, a common-precision positioning area and a connection area according to the positioning precision requirement, and UWB positioning anchor point equipment and BLE positioning anchor point equipment are arranged in different positioning areas at the same time. And an indoor fixed person navigation positioning model is formed by adopting UWB and BLE pedestrian navigation constraints for the connection area, so that the requirements of relatively high precision and relatively low cost are met.

Description

Technical Field

[0001] This invention relates to mobile wireless positioning technology, and more specifically, to a Bluetooth and UWB fusion positioning method and system based on Kalman filtering. Background Technology

[0002] With the development of vehicle-mounted wireless positioning technology, single-path positioning technology has gradually matured, but single technologies are no longer sufficient to meet the needs of complex vehicle positioning environments. Currently, vehicle-mounted Bluetooth technology suffers from unstable connections, limited coverage, susceptibility to signal interference, and vulnerability to relay attacks. While UWB positioning solutions offer accurate positioning, they face challenges such as high cost, low technological maturity, poor compatibility, and high power consumption.

[0003] In this context, multi-technology fusion has become a trend in vehicle positioning. By combining the advantages of various positioning technologies, a fusion positioning system is formed to achieve better positioning results. The combination of UWB and Bluetooth embodies this trend. Through algorithm optimization and dynamic adjustment of signal weights, this method can maintain high-precision positioning and anti-interference capabilities even in complex environments, meeting the high-precision requirements of vehicle positioning. Multi-technology fusion not only improves positioning accuracy but also enhances the system's robustness and adaptability. Through intelligent algorithm optimization, the system can dynamically adjust the weights of each technology, ensuring stable operation in different environments and meeting diverse application needs.

[0004] In terms of market demand, the global digital key market is projected to reach $3.5 billion by 2025, but the overall penetration rate of new energy vehicles with higher intelligent requirements remains low. There is an urgent need for high-security, low-cost key solutions for vehicle keys. With continuous technological advancements, integrated positioning systems will become the mainstream choice for future vehicle keys, contributing to the intelligent development of new energy vehicles and meeting the market's pressing demand for high-security, low-cost solutions. Summary of the Invention

[0005] To address the insufficient accuracy of current vehicle-mounted wireless positioning methods that employ a single technology approach, this invention proposes a positioning method that integrates UWB and BLE technologies. Through dynamic weight adjustment and error correction, it achieves high-precision and low-cost vehicle-mounted positioning functionality.

[0006] To achieve the above objectives, the present invention provides the following technical solution: A Kalman filter-based Bluetooth and UWB fusion positioning method includes the following steps: acquiring UWB ranging values ​​and the distance converted from Bluetooth signals using a distance conversion model to construct an observation vector; performing data fusion processing on the observation vector using a Kalman filter algorithm to determine the real-time location of the target; dividing the positioning service area into a high-precision positioning area, a normal-precision area, and a connecting area according to positioning accuracy requirements; employing a UWB positioning algorithm in the high-precision positioning area, a Bluetooth spatial fingerprint algorithm based on RSSI in the normal-precision area, and a UWB / BLE combined positioning algorithm in the connecting area. Specifically, the positioning area is divided according to positioning accuracy requirements, and UWB and BLE anchor point devices are deployed; in the connecting area, UWB and BLE motion trajectory constraints are applied to form a navigation and positioning model, achieving high-precision, low-cost positioning.

[0007] Furthermore, to address the differences in observation errors between UWB and Bluetooth ranging values, as well as the ranging errors caused by indoor environmental interference, a UWB / BLE Kalman positioning algorithm model is constructed. This algorithm model reduces the impact of environmental interference on ranging accuracy and improves the reliability of positioning data. For the real-time status of the positioning system in different environments, the weights of UWB and BLE signals are adaptively adjusted to optimize data fusion, reduce signal interference and multipath effects, and ensure positioning accuracy and stability, meeting the requirements for precise target tracking in complex vehicle environments.

[0008] Furthermore, when the positioning tag detects only a Bluetooth signal, and there is at least one valid signal source from the Bluetooth positioning anchor point, the data is processed using a Bluetooth spatial fingerprint algorithm based on Received Signal Strength Indicator (RSSI). For positioning in ordinary areas, the Bluetooth wireless signal can be functionally ranged using RSSI signal strength. Due to the limited spatial range of Bluetooth signals, a spatial fingerprint method is used for positioning. During the positioning process, the complex in-vehicle environment can affect the ideal model and the actual received signal strength, causing a certain error in the displayed signal strength. A series of machine learning algorithms are used to fit the positioning data and cluster the positioning data towards the actual model. The corresponding Bluetooth anchor points will provide the corresponding signal strength for target testing. During the offline phase, the target's signal is collected to establish a fingerprint database. Simultaneously, software is used to statistically analyze the collected Bluetooth signal strength, design the step size, and fit the signal strength versus distance curve. During the online phase, the positioning signal strength of each point was received, and offline data was used to match it accordingly. Furthermore, the positioning accuracy was optimized using KNN (K-nearest Neighbors Algorithm), SVM (Support Vector Machine), Random Forest machine learning, and Particle Swarm Optimization.

[0009] Furthermore, when the signals collected by the fusion positioning tag are only UWB signals, and there is at least one effective signal source for the positioning anchor point, UWB positioning is performed using ultra-wideband positioning algorithms based on Time of Arrival (ToA), Time of Difference of Arrival (TDoA), Angle of Arrival (AoA), Single-Side Two-Way Ranging (SS-TWR), and Double Side Two-Way Ranging (DS-TWR).

[0010] Furthermore, when the positioning tag acquires both UWB and BLE signals, the two signals are fused. Based on a weighted allocation mechanism of uncertainty in model predictions and actual measurements, the positioning results are optimized. The fused signal is then input into the UWB-BLE ranging model to obtain the distance between the fused positioning tag and each positioning anchor point. This distance is used as the input excitation for the Kalman filter / particle filter algorithm, and the initial position of the positioning tag and the trilateration algorithm are applied to calculate the initial position of the positioning target. Since the excitation changes constantly and new excitations are input, the positioning coordinates are continuously updated and predicted.

[0011] Furthermore, the acquired UWB and BLE signals are subjected to joint Kalman filtering, and accurate target identification is achieved through a dynamic weight adjustment mechanism. Based on the current state of the detected target, positioning calibration is performed at that moment. Simultaneously, the system error covariance for the next moment is predicted based on the previous system error covariance, and the Kalman gain of the fused positioning for the next moment is calculated. This leads to the calculation of the optimal state variables of the current fused positioning system, i.e., the optimal positioning coordinates. When the object moves, the system covariance at the current moment is recalculated, and this process is repeated to calculate the real-time positioning coordinates under moving conditions, ensuring dynamic tracking accuracy.

[0012] Furthermore, by using BLE and UWB ranging values ​​as observation vectors and combining them with historical positioning data of the target, observation errors are dynamically corrected to further improve positioning accuracy. Different positioning algorithms are applied in different positioning areas. Finally, a combined UWB / BLE vehicle navigation and positioning trajectory is obtained, thus achieving a relatively high-precision and relatively low-cost optimal vehicle positioning solution.

[0013] The Bluetooth and UWB fusion positioning system based on Kalman filtering includes UWB positioning anchors, Bluetooth positioning anchors, positioning tags, a main control chip (MCU), an NB-IoT module, a power management module, and a clock circuit. The UWB and Bluetooth positioning anchors are respectively deployed on the object under test. The positioning tags are used to collect UWB ranging values ​​and Bluetooth signal strength to construct an observation vector. The MCU receives and processes UWB and Bluetooth positioning information, using a Kalman filter algorithm for data fusion processing. The NB-IoT module wirelessly forwards positioning information. The power management module provides unified power to the UWB module, Bluetooth module, and MCU. The clock circuit calibrates the MCU clock signal to ensure time synchronization accuracy.

[0014] Furthermore, the MCU interacts with the UWB communication module and the Bluetooth communication module through the SPI bus interface to achieve synchronous information transmission between the modules; The power management module adopts a low-power design, and isolation circuits are used between modules to prevent electromagnetic interference and ensure stable system operation.

[0015] Furthermore, the positioning system also includes positioning engine software, which receives data sent by positioning anchor points, embeds a fusion positioning algorithm, calculates the real-time position coordinates of the target, saves the results to the database, and uploads them to the host computer for display.

[0016] Furthermore, the positioning system enables communication between the UWB anchor point, Bluetooth anchor point, and MCU via CAN BUS or LIN bus. The positioning information is sent to the background monitoring platform through the vehicle T-BOX and 4G module to achieve remote real-time monitoring.

[0017] By adopting the above technical solution, the beneficial effects of the present invention are as follows: The UWB / BLE combined localization algorithm is a fusion localization algorithm based on UWB and BLE. Its function is to comprehensively process data from multiple different types of sensors and optimize the localization result through a dynamic weight allocation mechanism. Unlike traditional single-sensor source localization algorithms, the fusion algorithm uses Kalman filtering and particle filtering methods to infer and estimate the localization data measured by different sensors.

[0018] The method of this invention combines the low power consumption of Bluetooth signals with the high precision of UWB. It uses the distance converted from UWB ranging values ​​and Bluetooth signal distance conversion model as observation vectors, and employs the Kalman Filtering Algorithm to perform data fusion processing for the real-time position of the detected target, thereby achieving real-time positioning of the observed target.

[0019] The algorithm divides the positioning service area into high-precision positioning areas, normal-precision positioning areas, and connecting areas according to their positioning accuracy requirements. At the same time, UWB and BLE positioning anchor devices are deployed in different positioning areas.

[0020] For the transition area, an indoor person-to-person navigation and positioning model is constructed using UWB and BLE pedestrian navigation constraints to meet the requirements of relatively high accuracy and relatively low cost. Simultaneously, in the transition area, UWB and Bluetooth positioning information are used for fusion positioning to achieve partitioning and smooth switching between the UWB and Bluetooth positioning systems.

[0021] For areas requiring standard precision positioning, the BLE testing distance is significantly greater than the UWB testing distance; therefore, signals received at greater distances are typically BLE signals. Given the complex real-world in-vehicle environment and the inherent limitations of ideal models, as well as the inherent errors in actual received signal strength, machine learning algorithms, such as least squares, are used for curve fitting to obtain a smooth RSSI curve. This allows for better fitting of the positioning data, clustering it as close as possible to the actual model.

[0022] The algorithm uses target motion features as a dynamic vector and stores the target velocity information and surrounding environment information as target navigation constraint information. By dynamically adjusting signal weights, it ensures stable operation in different environments, effectively overcoming the limitations of traditional positioning technologies. Through combined UWB and BLE navigation and positioning, the system achieves real-time correction of positioning errors, improving navigation accuracy. Optimized processing of positioning data using a Kalman filter algorithm enables seamless integration of UWB and Bluetooth data, enhancing positioning accuracy and stability. It effectively overcomes the limitations of single positioning methods, ensuring high-precision positioning even under signal interference and multipath effects, making it suitable for accurate target tracking in complex vehicular environments. Attached Figure Description

[0023] Figure 1 This is a block diagram of the positioning system of the present invention.

[0024] Figure 2 This is a block diagram of the integrated positioning circuit combining UWB and Bluetooth in this invention.

[0025] Figure 3 This is a block diagram of the integrated positioning and communication architecture in this invention.

[0026] Figure 4 This is a schematic diagram of the positioning accuracy area division in this invention.

[0027] Figure 5 This is a graph showing the original RSSI data from a Bluetooth key test in an embodiment of the present invention.

[0028] Figure 6 This is a graph of RSSI data after filtering during Bluetooth key testing according to an embodiment of the present invention.

[0029] Figure 7 This is a schematic diagram illustrating the principle of the SS-TWR ranging algorithm in an embodiment of the present invention.

[0030] Figure 8 This is a schematic diagram illustrating the principle of the DS-TWR ranging algorithm in an embodiment of the present invention.

[0031] Figure 9 This is a flowchart illustrating the software workflow for locating sub-anchor points according to an embodiment of the present invention.

[0032] Figure 10 This is a flowchart illustrating the software workflow for locating the main anchor point in an embodiment of the present invention.

[0033] Figure 11 This is a flowchart of the host computer software in an embodiment of the present invention.

[0034] Figure 12 This is a positioning result diagram obtained after analysis in an embodiment of the present invention. Detailed Implementation

[0035] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0036] In the description of this invention, it should be understood that the terms "upper", "lower", "front", "rear", "left", "right", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0037] This invention's vehicle-mounted fusion positioning system is based on one UWB anchor point and multiple Bluetooth anchor points for positioning. First, UWB and Bluetooth anchor points are installed at fixed locations on the vehicle. Since Bluetooth positioning is based on fingerprint positioning, an offline fingerprint database needs to be built in advance, and then real signal strength information can be collected during the online phase. When the tag is carried by the person to be located into the signal range of the UWB anchor point, distance measurements are taken with the anchor point at regular intervals. The tag, as the device being located, actively initiates a distance measurement request. The UWB anchor point receives the distance measurement request initiated by the tag and participates in the distance measurement process. Simultaneously, the tag collects Bluetooth signal strength information within its range. Afterward, the tag sends the distance measurement information and the collected Bluetooth positioning information to the UWB anchor point, which then transmits the location information to the server.

[0038] This paper presents a fusion positioning circuit design, integrating UWB and Bluetooth chips into a system architecture, employing a modular design to enhance scalability. The design incorporates a fusion circuit for UWB and BLE positioning, providing both Bluetooth and UWB positioning capabilities, as well as fusion positioning functionality. A single power supply circuit manages power to ensure stable power supply and efficient collaboration between the UWB and BLE modules. The power supply circuit uses the same voltage parameters for the UWB, NB-IoT, and BLE modules, reducing the number of power chips and saving design costs. The power management module features a low-power design to extend device battery life. Isolation circuits are used between modules to prevent interference and shield against electromagnetic interference.

[0039] The data interaction and synchronization between the UWB chip and the BLE chip are achieved through an MCU chip, enabling information transmission between the modules. Simultaneously, a clock circuit is used to calibrate the MCU module to ensure time synchronization accuracy and improve the accuracy of positioning data. The clock circuit design uses a high-precision crystal oscillator to ensure frequency stability. Data interaction is achieved through the SPI bus to improve transmission efficiency. In this embodiment, the IoT chip is the Quectel BC95 850MHz chip, the Bluetooth chip is the NXP-KW45, and the UWB chip is the NXP-NCJ29D6.

[0040] In the software system architecture, UWB anchor points communicate with the MCU via LIN or CAN bus to achieve information fusion. BLE sub-anchor points and the BLE main anchor point exchange information via LIN or CAN bus. The MCU processes the received UWB and BLE positioning information. Positioning information is transmitted to the vehicle-mounted T-BOX via CAN bus2 for vehicle-to-ground location data collection. Simultaneously, it is transmitted to the backend via the 4G module for monitoring.

[0041] Different positioning accuracy intervals are defined. Bluetooth anchors deployed in the positioning environment continuously transmit positioning signals. The positioning tag transmits the RSSI values ​​of surrounding Bluetooth signals received at the same time to the UWB positioning anchors and the positioning system via the UWB module. The positioning system stores the Bluetooth RSSI values ​​received by the positioning tag at the same time into the same frame array. Then, it finds the maximum RSSI value from these data.

[0042]

[0043] judge The signal source is determined by the fact that each Bluetooth anchor point marks its own location at the beginning. Therefore, the positioning area of ​​the anchor point can be determined by the following formula.

[0044]

[0045] In the formula The maximum RSSI value received by the tag is used. Then, median filtering is applied to smooth the data, eliminating random errors and improving the accuracy of region segmentation. A fixed-size data block is selected, and the median within that block is used as the filtering result. Median filtering is then applied to... Outliers in the data are corrected. A sliding window mode filter is used to correct for recurring outliers, resulting in filtered data. Its data value represents the final region determination result.

[0046] Based on the calculated partitioning results, the overall region can be divided into high-precision regions, normal-precision regions, and transitional regions using a partitioning strategy. Division into a high-precision region: When... If the partitioning result ultimately determines that the tag is in a high-precision zone, then the tag is considered to be in a high-precision zone, and the UWB positioning algorithm can be used to calculate the tag's position in the high-precision zone. If the partitioning result is classified as a transition zone: The final determination is that the tag is not in the high-precision zone. At this point, the tag can receive both Bluetooth positioning signals and UWB ranging information, indicating it is in the transition zone. In the transition zone, a UWB and Bluetooth fusion algorithm can be used for calculation. It is then classified as a normal positioning zone: If... If the final determination of the partitioning result is that the tag is not in the high-precision zone, it can be determined that the tag is in the normal-precision zone. At this time, the positioning tag is calculated according to the Bluetooth positioning signal.

[0047] The specific process of implementing the positioning sub-anchor point software is as follows: The positioning tag continuously collects RSSI signal data from Bluetooth anchor points within the positioning area, and divides the area into high-precision positioning zone, normal-precision positioning zone, and transition zone using a precise partitioning strategy. Based on the partitioning results, relevant positioning methods are used for positioning. In the normal-precision positioning zone, the RSSI algorithm is used for positioning; in the high-precision zone, the DS-TWR algorithm is used; and for the transition zone, a Kalman filter algorithm is used to filter and output the final value. Finally, the comprehensive coordinate value is selected for output.

[0048] For the software implementation of the positioning sub-anchor point: 1) When the positioning tag starts, the Bluetooth module begins receiving signals from the Bluetooth anchor point and stores the RSSI value and anchor point identifier in its internal memory; 2) When the positioning tag's UWB module needs to send a UWB positioning signal, it adds a data field to its UWB request frame to transmit the anchor point identifier and RSSI value stored in its internal memory; 3) The positioning tag sends a UWB request frame and waits for a response frame from the UWB anchor point; 4) If no response frame is received within a specified time, the system initializes and restarts broadcasting the request; 5) If a response frame is received within a specified time, the system continues to determine if there are any abnormalities in the received information (device malfunction, positioning coordinate malfunction, low battery malfunction, etc.). If the positioning information is abnormal, the UWB module on the positioning sub-anchor point will stop broadcasting request frames and wait for the anchor point to reply with a response frame. Otherwise, it will continue broadcasting request frames.

[0049] For the master anchor software design: The Bluetooth master anchor continuously broadcasts Bluetooth signals during operation so that the positioning tag can receive and calculate the RSSI value and anchor identifier. The broadcast frequency, power, and channel parameters of the Bluetooth master anchor can be adjusted to meet the positioning requirements of different scenarios. The UWB master anchor's workflow is the same as the positioning tag, including power-on, initialization, and anomaly detection. Afterward, the UWB master anchor waits for signals from the positioning tag. When a signal is received, it records the signal arrival time and parses the data in the request frame. After parsing the data, the UWB master anchor packages the data and transmits it to the server positioning engine for positioning calculation and result output.

[0050] Host Computer Software Design: The host computer application software used in this embodiment calculates the location of the positioning tag and the positioning results within the positioning engine and its connected host computer system, and displays the results on the host computer. The system automatically initializes after powering the anchor point, ensuring it is in the correct working state. The positioning anchor point packages the collected data and transmits it to the positioning engine. The positioning engine parses and processes the received data to obtain the raw data, which is then saved to the database for later use. The main anchor point parameter information in the database is then updated and transmitted to the positioning engine to help calculate accurate positioning coordinates. Finally, after the positioning engine calculates the position coordinates, it saves the calculated coordinate data to the database and uploads the data to the application platform to display the coordinate location.

[0051] In this embodiment, four Bluetooth anchor points and three UWB positioning anchor points are deployed. The UWB data ports are connected to the PoE switch via network cables. The UWB anchor point data forwards the positioning data to the server's switch via the network cable. Then, the data is forwarded to the positioning engine software on the host computer through the switch. The positioning data is obtained by analyzing and calculating the data.

[0052] Pre-calibrate both Bluetooth and UWB positioning anchors, configuring Bluetooth anchor transmission parameters such as transmission power, measurement power, transmission interval, and multicast interval. Maintain consistency in these key parameters, and ensure that the transmission interval modulation and the frequency of the UWB anchor's transmission signal are synchronized to keep the positioning system data in sync.

[0053] First, the positioning anchor sends UWB signal data and simultaneously receives positioning signal data transmitted by the Bluetooth anchor. The UWB positioning anchor forwards the positioning data to an intermediate switch, which then forwards the data to the positioning engine software on the server. A pre-written fusion positioning algorithm is embedded into the positioning engine, and the data is processed by the positioning engine to calculate the final positioning coordinates, which are then saved to the database and displayed on the host computer.

[0054] Specifically, the invention comprises one or more primary positioning anchor points and one or more auxiliary positioning anchor points. The primary and auxiliary positioning anchor points respectively collect signal strength under corresponding conditions.

[0055] The positioning system framework provides With several positioning anchor points, the signal strength vector of each point in space can be represented as... ,in Then the first corresponding to the current position point The Bluetooth signal strength of each BLE anchor point. Sufficient signal strength fingerprint data was collected from the area surrounding each spatial anchor point. The positioning signal strength at each location was numbered. During the online phase, the positioning strength of each point was received, matched accordingly, and optimized using algorithms to improve positioning accuracy, achieving signal strength matching with high time precision.

[0056] When the received signal is a UWB signal, the DS-TWR positioning algorithm is used for positioning. The DS-TWR algorithm increases the number of communications to obtain multiple time intervals based on the SS-TWR positioning algorithm, thereby further reducing ranging errors caused by frequency drift.

[0057] When the received signal is a mixture of BLE and UWB, the system automatically switches to the fusion positioning mode. The system's state transition model is defined as follows:

[0058] To merge the coordinates of the location tags, The state vector representing the current position, This represents the position and state vector at the previous moment. During the Kalman filtering process, The process noise of the positioning model is represented by the covariance matrix of the positioning system. .

[0059] Taking two-dimensional space as an example, the state transition matrix In this positioning system, it is represented as

[0060] in the formula This refers to the sampling time interval of the corresponding positioning sensor, also known as the clock interval, used to update the received position within the interval. Therefore, the measurement model of the entire positioning system can be defined as:

[0061] In the formula The current measurement vector is represented in the fusion positioning system as the measurement distance between the positioning sub-anchor point and the positioning master anchor point. for The measurement vector at time, For the observation matrix, for The observation noise matrix at time t.

[0062] This distance can be measured by UWB or by BLE. It is the observation matrix of the fused positioning system, and its expression can be represented by the formula.

[0063] It represents the distance between the location tag receiving distance information and the location anchor point. In the above formula... for The coordinates of the fused positioning tag received by the fused sensor at any given time. Then it means the first The coordinates of each positioning anchor point. (Using...) The two-dimensional space formed by the positioning anchor points The Jacobian determinant, i.e., the observation matrix representing the spatial coordinates of the anchor point. It can be expressed by the formula as follows:

[0064] If the fused positioning system receives ultra-wideband information from 3D positioning, the dimensional extension is correspondingly expanded. In the Kalman filtering process, the received signal state is divided into two stages: the prediction stage and the update stage.

[0065] During the prediction phase, the position state vector Make predictions.

[0066] In the formula, To predict the state vector, for The estimated state vector at time t.

[0067] The state prediction covariance matrix is ​​predicted as follows

[0068] In the formula, To predict the state covariance matrix, for The estimated state covariance matrix at each time step. For the position update stage, since the sensor interval was used as the update point when setting the transition matrix, updates can be performed whenever new measurement vector values ​​are available. The purpose of this stage is to refine the position state vector of the fused positioning system through updates. And the error covariance matrix of the fusion positioning system .

[0069] Simultaneously predict the error covariance matrix of the system.

[0070]

[0071] in System control variables, typically in IMU-assisted systems, are variables such as acceleration and angular velocity.

[0072] The core of the model is calculating the Kalman filter gain, which determines the ratio of the uncertainty of the model's predictions to the uncertainty of the actual measurements.

[0073] In the formula This is the transpose of the Jacobian determinant. The covariance matrix of the observation noise matrix represents the signal loss error or sensor physical error in the fusion positioning system. The updated state is calculated after combining the sensor information received by the fusion system with the prediction stage of the Kalman filter.

[0074] in, The difference in calculated values ​​from the fused positioning system is represented by the measurement vector. Subtract the predicted measurement value The result is obtained by subtraction. Using the formula...

[0075] Update the error covariance matrix. Then update the information vector as follows:

[0076] in For the new information vector, for The state vector at time step 1. The final vector update to determine the estimated state:

[0077] This allows us to determine the target's final location information.

[0078] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any ordinary changes and substitutions made by those skilled in the art within the scope of the technical solution of the present invention should be included within the protection scope of the present invention.

Claims

1. A Bluetooth and UWB fusion positioning method based on Kalman filtering, characterized in that, Includes the following steps: Obtain the distance from the UWB ranging value and the Bluetooth signal after transformation using a distance conversion model, and construct the observation vector; The observation vectors are fused using the Kalman filter algorithm to determine the real-time position of the target. Based on positioning accuracy requirements, the positioning service area is divided into a high-precision positioning area, a normal-precision positioning area, and a connecting area. The UWB positioning algorithm is used in the high-precision positioning area, the RSSI-based Bluetooth spatial fingerprint algorithm is used in the normal-precision positioning area, and the UWB / BLE combined positioning algorithm is used in the connecting area.

2. The Bluetooth and UWB fusion positioning method based on Kalman filtering according to claim 1, characterized in that: To address the differences in observation errors between UWB and Bluetooth ranging values, as well as the ranging errors caused by indoor environmental interference, a UWB / BLE Kalman positioning algorithm model is constructed. This algorithm model reduces the impact of environmental interference on ranging accuracy and improves the reliability of positioning data. Furthermore, it adaptively adjusts the UWB and BLE signal weights based on the real-time status of the positioning system in different environments, optimizes data fusion, reduces signal interference and multipath effects, and ensures positioning accuracy and stability, meeting the requirements for precise target tracking in complex vehicle environments.

3. A Bluetooth and UWB fusion positioning method based on Kalman filtering according to claim 1 or 2, characterized in that: When the positioning tag collects a single Bluetooth signal and there is at least one valid Bluetooth anchor signal source, the positioning is performed using a Bluetooth spatial fingerprint algorithm based on RSSI. When the positioning tag acquires a single UWB signal and there is at least one valid UWB anchor signal source, the DS-TWR, SS-TWR, ToA, TDoA or AoA ultra-wideband positioning algorithm is used for positioning. When the positioning tag simultaneously collects UWB and Bluetooth signals, the two signals are fused and processed. Based on the weighting mechanism of the uncertainty of model prediction and the uncertainty of actual measurement, the positioning results are optimized.

4. The Bluetooth and UWB fusion positioning method based on Kalman filtering according to claim 3, characterized in that: The Bluetooth spatial fingerprint algorithm includes an offline phase and an online phase: During the offline phase, target Bluetooth signal strength is collected to establish a fingerprint database, and KNN, SVM, random forest or particle swarm optimization algorithms are used to optimize positioning accuracy. During the online phase, location is achieved by matching the real-time received Bluetooth signal strength with the offline fingerprint database.

5. The Bluetooth and UWB fusion positioning method based on Kalman filtering according to claim 4, characterized in that: The Kalman filter algorithm is used to jointly process UWB and Bluetooth signals, and a dynamic weight adjustment mechanism is used to achieve accurate target recognition. Real-time positioning and calibration are performed based on the current state of the target being detected, the system error covariance is predicted for the next moment, the Kalman gain is calculated, the optimal state variables of the system are updated, and dynamic tracking is achieved.

6. A Bluetooth and UWB fusion positioning system based on Kalman filtering, characterized in that, include: UWB positioning anchors and Bluetooth positioning anchors are respectively placed on the object to be tested; Location tags are used to collect UWB ranging values ​​and Bluetooth signal strength to construct observation vectors; The main control chip MCU is used to receive and process UWB and Bluetooth positioning information, and uses the Kalman filter algorithm for data fusion processing; NB-IoT module, used for wirelessly forwarding location information; The power management module provides unified power to the UWB module, Bluetooth module, and MCU. The clock circuit is used for calibrating the MCU clock signal to ensure time synchronization accuracy.

7. A Bluetooth and UWB fusion positioning system based on Kalman filtering according to claim 6, characterized in that: The MCU interacts with the UWB communication module and the Bluetooth communication module through the SPI bus interface to achieve synchronous information transmission between the modules; The power management module adopts a low-power design, and isolation circuits are used between modules to prevent electromagnetic interference and ensure stable system operation.

8. A Bluetooth and UWB fusion positioning system based on Kalman filtering according to claim 6 or 7, characterized in that: The positioning system also includes positioning engine software, which is used to receive data sent by positioning anchor points, embed a fusion positioning algorithm, calculate the real-time position coordinates of the target, save the results to the database and upload them to the host computer for display.

9. A Bluetooth and UWB fusion positioning system based on Kalman filtering according to any one of claims 6 to 8, characterized in that: The positioning system enables communication between the UWB anchor point, Bluetooth anchor point and MCU via CAN BUS or LIN bus. The positioning information is sent to the background monitoring platform through the vehicle T-BOX and 4G module to achieve remote real-time monitoring.

Images