Multi-sensor fusion locator positioning method

Abstract

The present invention belongs to the field of positioning and navigation technology, and specifically relates to a multi-sensor fusion locator positioning method, which includes the following steps: first, collecting and synchronizing data from different sensors, and organizing them in a unified time-stamped sequence; then, extracting key positioning signals and environmental information from these synchronized data, and calculating environmental interference characteristics; finally, using environmental information to correct and purify the original positioning signal, thereby improving the accuracy and robustness of positioning. Through this whole set of processes, the influence of various environmental interferences on the positioning results can be effectively reduced, making the positioning more accurate and reliable, especially in complex and changeable indoor and outdoor environments.

Description

Technical Field

[0001] The invention belongs to the technical field of positioning and navigation, and in particular relates to a multi-sensor fusion locator positioning method. Background Art

[0002] In recent years, with the rapid development of mobile computing and wireless communication technologies, location services have played an increasingly important role in people's daily lives. Especially in the scenario of indoor and outdoor environment integration, how to provide an accurate and reliable positioning service has become a technical problem that needs to be solved urgently. Existing positioning technologies mainly rely on a single type of sensor, such as GPS, Wi-Fi or Bluetooth. However, each of these technologies has its own limitations, such as GPS signals perform poorly indoors, and Wi-Fi and Bluetooth signals are easily affected by environmental interference. Therefore, how to effectively combine the data of these sensors to make up for the shortcomings of single sensor positioning has become a hot topic and difficulty in current research. Summary of the invention

[0003] 1. Technical issues to be resolved The present invention mainly aims at the problem of inaccurate positioning of a single sensor in a specific environment in the prior art, and proposes a multi-sensor fusion locator positioning method. Its purpose is to improve the accuracy and reliability of positioning by comprehensively utilizing the data of multiple sensors, especially the positioning service in the indoor and outdoor fusion environment.

[0004] (II) Technical solution To achieve the above object, the present invention provides a multi-sensor fusion locator positioning method, comprising the following steps: Step 1: Collect data from multiple sensors and upload them to the backend server; Step 2: Perform time synchronization on the data collected from multiple sensors on the backend server to ensure that each data point has a unified time tag sequence T={t}; Step 3: extract key signals for positioning from the synchronized data, the key signals including but not limited to GPS coordinates, Wi-Fi signal strength and Bluetooth signal strength as position frames, and the position frame set constitutes a second frame sequence; Step 4: extracting environmental information frames, and forming a third frame sequence with the environmental information frame set; Step 5: Calculate the environmental interference feature based on the environmental information frame in the third frame sequence, and convert and calculate the transformed feature using the position frame in the second frame sequence; Step 6: Determine the temporal proximity of the location frame and the environment information frame through the time tag sequence T={t}, and use the environmental interference features of the adjacent environment information frame to correct the conversion features of the location frame to produce purified positioning features.

[0005] Furthermore, in step 1, the sensor includes a GPS receiver, a Wi-Fi signal detector, and a Bluetooth signal detector.

[0006] Furthermore, in step 2, the step of performing time synchronization processing on the collected data of multiple sensors on the backend server includes: Receive data packets uploaded from each sensor to the backend server; each data packet should contain the sensor ID, timestamp and collected raw data; Convert the timestamps in all data packets to the server's unified time format; Time-align all collected sensor data based on the timestamp of each data packet; The timestamp is adjusted by the clock deviation estimation algorithm; After completing the above steps, each data point is assigned a unified time tag to form a complete time tag sequence T = {t}.

[0007] Furthermore, the specific steps of step 3 include: Filter out positioning-related signals from the synchronized data, including GPS coordinates, Wi-Fi signal strength, and Bluetooth signal strength; Preprocess the selected data; Extract key features for positioning from preprocessed data; Encapsulate the extracted key features into location frames, each of which contains a timestamp and a set of key signals for positioning; All position frames are arranged in time sequence to form a second frame sequence for subsequent processing.

[0008] Further, in step 5, the environmental interference feature is calculated based on the environmental information frame in the third frame sequence, and the transformed feature is obtained by converting and calculating the position frame in the second frame sequence, and the method is: The position frame sequence and environment information frame sequence are defined as L(n,m) and H(n,m), respectively, where n represents the number of frames in the sequence and m represents the dimension of the data points in each frame; L(i,j) and H(i,j) refer to the position information value and environment information value on the j-dimensional data of the i-th frame, respectively; For each frame H in the environmental information frame sequence, an algorithm is used to calculate the environmental impact factor of the frame; For each frame L in the position frame sequence, based on the associated environmental influencing factors, the adjusted position feature is obtained through the calculation model. .

[0009] Furthermore, in step 6, the method for determining the temporal proximity of the position frame and the environment information frame through the time tag sequence T={t} is: For each location frame L, find the environment information frame H closest to its acquisition time in the time tag sequence T={t} to ensure that the two are close in time; Select the environment information frame H that is closest to the position frame L in time, and use the environment impact factor of the environment information frame H to calculate the original position feature of the position frame L make corrections; Each position frame L is corrected by environmental factors to obtain accurate position information .

[0010] Furthermore, the step of using an algorithm to calculate the environmental impact factor of the frame includes: Assume that each frame H in the environmental information frame sequence has m dimensions of data, which respectively represent different aspects of the environment information; represent the environmental information frame sequence as a matrix H, where each row represents a frame and each column represents a dimension of data; The environmental impact factor of the i-th frame is calculated and expressed as a vector H(i), which is obtained by fitting the following linear model: ; in, is the vector represented by the environmental impact factor of the i-th frame to be calculated, It is the data of each dimension of the environment information frame. are the parameters of the model, is the error term.

[0011] Furthermore, the adjusted position feature is obtained through the following calculation model: : ; in, is the environmental impact factor of the i-th position frame; is the original position feature of the i-th position frame; is the adjusted position feature.

[0012] (III) Beneficial effects Compared with the prior art, the multi-sensor fusion locator positioning method provided by the present invention solves the technical problem through the following steps: first, collect and synchronize data from different sensors and organize them in a unified time-stamped sequence; then, extract key positioning signals and environmental information from these synchronized data, and calculate environmental interference characteristics; finally, use environmental information to correct and purify the original positioning signal, thereby improving the accuracy and robustness of positioning. Through this whole set of processes, the influence of various environmental interferences on the positioning results can be effectively reduced, making the positioning more accurate and reliable, especially in complex and changeable indoor and outdoor environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 This is a flow chart of a multi-sensor fusion locator positioning method disclosed in this application.

[0014] Figure 2 This is a time synchronization processing flow chart disclosed in this application.

[0015] Figure 3 This is a framework diagram of a dynamic position adjustment model disclosed in this application. DETAILED DESCRIPTION

[0016] The present invention will be described in detail below in conjunction with the accompanying drawings, and the technical solutions in the embodiments of the present invention will be clearly and completely described. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] This embodiment provides a multi-sensor fusion locator positioning method, which aims to improve the accuracy and robustness of positioning by comprehensively utilizing data from multiple sensors. The following describes the implementation of the present invention in detail. Figure 1-Figure 3 : Step 1: Data Collection First, various types of sensors are installed on the device that determines the positioning target, including GPS receivers, Wi-Fi signal detectors, and Bluetooth signal detectors. As the positioning target moves, these sensors will continue to collect relevant data about the surrounding environment, including but not limited to GPS coordinates, Wi-Fi signal strength, and Bluetooth signal strength. Subsequently, the collected data will be uploaded to the backend server for further processing.

[0018] Step 2: Time synchronization After the backend server receives data uploaded by multiple sensors, the first task is to synchronize the data. Specifically, the server will assign a unified time tag sequence T={t} to each data point based on the time when the data was received, ensuring that the data from all sensors can correspond in time, laying the foundation for subsequent data processing.

[0019] Step 3: Extraction of key signals After completing the time synchronization process, the system will extract key signals for positioning from the synchronized data. These key signals include GPS coordinates, Wi-Fi signal strength, and Bluetooth signal strength, which are organized into location frames and constitute the second frame sequence.

[0020] Step 4: Environmental information extraction In addition to location information, environmental information also has an important impact on the positioning results. Therefore, the method also includes the step of extracting environmental information frames, such as the layout of surrounding buildings, indoor and outdoor environmental characteristics, etc. These environmental information frames are aggregated into a third frame sequence for subsequent data processing and analysis.

[0021] Step 5: Calculation of environmental interference characteristics Based on the environmental information frame in the third frame sequence, the system calculates the environmental interference features. Then, combined with the position frame in the second frame sequence, these environmental interference features are transformed and calculated to obtain their transformation features. This step is the key to identifying and correcting positioning deviations.

[0022] Step 6: Positioning feature purification Finally, the system determines the temporal proximity of the location frame and the environmental information frame through the time tag sequence T={t}. The environmental interference features of the adjacent environmental information frame are used to correct the conversion features of the location frame to produce purified positioning features. The purpose of this step is to remove or reduce the impact of environmental factors on the positioning results and improve the accuracy and reliability of positioning.

[0023] Through the above steps, the multi-sensor fusion positioning method provided by the present invention can effectively process and analyze data from different sensors to achieve more accurate and stable positioning results, and is particularly suitable for scenarios requiring high-precision positioning services.

[0024] In step 1, all the pre-determined sensors are started to start data collection. In this phase, each sensor collects environmental data independently according to its configuration; a timestamp is attached to each collected data point, which can be provided by the sensor itself (if it has the ability to provide accurate time) or added by the collection system immediately after the data is collected; the collected data is sent to the back-end server through the network, on which a data receiving service is established to receive data from various sensors. This service is responsible for parsing the received data and performing preliminary formatting processing to facilitate subsequent synchronization and analysis. A preliminary quality assessment is performed on the received data to exclude obvious erroneous data or abnormal data to improve the accuracy and efficiency of subsequent processing. This includes basic checks on data values, such as whether the GPS coordinates are within a reasonable range and whether the signal strength is within the expected range.

[0025] In step 2, if Figure 2 First, the backend server receives data packets from each sensor. Each data packet contains the sensor's identifier, timestamp, and collected raw data. Subsequently, the server converts the timestamps in all data packets into a unified time format and performs time alignment on all collected data according to the timestamp of each data packet. In order to further improve the accuracy of time synchronization, the system uses a clock deviation estimation algorithm to adjust the timestamp, and finally assigns a unified time tag sequence T={t} to each data point, forming a complete time tag sequence for subsequent data processing.

[0026] After the time synchronization process is completed, the system begins to filter out signals related to positioning, including GPS coordinates, Wi-Fi signal strength, and Bluetooth signal strength. The selected data is preprocessed, such as filtering and denoising, to improve data quality. After preprocessing, key features for positioning are extracted from the data and encapsulated into location frames. Each location frame contains a timestamp and a set of key signals for positioning. In this way, all location frames are arranged in chronological order to form a second frame sequence for subsequent analysis and processing.

[0027] In order to solve the positioning interference problem caused by environmental changes, this method adopts a calculation model in step 5 to calculate the environmental interference characteristics based on the environmental information frame in the third frame sequence, and performs corresponding conversion calculations on the position frame in the second frame sequence. Specifically, Assume that each frame H in the environmental information frame sequence has m dimensions of data, which respectively represent different aspects of the environment. The environmental information frame sequence is represented as a matrix H, where each row represents a frame and each column represents a dimension of data.

[0028] Assuming that we want to calculate the environmental impact factor of the i-th frame, we can express it as a vector H(i), which can be obtained by linear regression fitting. The linear model is: in, is the vector represented by the environmental impact factor of the i-th frame to be calculated, It is the data of each dimension of the environment information frame. are the parameters of the model, is the error term.

[0029] Use the least squares method or other regression algorithms to fit the above model and get the parameters , and then obtain the environmental impact factors.

[0030] In this embodiment, the adjusted position feature is obtained by calculating the model , the specific steps are as follows: Suppose there is a sequence of position frames L(n,m), where each frame L(i,j) contains the dimension j of position information. The above steps have obtained the environmental impact factor of each position frame, and the following linear weighted average model is used to adjust the position features: in, is the environmental impact factor of the i-th position frame; is the original position feature of the i-th position frame; is the adjusted position feature.

[0031] The above model performs weighted averaging on each location frame, where the weight is determined by the environmental factors, so the adjusted location features can be obtained. .

[0032] In step 6, the system determines the temporal proximity of the location frame and the environmental information frame through the established time tag sequence T={t}. The environmental information frame H that is closest in time to each location frame L is selected, and the environmental impact factor of the environmental information frame H is used to correct the original location feature of the location frame L to obtain more accurate location information. This process further improves the positioning accuracy.

[0033] Through the implementation steps described in detail above, the present invention realizes a method for improving positioning accuracy and robustness through multi-sensor data fusion and environmental information consideration, which is suitable for precise positioning needs in a variety of complex environments.

[0034] To better understand this multi-sensor fusion positioning method and its various steps, a specific example is used to illustrate. Imagine a project whose goal is to accurately locate the location of customers in a large shopping mall to provide navigation services, push relevant offers, etc. The shopping mall is covered by GPS, Wi-Fi, and Bluetooth signals.

[0035] First, through the application installed on the customer's smartphone, data from the phone's GPS receiver, Wi-Fi signal detector, and Bluetooth signal detector are collected and uploaded to the backend server. For example, at a certain moment, the phone may capture the following data: GPS coordinates (30.2672°N, 97.7431°W), the signal strength of three nearby Wi-Fi access points (-80dBm, -70dBm, -90dBm), and the signal strength of two Bluetooth devices (-75dBm, -65dBm).

[0036] After receiving data from multiple sensors, the backend server synchronizes them. Assuming the timestamp of GPS data is 08:00:01.123, Wi-Fi data is 08:00:01.125, and Bluetooth data is 08:00:01.124, the server will align them to ensure that each data point has a unified time stamp, such as 08:00:01.

[0037] From the synchronized data, select the signals related to positioning: GPS coordinates, Wi-Fi and Bluetooth signal strength. Then, preprocess these signals (such as filtering, denoising, etc.) to extract the key features used for positioning, and encapsulate these features into location frames. Each location frame contains a timestamp and a set of key signals used for positioning.

[0038] At the same time, the system also needs to extract environmental information frames, such as crowd density information in different areas of the shopping mall, electromagnetic interference information, etc., which will also affect the accuracy of positioning.

[0039] Next, the environmental interference characteristics are calculated based on the environmental information frame. For example, if the Wi-Fi signal is found to be abnormally strong in a certain area, it may be because there is a new Wi-Fi hotspot there that is not registered on the map, and this information should be taken into account.

[0040] Through the time-stamp sequence, the temporal proximity of the location frame and the environmental information frame is determined, and then the environmental interference features of the adjacent environmental information frame are used to correct the conversion features of the location frame to produce a purified positioning feature. For example, if the environmental information shows that the GPS signal at the current location is strongly interfered, the GPS positioning result can be adjusted by relying on Wi-Fi and Bluetooth signals.

[0041] Through the above steps, this multi-sensor fusion positioning method can effectively improve the accuracy and robustness of positioning. In practical applications, this method can be adjusted and optimized according to specific environments and needs to achieve the best positioning effect.

[0042] The above embodiments are only used to illustrate the technical solution of the present application and are not intended to limit it. Although the present application has been described in detail with reference to the preferred embodiments, a person skilled in the art should understand that the technical solution of the present application may be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present application.

Claims

1. A multi-sensor fusion locator positioning method, characterized in that: The steps include: Step 1: Collect data from multiple sensors and upload them to the backend server; Step 2: Perform time synchronization on the data collected from multiple sensors on the backend server to ensure that each data point has a unified time tag sequence T={t}; Step 3: extract key signals for positioning from the synchronized data, the key signals including but not limited to GPS coordinates, Wi-Fi signal strength and Bluetooth signal strength as position frames, and the position frame set constitutes a second frame sequence; Step 4: extracting environmental information frames, and forming a third frame sequence with the environmental information frame set; Step 5: Calculate the environmental interference feature based on the environmental information frame in the third frame sequence, and convert and calculate the transformed feature using the position frame in the second frame sequence; Step 6: Determine the temporal proximity of the location frame and the environment information frame through the time tag sequence T={t}, use the environmental interference features of the adjacent environment information frame to correct the conversion features of the location frame, and produce purified positioning features.

2. A multi-sensor fusion locator positioning method according to claim 1, characterized in that: In step 1, the sensors include a GPS receiver, a Wi-Fi signal detector, and a Bluetooth signal detector.

3. A multi-sensor fusion locator positioning method according to claim 1, characterized in that: In step 2, the step of performing time synchronization processing on the collected data of multiple sensors on the backend server includes: Receive data packets uploaded from each sensor to the backend server; each data packet should contain the sensor ID, timestamp and collected raw data; Convert the timestamps in all data packets to the server's unified time format; Time-align all collected sensor data based on the timestamp of each data packet; The timestamp is adjusted by the clock deviation estimation algorithm; After completing the above steps, each data point is assigned a unified time tag to form a complete time tag sequence T = {t}.

4. A multi-sensor fusion locator positioning method according to claim 1, characterized in that: The specific steps of step 3 include: Filter out positioning-related signals from the synchronized data, including GPS coordinates, Wi-Fi signal strength, and Bluetooth signal strength; Preprocess the selected data; Extract key features for positioning from preprocessed data; Encapsulate the extracted key features into location frames, each of which contains a timestamp and a set of key signals for positioning; All position frames are arranged in time sequence to form a second frame sequence for subsequent processing.

5. The multi-sensor fusion locator positioning method according to claim 1, characterized in that: In step 5, the environmental interference feature is calculated based on the environmental information frame in the third frame sequence, and the transformed feature is obtained by converting and calculating the position frame in the second frame sequence. for: The position frame sequence and environment information frame sequence are defined as L(n,m) and H(n,m), respectively, where n represents the number of frames in the sequence and m represents the dimension of the data points in each frame; L(i,j) and H(i,j) refer to the position information value and environment information value on the j-dimensional data of the i-th frame, respectively; For each frame H in the environmental information frame sequence, an algorithm is used to calculate the environmental impact factor of the frame; For each frame L in the position frame sequence, based on the associated environmental influencing factors, the adjusted position feature is obtained through the calculation model. .

6. A multi-sensor fusion locator positioning method according to claim 5, characterized in that: In step 6, the method for determining the temporal proximity of the position frame and the environment information frame through the time tag sequence T={t} is: For each location frame L, find the environment information frame H closest to its acquisition time in the time tag sequence T={t} to ensure that the two are close in time; Select the environment information frame H that is closest to the position frame L in time, and use the environment impact factor of the environment information frame H to calculate the original position feature of the position frame L make corrections; Each position frame L is corrected by environmental factors to obtain accurate position information .

7. A multi-sensor fusion locator positioning method according to claim 5, characterized in that: The steps of using an algorithm to calculate the environmental impact factor of the frame include: Assume that each frame H in the environmental information frame sequence has m dimensions of data, which respectively represent different aspects of the environment information; represent the environmental information frame sequence as a matrix H, where each row represents a frame and each column represents a dimension of data; The environmental impact factor of the i-th frame is calculated and expressed as a vector H(i), which is obtained by fitting the following linear model: ； in, is the vector represented by the environmental impact factor of the i-th frame to be calculated, It is the data of each dimension of the environment information frame. are the parameters of the model, is the error term.

8. The multi-sensor fusion locator positioning method according to claim 5, characterized in that: The adjusted position features are obtained through the following calculation model : ； in, is the environmental impact factor of the i-th position frame; is the original position feature of the i-th position frame; is the adjusted position feature.

Images