Multi-sensor fusion-based indoor and outdoor integrated positioning method and system

Abstract

The present invention relates to a multi-sensor fusion-based indoor and outdoor integrated positioning method and system. The method comprises: acquiring, by means of a receiving end of a current device, positioning signals of an indoor positioning system and an outdoor positioning system, and executing a network decision process to select an optimal positioning network, the network decision process comprising: on the basis of the quality of the positioning signals of the indoor positioning system and the outdoor positioning system, switching and selecting between positioning schemes; when signals of both the indoor positioning system and the outdoor positioning system are lost, on the basis of acceleration information and angular velocity information of an IMU reference system, using a pre-trained neural network model to predict a position change of the current device; and, when the position enters an indoor and outdoor transition area, performing fusion positioning on the basis of the indoor positioning system and the outdoor positioning system. Compared with the prior art, the present invention has the advantages of improving the positioning accuracy and robustness, achieving seamless indoor and outdoor navigation, and being capable of handling situations where signals are completely lost.

Description

An Indoor and Outdoor Integrated Positioning Method and System Based on Multi-Sensor Fusion Technical Field

[0001] This invention relates to the field of positioning technology, and in particular to an integrated indoor and outdoor positioning method and system based on multi-sensor fusion. Background Technology

[0002] In modern society, navigation and positioning technologies have become an indispensable part of daily life and industrial production. Currently, outdoor navigation systems primarily rely on GNSS (Global Navigation Satellite System). GNSS technology performs well in outdoor environments, providing high positioning accuracy. However, indoors, underground, or in environments with severe signal obstruction, GNSS signals are significantly affected, leading to a substantial decrease in positioning accuracy. To address this, indoor positioning technologies such as Wi-Fi, Bluetooth, and UWB are widely used. However, indoor positioning signal coverage is limited, making it difficult to apply in large-scale outdoor environments. To overcome the limitations of indoor and outdoor positioning technologies, existing technologies typically employ multi-sensor fusion methods, such as fusing multiple indoor and outdoor positioning data, to improve positioning accuracy and system robustness. However, existing multi-sensor fusion methods have the following drawbacks:

[0003] Inconsistent positioning accuracy: When moving from outdoors to indoors or from indoors to outdoors, the difference in the characteristics of indoor and outdoor positioning signals can cause sudden changes in positioning accuracy, affecting the continuity and reliability of the navigation system.

[0004] Lack of effective network switching mechanism: Existing multi-sensor fusion methods lack an effective network switching mechanism when dealing with positioning switching in different environments, which can easily lead to the ping-pong effect and further aggravate positioning errors.

[0005] Inadequate signal loss handling: When both indoor and outdoor positioning signals are unavailable, existing technologies lack effective countermeasures, resulting in the system being unable to provide reliable location information support. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the prior art, such as inconsistent positioning accuracy, lack of effective network switching mechanism and insufficient signal loss handling, and to provide an integrated indoor and outdoor positioning method and system based on multi-sensor fusion.

[0007] The objective of this invention can be achieved through the following technical solutions:

[0008] An integrated indoor-outdoor positioning method based on multi-sensor fusion is used in indoor and outdoor environments with an indoor positioning system and an outdoor positioning system. The outdoor positioning system includes a GNSS system and an IMU reference system. The method includes the following steps:

[0009] The device collects positioning signals from indoor and outdoor positioning systems through its receiver and performs a network decision process to select the best positioning network.

[0010] The online judgment process includes:

[0011] Based on the positioning signal quality of the indoor and outdoor positioning systems, switch between the positioning scheme based on the indoor positioning system, the fusion positioning scheme based on the indoor and outdoor positioning systems, and the positioning scheme based on the outdoor positioning system.

[0012] When both indoor and outdoor positioning system signals are lost, a pre-trained neural network model is used to predict the current position change of the device based on the acceleration and angular velocity information of the IMU reference system.

[0013] When the current location of the device enters the boundary area between indoor and outdoor environments, it performs fusion positioning based on the indoor positioning system and the outdoor positioning system.

[0014] Furthermore, the specific process of switching between the indoor and outdoor positioning systems based on their signal quality includes:

[0015] If the number of available positioning satellites in the GNSS system is less than the preset minimum threshold, the HDOP value is greater than the preset maximum HDOP value, and the RSSI value of the outdoor positioning system base station is less than the lower limit and greater than or equal to the preset optimal threshold, then positioning is performed based on the indoor positioning system.

[0016] When the number of available positioning satellites in the GNSS system is between the preset minimum and maximum thresholds, the HDOP value is less than the maximum HDOP value, and the RSSI value of the base station in the indoor positioning system is less than the lower limit and greater than or equal to the preset optimal threshold, then fusion positioning is performed based on the indoor positioning system and the outdoor positioning system.

[0017] If the number of available positioning satellites in the GNSS system is greater than the maximum number threshold, the HDOP value is less than the maximum HDOP value, and the RSSI value of the indoor positioning system base station is less than the lower limit number which is less than the preset optimal threshold, then the outdoor positioning system is selected for positioning.

[0018] Furthermore, during the process of switching the selected positioning scheme, a switching count is also performed. If the number of switching counts exceeds a preset switching threshold, the positioning scheme is reset and the switching count is cleared to zero.

[0019] Furthermore, the training process of the neural network model includes:

[0020] The acceleration and angular velocity information collected by the IMU reference system are used as input to the neural network model. The position information obtained after switching the positioning scheme according to the positioning signal quality of the indoor and outdoor positioning systems is used as the output of the neural network model. A training dataset is constructed to train the neural network model.

[0021] Furthermore, the neural network model is a BP neural network.

[0022] Furthermore, in the indoor-outdoor boundary area, a federal Kalman filter is used to perform fusion positioning based on the indoor positioning system and the outdoor positioning system.

[0023] Furthermore, the federated Kalman filter includes a first sub-filter, a second sub-filter, and a main filter. The first sub-filter performs filtering and fusion based on the positioning signals from the GNSS system and the IMU reference system to obtain a first positioning result, which is then output to the main filter. The second sub-filter performs filtering and fusion based on the positioning signals from the indoor positioning system and the IMU reference system to obtain a second positioning result, which is then output to the main filter.

[0024] The main filter performs a weighted average fusion of the positioning signal from the IMU reference system, the first positioning result, and the second positioning result, and generates a first adjustment parameter and a second adjustment parameter. The first adjustment parameter is fed back to the first sub-filter for parameter optimization, and the second adjustment parameter is fed back to the second sub-filter for parameter optimization.

[0025] An integrated indoor and outdoor positioning system based on multi-sensor fusion includes an indoor positioning system, an outdoor positioning system, and a main processor unit. The outdoor positioning system includes a GNSS system and an IMU reference system.

[0026] The main processor unit is used to execute the steps of an indoor-outdoor integrated positioning method based on multi-sensor fusion as described above.

[0027] Furthermore, the indoor positioning system is a Bluetooth AOA device used to obtain directional information of relative position and realize positioning in the indoor environment.

[0028] Furthermore, the GNSS system includes a GNSS receiver for acquiring global navigation satellite system signals to achieve positioning in outdoor environments;

[0029] The IMU reference system includes accelerometers and gyroscopes to provide displacement and attitude change information when the global navigation satellite system signal is unstable or lost.

[0030] Compared with the prior art, the present invention has the following advantages:

[0031] (1) Improve positioning accuracy and robustness: In the positioning process, the present invention first judges the positioning signal quality of the indoor positioning system and the outdoor positioning system. After judging based on parameters such as the number of available GNSS satellite signals, the GNSS horizontal accuracy factor HDOP value, and the RSSI value of the indoor base station, the optimal positioning network is selected to ensure the quality of the positioning signal.

[0032] Furthermore, it is proposed to use a federated Kalman filter to fuse the positioning signals of indoor and outdoor positioning systems. By utilizing the multi-sensor fusion capability of the federated Kalman filter, high-precision location determination can be achieved in indoor and outdoor boundary areas. This effectively avoids the problem of decreased positioning accuracy caused by the performance differences of a single technology in different environments and improves the robustness of the system.

[0033] (2) Achieve seamless indoor and outdoor navigation: This invention proposes a positioning scheme that integrates indoor and outdoor positioning systems, which can achieve a smooth transition of positioning data when switching between indoor and outdoor environments, avoid the problem of sudden changes in accuracy, achieve seamless indoor and outdoor navigation, and improve the user experience.

[0034] (3) Dealing with the situation of complete signal loss: When all positioning signals are lost, the simulation experiment scheme proposed in this invention, which uses a BP neural network combined with historical data for location prediction, can maintain the continuity of navigation services and solve the problem of location prediction when the signal is completely lost in the existing technology. Attached Figure Description

[0035] Figure 1 is a schematic diagram of the overall process of an indoor-outdoor integrated positioning method based on multi-sensor fusion provided in an embodiment of the present invention;

[0036] Figure 2 is a schematic diagram of a network handover decision process provided in an embodiment of the present invention;

[0037] Figure 3 is a schematic diagram of a data fusion process of a federated Kalman filter provided in an embodiment of the present invention. Detailed Implementation

[0038] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0039] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0040] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0041] In the description of this invention, it should be noted that the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "a plurality of" means two or more, unless otherwise explicitly specified.

[0042] Example 1

[0043] As shown in Figure 1, this embodiment provides an integrated indoor and outdoor positioning method based on multi-sensor fusion, which is used in indoor and outdoor environments with indoor positioning systems and outdoor positioning systems. The outdoor positioning system includes a GNSS system and an IMU reference system. The method includes the following steps:

[0044] The device collects positioning signals from indoor and outdoor positioning systems through its receiver and performs a network decision process to select the best positioning network.

[0045] The online judgment process includes:

[0046] Based on the positioning signal quality of the indoor and outdoor positioning systems, switch between the positioning scheme based on the indoor positioning system, the fusion positioning scheme based on the indoor and outdoor positioning systems, and the positioning scheme based on the outdoor positioning system.

[0047] When both indoor and outdoor positioning system signals are lost, a pre-trained neural network model is used to predict the current position change of the device based on the acceleration and angular velocity information of the IMU reference system.

[0048] When the current location of the device enters the boundary area between indoor and outdoor environments, it performs fusion positioning based on the indoor positioning system and the outdoor positioning system.

[0049] Essentially, this solution divides the indoor and outdoor positioning process into a network switching phase, a positioning prediction phase for signal loss, and a phase for handling the boundary area between indoor and outdoor areas separately, which will be described in detail below.

[0050] 1. Network handover phase

[0051] During the network handover phase, the network handover problem is modeled as a multi-attribute decision-making problem, encompassing network handover discovery, network selection decision, and elimination of the ping-pong effect. It comprehensively considers parameters such as the number of available GNSS satellite signals, the GNSS horizontal accuracy factor (HDOP) value, and the RSSI value of indoor base stations to determine the current positioning status of the device. When the receiver simultaneously receives positioning signals from two heterogeneous networks, a network decision process is triggered to select the optimal positioning network.

[0052] The specific judgment process is as follows:

[0053] If the number of available positioning satellites in the GNSS system is less than the preset minimum threshold, the HDOP value is greater than the preset maximum HDOP value, and the RSSI value of the outdoor positioning system base station is less than the lower limit and greater than or equal to the preset optimal threshold, then positioning is performed based on the indoor positioning system.

[0054] When the number of available positioning satellites in the GNSS system is between the preset minimum and maximum thresholds, the HDOP value is less than the maximum HDOP value, and the RSSI value of the base station in the indoor positioning system is less than the lower limit and greater than or equal to the preset optimal threshold, then fusion positioning is performed based on the indoor and outdoor positioning systems.

[0055] If the number of available positioning satellites in the GNSS system is greater than the maximum threshold, the HDOP value is less than the maximum HDOP value, and the RSSI value of the indoor positioning system base station is less than the lower limit and less than the preset optimal threshold, then the outdoor positioning system will be selected for positioning.

[0056] If none of the above conditions are met, the default mode will be used for positioning.

[0057] Preferably, as shown in Figure 2, a switching count is also performed during the switching selection of the positioning scheme. If the number of switching counts is greater than the preset switching threshold, the positioning scheme is reset, the default mode is used for positioning, and the switching count is cleared to zero, thereby realizing the cyclic switching process.

[0058] 2. Location prediction in signal loss situations

[0059] The neural network model mentioned above can be a BP neural network or other network models, and is not limited to a single one.

[0060] This embodiment proposes a positioning prediction method based on a backpropagation (BP) neural network to predict the location using historical data when both GNSS and indoor base station signals are lost, thus ensuring system continuity. When the tag meets the aforementioned positioning conditions, the neural network learns the motion pattern of the device being located from historical data. The acceleration and angular velocity information of the IMU are used as inputs to the neural network model, and the location information calculated from indoor and outdoor fusion positioning is used as the output to train the model. When the positioning signal is interrupted, the neural network predicts the device's position change, ensuring the stability and continuity of the location information.

[0061] 3. Indoor and Diplomatic Boundary Areas

[0062] In indoor-outdoor boundary areas, a Federal Kalman Filter (FKF) is used for fusion positioning based on indoor and outdoor positioning systems. FKF effectively combines data from multiple sensors, fully leveraging the advantages of different technologies to improve positioning accuracy and system robustness. FKF optimizes the combination of information from multiple sensors through weighted averaging, thereby achieving a smooth transition between indoor and outdoor positioning data and avoiding sudden changes in accuracy. The parameter settings of FKF need to be adjusted according to the actual application scenario to ensure optimal filtering performance.

[0063] Specifically, as shown in Figure 3, the federated Kalman filter includes a first sub-filter, a second sub-filter, and a main filter. The first sub-filter performs filtering and fusion based on the positioning signals from the GNSS system and the IMU reference system to obtain a first positioning result, which is then output to the main filter. The second sub-filter performs filtering and fusion based on the positioning signals from the indoor positioning system and the IMU reference system to obtain a second positioning result, which is then output to the main filter.

[0064] The main filter performs weighted averaging fusion based on the positioning signal from the IMU reference system, the first positioning result, and the second positioning result, and generates a first adjustment parameter and a second adjustment parameter. The first adjustment parameter is fed back to the first sub-filter for parameter optimization, and the second adjustment parameter is fed back to the second sub-filter for parameter optimization.

[0065] Example 2

[0066] This embodiment provides an integrated indoor and outdoor positioning system based on multi-sensor fusion, including an indoor positioning system, an outdoor positioning system, and a main processor unit. The outdoor positioning system includes a GNSS system and an IMU reference system.

[0067] The main processor unit is used to execute the steps of an indoor-outdoor integrated positioning method based on multi-sensor fusion, as described in Example 1.

[0068] Specifically, the indoor positioning system is a Bluetooth AOA device, the GNSS system includes a GNSS receiver, and the IMU reference system includes an accelerometer and a gyroscope;

[0069] The main processor unit (MCU / MPU) is responsible for processing data from various sensors, executing fusion algorithms, processing the raw positioning information, and outputting the final positioning result. The main processor receives raw data from various sensors and applies fusion algorithms to improve the overall positioning accuracy and robustness.

[0070] GNSS receiver: Used to acquire Global Navigation Satellite System (GNSS) signals and provide global coordinate information. GNSS receivers can provide high-precision position data in open environments, providing fundamental position information for the overall fusion system.

[0071] Inertial Measurement Unit (IMU): An inertial measurement unit typically contains accelerometers and gyroscopes, used to provide short-term displacement and attitude change information when GNSS signals are unstable or lost. IMU data is fused to compensate for the deficiencies of GNSS in obstructed environments.

[0072] Bluetooth AOA devices: Bluetooth AOA devices are used to acquire directional information about a relative position, which is typically helpful for accurate short-range positioning in indoor environments. By using AOA information from multiple base stations, the system can perform multilateral positioning, thereby improving positioning capabilities in areas where GNSS signals cannot cover.

[0073] Data acquired from the various navigation technologies mentioned above is input into the main processor unit. The main processor unit determines the current positioning status based on the received base station data. When in outdoor positioning mode, a GNSS / IMU fusion positioning scheme is used; when in indoor-outdoor transition mode, the main processor unit uses FKF (Fixed Kinematic Filter) to optimize and combine GNSS / AOA / IMU information, thus achieving a smooth transition between indoor and outdoor positioning data and avoiding sudden changes in accuracy. The FKF parameter settings need to be adjusted according to the actual application scenario to ensure optimal filtering effect. When all positioning signals are lost, a BP neural network combined with historical data is used for location prediction. The BP neural network needs to be trained to predict future locations based on historical location data. Historical data from the positioning process is used during training.

[0074] It should be noted that the specific content and beneficial effects of the system in this application can be found in the above method embodiments, and will not be repeated here.

[0075] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. An indoor-outdoor integrated positioning method based on multi-sensor fusion, characterized in that, For use in indoor and outdoor environments with indoor and outdoor positioning systems, wherein the outdoor positioning system includes a GNSS system and an IMU reference system, the method includes the following steps: The device collects positioning signals from indoor and outdoor positioning systems through its receiver and performs a network decision process to select the best positioning network. The online judgment process includes: Based on the positioning signal quality of the indoor and outdoor positioning systems, switch between the positioning scheme based on the indoor positioning system, the fusion positioning scheme based on the indoor and outdoor positioning systems, and the positioning scheme based on the outdoor positioning system. When both indoor and outdoor positioning system signals are lost, a pre-trained neural network model is used to predict the current position change of the device based on the acceleration and angular velocity information of the IMU reference system. When the current location of the device enters the boundary area between indoor and outdoor environments, it performs fusion positioning based on the indoor positioning system and the outdoor positioning system. 2.The indoor-outdoor integrated positioning method based on multi-sensor fusion according to claim 1, characterized in that, The specific process of switching between positioning schemes based on the positioning signal quality of indoor and outdoor positioning systems includes: If the number of available positioning satellites in the GNSS system is less than the preset minimum threshold, the HDOP value is greater than the preset maximum HDOP value, and the RSSI value of the outdoor positioning system base station is less than the lower limit and greater than or equal to the preset optimal threshold, then positioning is performed based on the indoor positioning system. When the number of available positioning satellites in the GNSS system is between the preset minimum and maximum thresholds, the HDOP value is less than the maximum HDOP value, and the RSSI value of the base station in the indoor positioning system is less than the lower limit and greater than or equal to the preset optimal threshold, then fusion positioning is performed based on the indoor positioning system and the outdoor positioning system. If the number of available positioning satellites in the GNSS system is greater than the maximum number threshold, the HDOP value is less than the maximum HDOP value, and the RSSI value of the indoor positioning system base station is less than the lower limit number which is less than the preset optimal threshold, then the outdoor positioning system is selected for positioning. 3.The indoor-outdoor integrated positioning method based on multi-sensor fusion according to claim 2, characterized in that, During the process of switching the selected positioning scheme, a switching count is also performed. If the number of switching counts exceeds a preset switching threshold, the positioning scheme is reset and the switching count is cleared to zero.

4. The indoor-outdoor integrated positioning method based on multi-sensor fusion according to claim 1, characterized in that, The training process of the neural network model includes: The acceleration and angular velocity information collected by the IMU reference system are used as input to the neural network model. The position information obtained after switching the positioning scheme according to the positioning signal quality of the indoor and outdoor positioning systems is used as the output of the neural network model. A training dataset is constructed to train the neural network model.

5. The indoor-outdoor integrated positioning method based on multi-sensor fusion according to claim 1, characterized in that, The neural network model is a BP neural network.

6. The indoor-outdoor integrated positioning method based on multi-sensor fusion according to claim 1, characterized in that, The method employs a federated Kalman filter to perform fusion positioning based on indoor and outdoor positioning systems.

7. The indoor-outdoor integrated positioning method based on multi-sensor fusion according to claim 6, characterized in that, The federal Kalman filter includes a first sub-filter, a second sub-filter, and a main filter. The first sub-filter performs filtering and fusion based on the positioning signals from the GNSS system and the IMU reference system to obtain a first positioning result, which is then output to the main filter. The second sub-filter performs filtering and fusion on the positioning signals from the indoor positioning system and the IMU reference system to obtain a second positioning result, which is then output to the main filter. The main filter performs a weighted average fusion of the positioning signal from the IMU reference system, the first positioning result, and the second positioning result, and generates a first adjustment parameter and a second adjustment parameter. The first adjustment parameter is fed back to the first sub-filter for parameter optimization, and the second adjustment parameter is fed back to the second sub-filter for parameter optimization.

8. An indoor-outdoor integrated positioning system based on multi-sensor fusion, characterized in that, It includes an indoor positioning system, an outdoor positioning system, and a main processor unit, wherein the outdoor positioning system includes a GNSS system and an IMU reference system; The main processor unit is used to execute the steps of the indoor and outdoor integrated positioning method based on multi-sensor fusion as described in any one of claims 1-7.

9. The indoor-outdoor integrated positioning system based on multi-sensor fusion according to claim 8, characterized in that, The indoor positioning system is a Bluetooth AOA device used to obtain directional information of relative position and realize positioning in the indoor environment.

10. The indoor-outdoor integrated positioning system based on multi-sensor fusion according to claim 8, characterized in that, The GNSS system includes a GNSS receiver, used to acquire global navigation satellite system signals to achieve positioning in outdoor environments; The IMU reference system includes accelerometers and gyroscopes to provide displacement and attitude change information when the global navigation satellite system signal is unstable or lost.

Images