Positioning method of mobile device, program product and mobile device

Abstract

The embodiment of the invention provides a positioning method of a mobile device, a program product and the mobile device. For movable equipment driven by a wheel type chassis, the course angle direction of the movable equipment is consistent with the speed direction of the movable equipment, so that mileage data and TDOA measurement data acquired in the movement process of the movable equipment can be utilized; the method comprises the following steps: constructing an optimization model which takes the course angle of the mobile equipment as an optimization variable and minimizes the residual sum of a measured value and a predicted value of a target parameter related to TDOA as an optimization target, solving to obtain the course angle, and then determining the position information of the mobile equipment based on the course angle and TDOA measurement data. For example, the position information can be comprehensively determined by combining a course angle, motion state data acquired by an inertial measurement unit of the movable equipment, air pressure data acquired by an altitude barometer and mileage data acquired by a wheel type odometer. Therefore, under the condition that the number of the base stations is small (for example, only two base stations exist), course angle constraint is provided, and accurate positioning is achieved.

Description

Technical Field

[0001] This application relates to the field of positioning technology, and more specifically, to a positioning method, software product, and mobile device for a mobile device. Background Technology

[0002] In scenarios involving the positioning of mobile devices, global navigation satellite systems (such as GPS and BeiDou) are the mainstream solution for outdoor positioning. However, in environments where satellite signals are blocked or absent, such as inside buildings, underground parking lots, tunnels, and canyons, the positioning accuracy of global navigation satellite systems drops sharply or even fails. Positioning in these environments with blocked or absent satellite signals can be collectively referred to as indoor positioning. Currently, a commonly used indoor positioning technology is TDOA (Time Difference of Arrival) positioning technology based on radio signals. The basic principle of TDOA positioning technology is: by converting the time difference between radio signals received by the mobile device from multiple fixed base stations synchronously into the distance difference between the mobile device and each base station, and then combining this with the base station location information, the location information of the mobile device is determined through geometric calculation.

[0003] However, traditional TDOA positioning typically requires the deployment of multiple base stations to achieve accurate positioning, increasing the hardware cost and space constraints of system deployment, and is particularly unsuitable for scenarios where base station deployment is limited (such as old underground parking lots and narrow passages). When the number of deployed base stations is small, TDOA measurement can only provide positional constraints based on distance differences, lacking effective constraints on heading angles. In scenarios where mobile devices are in continuous motion, heading drift will occur over long periods of positioning, resulting in large positioning errors. Therefore, it is necessary to provide a solution that can achieve accurate positioning even with a small number of base stations. Summary of the Invention

[0004] In view of this, this application provides a positioning method for a mobile device, a program product, and a mobile device.

[0005] According to a first aspect of this application, a positioning method for a mobile device is provided, wherein the heading angle of the mobile device is consistent with its own velocity direction, the method comprising: During the movement of the mobile device, the mileage data of the mobile device at multiple time points and the TDOA measurement data at each of the multiple time points are acquired. The TDOA measurement data includes the time difference between the arrival of signals transmitted synchronously by at least two base stations to the signal receiver of the mobile device. An optimization model is constructed with the heading angle of the mobile device as the optimization variable and minimizing the sum of the target residuals corresponding to each of the multiple time points as the optimization objective. The target residual corresponding to each time point is the deviation between the measured value and the predicted value of the target parameter related to TDOA. The measured value of the target parameter is determined based on the TDOA measurement data at that time point, and the predicted value of the target parameter is determined based on the heading angle, the mileage data, and the location information of the at least two base stations. Solve the optimization model to obtain the heading angle of the mobile device, and determine the position information of the mobile device based on the heading angle and TDOA measurement data.

[0006] According to a second aspect of this application, a computer program product is provided, the computer program product comprising a computer program that, when executed, implements the method mentioned in the first aspect above.

[0007] According to a third aspect of this application, a mobile device is provided, the mobile device including a processing device, a signal receiver communicatively connected to the processing device, a wheeled odometer, an inertial measurement unit, and an altimeter. The signal receiver is used to receive signals transmitted synchronously by at least two base stations; The wheeled odometer is used to collect mileage data of the mobile device; The inertial measurement unit is used to collect motion state data of the mobile device; The altimeter is used to collect ambient air pressure data; The processing device includes a processor, a memory, and computer instructions stored in the memory that can be executed by the processor. When the processor executes the computer instructions, it implements the method mentioned in the first aspect above. The location information of the mobile device is determined based on a comprehensive analysis of the determined heading angle, the mileage data, the motion state data, and the air pressure data.

[0008] According to a fourth aspect of this application, a computer-readable storage medium is provided, characterized in that a computer program is stored on the computer-readable storage medium, and when the computer program is executed, it implements the method mentioned in the first aspect above.

[0009] Applying the solution provided in this application, for mobile devices driven by wheeled chassis, since the heading angle of such mobile devices is consistent with their own velocity direction, the mileage data and TDOA measurement data collected during the movement of the mobile device can be fully utilized to construct an optimization model with the heading angle of the mobile device as the optimization variable and minimizing the sum of the residuals between the measured and predicted values ​​of the target parameters related to TDOA as the optimization objective. After solving for the accurate heading angle, the location information of the mobile device is determined based on this heading angle. This solution can provide heading angle constraints and achieve accurate positioning even with a small number of base stations (e.g., only two base stations), significantly reducing system deployment costs. Furthermore, the deep fusion of mileage data and TDOA measurement data compensates for the lack of heading angle constraints in traditional dual-base station TDOA positioning, effectively suppressing positioning drift and improving the accuracy and robustness of indoor positioning. In addition, since only the base station needs to send downlink signals to the mobile device, this solution can accommodate multiple wireless mobile devices, meaning that multiple wireless mobile devices can be located simultaneously.

[0010] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a flowchart of a positioning method for a mobile device according to an embodiment of this application.

[0013] Figure 2 This is a schematic diagram of a base station deployment according to an embodiment of this application.

[0014] Figure 3 A schematic diagram of a downlink TDOA measurement method according to an embodiment of this application.

[0015] Figure 4 This is a schematic diagram of a method for estimating vehicle heading angle based on TDOA measurement using a geometric model, according to one embodiment of this application.

[0016] Figure 5 This is a flowchart of a method for estimating vehicle heading angle based on a TDOA differential implementation of an ultra-wideband time difference of arrival gradient field model, according to an embodiment of this application.

[0017] Figure 6This is a schematic diagram of a vehicle positioning system deployment method based on downlink TDOA, IMU, wheel speed encoder, and altimeter according to an embodiment of this application.

[0018] Figure 7 This is a flowchart of a high-precision multi-source fusion positioning error state extended Kalman filter processing algorithm according to an embodiment of this application.

[0019] Figure 8 This is a schematic diagram illustrating a situation where the system is not fully observable in a map of isogradient lines when the velocity direction is parallel to the gradient direction, according to one embodiment of this application.

[0020] Figure 9 This is a heat map showing the distribution of the TDOA gradient field strength according to an embodiment of this application.

[0021] Figure 10 This is a schematic diagram illustrating the improved positioning effect according to one embodiment of this application.

[0022] Figure 11 This is a schematic diagram of the structure of a mobile device according to an embodiment of this application. Detailed Implementation

[0023] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0024] In environments such as urban canyons, tunnels, underground parking lots, or inside buildings, the signal of global satellite navigation systems (such as GPS and BeiDou) is severely blocked and attenuated, resulting in a significant decrease in positioning accuracy or even complete failure. Therefore, indoor positioning technologies have been proposed for these environments. Indoor positioning technology refers to the technology of determining the position and attitude of a moving target in space in real time through various sensors, wireless signals, or visual information. It aims to accurately, in real time, and continuously acquire the position and motion information of moving targets in complex environments without satellite signals.

[0025] Currently, commonly used indoor positioning technologies include the following categories: (1) Bluetooth positioning technology Bluetooth positioning technology works by using pre-deployed Bluetooth beacons and estimating the relative distance between the target and the beacon by receiving signal strength from the positioning terminal. Combining data from multiple beacons, trilateration or fingerprint matching algorithms can be used to obtain the positioning result. This method has advantages such as low cost, simple deployment, and low power consumption, making it suitable for large-scale scenarios such as personnel positioning and route navigation in shopping malls, exhibition halls, or office buildings. However, Bluetooth signals are significantly affected by obstruction, reflection, and environmental noise, resulting in large errors in signal strength measurement. This leads to a positioning accuracy that is generally only 1 to 3 meters, making it difficult to meet the needs of warehousing logistics, smart buildings, and other applications.

[0026] (2) Wi-Fi positioning technology Wi-Fi fingerprint positioning is a statistical positioning method based on matching signal features in a database. Its basic process includes: first, collecting Received Signal Strength Indicator (RSSI) values ​​at various locations in the environment to establish a fingerprint database; then, during real-time positioning, matching the RSSI vectors measured by the mobile device with the fingerprints in the database (e.g., using k-nearest neighbors, Bayesian, or machine learning methods) to obtain the most probable location. The main advantage of this method is that it requires no additional hardware and can be deployed at low cost using existing Wi-Fi infrastructure. Its disadvantages include the significant impact of environmental changes on RSSI distribution, requiring periodic resampling or dynamic updates to maintain accuracy; and the relatively large error in signal strength measurement, resulting in an accuracy of only 2–5 meters for Wi-Fi fingerprint positioning, making it unsuitable for applications with high positioning accuracy requirements, such as smart manufacturing and robot navigation.

[0027] (3) Visual / laser positioning technology based on camera and lidar With the development of computer vision and SLAM (Simultaneous Localization and Mapping) technologies, camera-based and lidar-based perception and localization methods have gradually become core methods for autonomous mobile robots and unmanned systems. Camera localization extracts image feature points (such as ORB, SIFT, etc.) and performs inter-frame matching or alignment with map features, using visual odometry and optimization algorithms to estimate the camera pose. LiDAR localization relies on the geometric features of point clouds (such as walls, corners, etc.), calculating relative displacement through point cloud matching algorithms (ICP, NDT, etc.) to achieve centimeter-level localization accuracy. The advantage of visual and lidar localization is that it does not require an external signal source and can still operate in environments without infrastructure deployment. However, their performance is highly dependent on the richness of environmental features and the computing power of the device. Visual navigation systems are prone to degradation under changes in lighting or occlusion, while lidar systems have higher overall computing power requirements and equipment costs.

[0028] (4) TOF / TDOA positioning technology based on wireless signals Time-of-Flight (TOF) positioning technology works by measuring the propagation time of a signal from the transmitter (base station / moving target) to the receiver (moving target / base station). This time, combined with the speed of light, determines the distance between the moving target and the base station. By combining distance information from multiple base stations, spatial location can be calculated to pinpoint the moving target's location. This approach requires a large number of base stations and operates in a "two-way interaction between the moving target and base station." The moving target must first send an uplink signal to the base station, which then replies with a downlink signal to complete the time-of-flight measurement. Since the wireless communication channels and signal processing capabilities of base stations are limited, they can only handle a limited number of uplink requests from moving targets simultaneously. If the number of moving targets is too large, channel congestion and a surge in processing delays can occur. Therefore, this positioning system has a definite upper limit on the number of moving targets it can support.

[0029] In view of this, TDOA positioning technology was proposed. The basic principle of TDOA positioning technology is to convert the time difference of radio signals received by the mobile device from multiple fixed base stations into the distance difference between the mobile device and each base station, and then combine the base station location information to determine the location of the mobile device through geometric calculation.

[0030] Traditional TDOA positioning typically requires the deployment of multiple base stations to achieve accurate positioning. For example, at least three base stations are needed to achieve a unique positional solution in a two-dimensional plane, and even more base stations are required for three-dimensional positioning. This increases the hardware cost and space constraints of system deployment, making it particularly unsuitable for scenarios with limited base station deployment (such as old underground parking lots or narrow passages). When the number of deployed base stations is small (e.g., only two), TDOA measurement can only provide positional constraints based on distance differences, lacking effective constraints on heading angles. In scenarios where mobile devices are in continuous motion, heading drift can occur over long periods, leading to significant positioning errors. Therefore, it is necessary to provide a solution that can achieve accurate positioning even with a limited number of base stations.

[0031] In view of this, this application provides a positioning method for mobile devices. For mobile devices driven by wheeled chassis, since the heading angle of such mobile devices is consistent with their own velocity direction, the mileage data and TDOA measurement data collected during the movement of the mobile device can be fully utilized to construct an optimization model with the heading angle of the mobile device as the optimization variable and minimizing the sum of the residuals between the measured and predicted values ​​of the target parameters related to TDOA as the optimization objective. After solving for the accurate heading angle, the position information of the mobile device is determined based on the heading angle. Through the above scheme, heading angle constraints can be provided to achieve accurate positioning even with a small number of base stations (e.g., only 2 base stations), which can significantly reduce the system deployment cost. At the same time, the deep fusion of mileage data and TDOA measurement data makes up for the lack of heading angle constraints in traditional dual-base station TDOA positioning, effectively suppressing positioning drift and improving the accuracy and robustness of indoor positioning. In addition, since only the base station needs to send downlink signals to the mobile device, this scheme can accommodate multiple wireless mobile devices, that is, it can simultaneously locate multiple wireless mobile devices.

[0032] The mobile device in this application refers to a mechanical device driven by a wheeled chassis, such as a regular car, an unmanned vehicle, or a wheeled robot, which can deploy a wheeled encoder. Its main characteristic is that the wheeled chassis cannot move laterally; it can only move forward along the direction of the tires. Therefore, the velocity direction of the mobile device is the same as the heading angle direction, and it can be estimated that the velocity direction is equivalent to the heading angle direction.

[0033] The positioning method of this application embodiment can be executed by the mobile device, or it can be executed by other devices that are connected to the mobile device, such as the control device of the mobile device, cloud server, etc. After determining the location information of the mobile device, the other devices send it to the mobile device.

[0034] like Figure 1 As shown, the positioning method for a mobile device provided in this application embodiment may include the following steps: S102. During the movement of the mobile device, acquire the mileage data of the mobile device at multiple time points, and the TDOA measurement data at each of the multiple time points. The TDOA measurement data includes the time difference between the arrival of signals transmitted synchronously by at least two base stations to the signal receiver of the mobile device. In step S102, during the movement of the mobile device, mileage data and TDOA measurement data of the mobile device at multiple time points can be acquired. Mileage data can be collected by a wheel speed odometer or by other devices capable of determining mileage; this application does not impose any limitations. The mobile device is equipped with a signal receiver for receiving signals transmitted by base stations. These signals can be various types of wireless signals with strong anti-interference capabilities. TDOA measurement data can include the time difference between the arrival of signals synchronously transmitted by at least two preset base stations at the mobile device's signal receiver. For example, in some scenarios, the preset base stations include only two, such as base station A and base station B. In this case, the TDOA measurement data at each time point only includes one set of time differences, i.e., the time difference between the arrival of signals synchronously transmitted by base station A and base station B at that time point at the mobile device's signal receiver. In some scenarios, if the preset base stations include three, such as base station A, base station B, and base station C, the TDOA measurement data at each time point can include one or more sets of time differences. For example, it can include only the time difference between the arrival of signals synchronously transmitted by two of the base stations at the mobile device's signal receiver, or it can include multiple sets of time differences, each set corresponding to two of the base stations.

[0035] S104. Construct an optimization model with the heading angle of the mobile device as the optimization variable and minimizing the sum of the target residuals corresponding to each of the multiple time points as the optimization objective. The target residual corresponding to each time point is the deviation between the measured value and the predicted value of the target parameter related to TDOA. The measured value of the target parameter is determined based on the TDOA measurement data at that time point, and the predicted value of the target parameter is determined based on the heading angle, the mileage data, and the location information of the at least two base stations. In step S104, after obtaining the aforementioned data, an optimization model can be constructed based on this data. The core is to establish the correlation between "heading angle - residual - multi-source data" to achieve mathematical constraints on the heading angle. The optimization model can use the heading angle of the mobile device as the optimization variable, with the goal of minimizing the sum of the target residuals corresponding to multiple time points. The target residual at each time point is the deviation between the measured and predicted values ​​of the target parameters related to TDOA. The target parameters related to TDOA can be the distance difference between the mobile device and the base station, or the rate of change of this distance difference per unit mileage, the differential value of the distance difference, the rate of change of the distance difference per unit time, etc. These target parameters can be calculated from TDOA measurement data or based on the mobile device's trajectory. The measured value of the target parameter is determined based on the TDOA measurement data at that time point; that is, the measured value can be obtained based on the actual observations of TDOA. The predicted value of the target parameter can be determined based on the heading angle, mileage data, and the location information of at least two base stations; that is, the predicted value can be obtained based on the vehicle's trajectory.

[0036] In general, the core of the optimization model construction is based on the geometric constraint of "TDOA coupling with vehicle motion". It uses the displacement provided by wheel speed odometers to correlate TDOA (or differential TDOA, etc.) at multiple time points with the heading angle. The heading of the mobile device determines its trajectory over a short period, and the measured TDOA value (or its difference) must be consistent with the change in the TDOA field caused by the mobile device's trajectory along that heading. Therefore, the heading can be obtained by minimizing the "residual between the predicted target parameters and the measured target parameters".

[0037] S106. Solve the optimization model to obtain the heading angle of the mobile device, and determine the position information of the mobile device based on the heading angle and TDOA measurement data.

[0038] In step S106, after constructing the optimized model, the model can be solved to obtain the heading angle of the mobile device. During the solution process, candidate initial values ​​for the heading angle can be obtained first through a global grid search, and then the precise heading angle can be solved using local optimization. For example, the possible range of the heading angle can be traversed through a grid search (assuming it's 0°~180°, with the step size set according to accuracy requirements, such as 0.1°~1°), calculating the sum of residuals for each grid point, and selecting the grid point with the smallest residual as the candidate initial value. Then, a local optimization algorithm can be used iteratively to solve the problem; for example, the Levenberg-Marquardt algorithm or the Gauss-Newton method can be used to solve for the optimized value of the heading angle.

[0039] Once the heading angle is determined, the location information of the mobile platform can be determined based on the heading angle and TDOA measurement data. For example, the distance difference between the mobile device and the base station can be determined based on the TDOA measurement data. By combining the location information of the base station, the distance difference, and the heading angle, the location information of the mobile device can be solved.

[0040] In some embodiments, such as Figure 2 As shown, the at least two base stations include a first base station and a second base station, and the target parameters include the distance difference between the first base station and the second base station and the mobile device, respectively.

[0041] For example, the first and second base stations can synchronize their signal transmissions via GPS timing or the IEEE 1588 precise time synchronization protocol. The mobile device's signal receiver records the arrival timestamps of the signals from the first and second base stations in real time, calculates the difference between the two timestamps (i.e., the time difference of arrival), and then combines this with the propagation speed of the corresponding signal (e.g., the speed of light for electromagnetic waves) to calculate the difference between the distance from the mobile device to the first base station and the distance to the second base station. This is the measured value of the target parameter (distance difference). Furthermore, the mobile device collects mileage data at each time point using a wheel speed odometer. Based on the mileage data, the mobile device's heading angle (consistent with the speed direction), and the mobile device's initial position prior information (e.g., determined based on GPS positioning signals or the positioning results of the previous cycle), the position coordinates of the mobile device at each time point are calculated. Then, based on the known position coordinates of the first and second base stations, the theoretical distances from the mobile device to the first and second base stations are calculated using spatial distance formulas, and the difference is used to obtain the predicted value of the target parameter (distance difference). An optimization model is constructed with the heading angle as the optimization variable. The target residual at each time point is defined as "the absolute or squared deviation between the measured and predicted distance difference at that time point." The optimization objective is set as "minimizing the sum of the target residuals at all time points." Finally, the optimization model is solved to obtain the heading angle. This embodiment only requires the deployment of two base stations to achieve two-dimensional positioning, significantly reducing hardware costs. At the same time, by directly constraining the original distance difference, it is suitable for short-to-medium distance scenarios with stable signal environments (such as indoor warehouses and underground parking lots), ensuring positioning accuracy and robustness.

[0042] Considering that distance difference measurement is susceptible to systematic errors such as initial base station calibration errors, TDOA measurement data errors, and base station clock deviations, which may lead to poor accuracy of the heading angle obtained based on distance difference optimization, in some embodiments, differential operations can be performed on the distance difference to eliminate the influence of systematic errors. The heading angle can then be optimized based on the differential distance difference to eliminate systematic errors.

[0043] For example, in some embodiments, such as Figure 2 As shown, the at least two base stations include a first base station and a second base station, and the aforementioned target parameters include the rate of change of the distance difference between the first base station and the second base station and the mobile device per unit mileage.

[0044] For example, the first and second base stations can transmit radio signals (e.g., UWB signals) through precise time synchronization (such as GPS timing or the IEEE 1588 protocol). Mobile devices simultaneously collect both types of data: TDOA measurement data at multiple consecutive time points. Based on this TDOA measurement data, the measured TDOA distance difference and the mileage data at each time point are determined. Then, the measured values ​​of the target parameters can be calculated. For instance, two adjacent time points can be selected, and the TDOA distance difference at the later time point can be subtracted from the TDOA distance difference at the earlier time point to obtain the change in TDOA distance difference. This change is then divided by the distance traveled by the mobile device between these two time points (mileage at the later time point minus mileage at the earlier time point), yielding the measured value of the "TDOA distance difference change rate per unit mileage." A gradient field vector (calculated using the Jacobian matrix, reflecting the direction and rate of change of distance difference with position) can be constructed based on the base station's location information and the prior location information of the mobile device. This gradient field vector is then multiplied by the unit direction vector corresponding to the heading angle to obtain a predicted value for the "TDOA distance difference change rate per unit mile." This predicted value reflects the theoretical change in TDOA distance difference for each unit mile traveled by the mobile device along the current heading. An optimization model can be constructed, using the heading angle as the optimization variable. The residual for each pair of adjacent time points is defined as "measured rate of change of distance difference per unit mile - predicted rate of change of distance difference per unit mile." The optimization objective is set as "minimizing the sum of all residuals." The model is solved using a "grid search + local iteration" method to obtain the optimal heading angle, which can then be used to determine the position of the mobile device.

[0045] In some embodiments, the target parameter is the distance difference. The predicted value of the distance difference at any of the plurality of time points can be determined based on the predicted location information of the mobile device at that time point and the location information of at least two base stations. The predicted location information is determined based on the heading angle of the mobile device, the distance traveled by the mobile device between the reference time point and that time point, and the initial location information of the mobile device at the reference time point.

[0046] For example, the reference time point can be the first sampling point among multiple time points or any known location. The location information of the mobile device at the reference point can be determined by GPS or other positioning methods, or it can be the positioning result determined in the previous cycle. For any given time point, the predicted location information of the mobile device at that time point (i.e., the theoretical location calculated based on the motion trajectory) can be determined based on the initial location information of the mobile platform at the reference time point, the distance traveled by the mobile device between the reference time point and that time point, and the heading angle. Combined with the location information of the base station, the predicted value of the distance difference can be obtained.

[0047] In some embodiments, the target parameter is the rate of change of the distance difference per unit mileage. When determining the predicted value of the rate of change of the distance difference per unit mileage at any of the multiple time points, the gradient field at that time point can be determined first. The gradient field is used to characterize the rate of change and direction of the distance difference at that time point. Then, the dot product of the gradient field and the unit vector of the heading angle is determined, and this dot product is used as the predicted value of the rate of change of the distance difference per unit mileage at that time point. For example, the TDOA distance difference can be represented as a function of the vehicle position, and the gradient field can be the Jacobian matrix (gradient vector) of this function, which describes the rate and direction of change of the TDOA distance difference at a point in space. The displacement generated by the movement of the mobile device along the heading angle is the direct cause of the change in the TDOA distance difference, and the gradient field is the bridge connecting this "displacement" and the "change in the TDOA distance difference". The change in the TDOA distance difference is approximately equal to the dot product of the gradient field and the displacement of the mobile platform. Therefore, the change in the TDOA distance difference per unit mileage is equal to the dot product of the gradient field and the unit vector of the heading angle.

[0048] In some embodiments, when determining the gradient field at a given time point, the predicted change in the distance difference between the previous time point and the current time point can be determined based on the predicted location information of the mobile device at that time point, the predicted location information of the mobile device at the previous time point, and the location information of the at least two base stations. Then, the mileage of the mobile device during the time period between the previous time point and the current time point can be determined based on the mileage data at the current time point and the mileage data at the previous time point. The ratio of the predicted change to the mileage is used as the gradient field. The predicted location information at any given time point is determined based on the heading angle of the mobile device, the mileage of the mobile device between the reference time point and the current time point, and the initial location information of the mobile device at the reference time point. Of course, the gradient field can also be determined in other ways, as long as it can characterize the rate and direction of change of the TDOA distance difference at a certain time point; this embodiment of the application does not impose any limitations.

[0049] In some embodiments, the target parameter is the rate of change of distance difference per unit mileage. When determining the measured value of the rate of change of distance difference per unit mileage at any of the plurality of time points, the TDOA measurement data of the mobile device at that time point, the mileage data of that time point, the TDOA measurement data of the mobile device at the previous time point, and the mileage data of the previous time point can be acquired. Based on the mileage data of that time point and the mileage data of the previous time point, the travel mileage of the mobile device in the time period between the previous time point and the current time point is determined. Based on the TDOA measurement data of the mobile device at that time point and the TDOA measurement data of the previous time point, the change in distance difference (the measured actual change) in the time period between the previous time point and the current time point is determined. The ratio of the change in distance difference to the travel mileage is used as the measured value of the rate of change of distance difference per unit mileage.

[0050] Considering that the estimated heading angle may also be inaccurate, if the heading angle is used as a constraint to determine the position information of the mobile device, the accuracy of the determined position information may be low when the accuracy of the heading angle is too low. Therefore, in some embodiments, after determining the heading angle, a target accuracy of the heading angle can also be determined. The target accuracy can be determined based on one or more of the following parameters: the standard deviation of the TDOA measurement data at each of the multiple time points, the number of the multiple time points, the average mileage of the mobile device at the multiple time points, and the difference between the angle and the right angle between the gradient field direction and the heading angle direction at each of the multiple time points. The gradient field at each time point is used to characterize the rate of change of the distance difference at that time point. The specific method for determining the gradient field can be referred to the description of the above embodiments, and will not be repeated here.

[0051] Specifically, the target accuracy is negatively correlated with the standard deviation of the TDOA measurement data at each of the multiple time points and the difference between the angle between the gradient field direction and the heading angle direction at each of the multiple time points and a right angle, and the target accuracy is positively correlated with the number of multiple time points and the average mileage of the mobile equipment at the multiple time points.

[0052] If the target accuracy is greater than the preset accuracy threshold, the location information of the mobile device is determined based on the heading angle.

[0053] For example, a reliability barrier for the positioning results can be built through the "heading angle accuracy verification" step to avoid position estimation deviation caused by low-precision heading angle. Specifically, after obtaining the heading angle of the mobile device through the optimization model, it is not used directly for positioning. Instead, the target accuracy of the heading angle is quantitatively calculated based on the characteristics of the TDOA measurement data and the mileage data of the mobile device. The calculation logic of this target accuracy is strongly correlated with three key parameters: (1) the standard deviation of TDOA measurement data at multiple time points; the smaller the standard deviation, the lower the noise of the TDOA measurement data, the stronger the signal stability, and the more reliable the basic data for heading angle estimation, thus the higher the target accuracy. (2) the number of time points (i.e., the number of sets of TDOA measurement data and mileage data collected). The more time points, the more sufficient the constraint samples for constructing residual optimization are, which can effectively offset the random error of a single time point. The target accuracy increases with the increase of the number of time points (positive correlation). (3) The average mileage of the equipment at multiple time points. The larger the average mileage, the more significant the displacement of the equipment within the sampling period, the more obvious the change characteristics of the TDOA measurement data with position, and the stronger the constraint on the heading angle. Therefore, the target accuracy increases with the increase of the average mileage (positive correlation). (4) The difference between the angle between the gradient field direction and the heading angle direction and the right angle. Since the gradient field is characterized by the first-order partial derivative of the TDOA distance difference function with respect to position (Jacobi matrix), its vector direction is the "spatial direction of the fastest change of TDOA distance difference", and its magnitude is the "spatial rate of change of distance difference". It is the core basis for calculating the predicted value of the unit mileage change rate. Therefore, when determining the accuracy of the heading angle target, in addition to combining the traditional TDOA measurement data standard deviation (noise level), number of time points (sample sufficiency), and average mileage (motion constraint strength), we can also introduce the "difference between the angle and a right angle between the gradient field direction and the heading angle direction". The smaller the difference (the closer the angle is to 90°), the more perpendicular the equipment velocity direction (heading angle direction) is to the direction of the fastest change in TDOA (gradient field direction). At this time, the measured value and the predicted value of the rate of change per unit mileage have a higher degree of matching, the constraint signal of the heading angle estimation is stronger, the observability is better, and the heading angle target accuracy is higher. When the difference is larger (the closer the angle is to 0° or 180°, i.e., the velocity and gradient direction are parallel), the constraint signal of the rate of change per unit mileage is weakened, the observability is poor, and the heading angle target accuracy is lower. Therefore, the heading angle target accuracy is negatively correlated with this difference value.

[0054] In actual calculations, the variance or standard deviation of the estimated heading angle can be obtained using accuracy assessment models such as the Cramer-Rao boundary (CRB), which serves as a quantitative indicator of target accuracy. Then, the target accuracy can be compared with a preset accuracy threshold, which can be set according to the positioning requirements of the application scenario. If the target accuracy is greater than the threshold, the heading angle is confirmed to be reliable. Based on this heading angle, mileage data at various time points, and initial position information, the precise position of the device is calculated through kinematic integration. If the target accuracy is less than the threshold (accuracy not met), a remedial mechanism is triggered, such as re-collecting more TDOA measurement data and mileage data at more time points, resolving the heading angle after removing abnormal TDOA measurement data, or fusing data from other sensors such as the IMU to assist in correcting the heading angle, until the target accuracy meets the threshold requirements before positioning is performed.

[0055] By using the above methods, positioning risks can be avoided from the source. Quantitative accuracy verification can replace the crude method of "directly using the heading angle" and avoid position drift caused by low-precision heading angles, thus ensuring the accuracy of positioning results.

[0056] In some embodiments, the target parameter is the rate of change of the distance difference per unit mileage. The predicted value of the rate of change of the distance difference per unit mileage at any of the plurality of time points can be determined based on the dot product of the gradient field and the unit vector of the heading angle at that time point, wherein the gradient field is used to characterize the rate of change and direction of the distance difference at that time point. In this case, when determining the target accuracy of the heading angle, it can also be determined based on the difference between the angle and a right angle between the gradient field direction and the heading angle direction, and the target accuracy of the heading angle is negatively correlated with this difference.

[0057] For scenarios where the target parameter is the rate of change of distance difference per unit mileage, the heading angle accuracy assessment stage can be enhanced by adding an "angle constraint between the gradient field and the heading angle direction" to construct an accuracy judgment system that better fits the positioning logic of the gradient field model.

[0058] In some scenarios, the difference between the angle between the gradient field direction and the heading angle direction and a right angle can be represented by the three-dimensional cross product of the gradient field and the heading angle unit vector.

[0059] By introducing the "angle between the gradient field direction and the heading angle direction" as a constraint, the heading angle accuracy is determined. Compared with the traditional evaluation method that only relies on statistical data characteristics, the judgment logic is more accurate and the accuracy of heading angle accuracy evaluation is improved.

[0060] In some embodiments, a pre-verification step of "determining the straightness of the motion path" can be added before heading angle estimation to ensure that the assumptions upon which subsequent heading angle estimation relies—namely, "the heading angle and velocity direction are consistent, and the motion trajectory has no significant turning"—are valid, thus laying the foundation for positioning accuracy. For example, before acquiring the location information of at least two preset base stations, the mileage data of the mobile device at multiple time points, and the TDOA measurement data at multiple time points, it can be determined whether the motion path of the mobile device is a straight line. If so, the subsequent step of determining the heading angle is then executed.

[0061] In some embodiments, determining whether the movement path of a mobile device is a straight line can be achieved through one or more of the following methods: Method 1: If the mobile device is equipped with an inertial measurement unit (IMU), the angular velocity collected by the inertial measurement unit can be obtained, and it can be determined whether the angular velocity is less than a preset angular velocity threshold. If it is, the movement path of the mobile device is determined to be a straight line. Method 2: If the mobile device is equipped with a wheeled odometer, the mileage of the left wheel and the right wheel of the mobile device can be obtained from the wheeled odometer. If the difference between the mileage of the left wheel and the mileage of the right wheel is less than a preset threshold, the movement path of the mobile device is determined to be a straight line.

[0062] In some embodiments, the mobile device is further provided with a target sensor. When determining the position information of the mobile device based on the heading angle, data collected by the target sensor can be acquired, and then the position information of the mobile device can be determined by combining the data collected by the target sensor and the heading angle. The target sensor includes one or more of the following: an inertial measurement unit, an altimeter, and a barometer.

[0063] The positioning accuracy of mobile devices can be further improved through multi-source fusion positioning logic combining "heading angle + multi-sensor data". For example, data collected by the inertial measurement unit (IMU) and altimeter / barometer on the mobile device can be combined with the previously determined heading angle to locate the mobile device. The IMU provides dynamic motion data of the mobile device (including three-axis angular velocity and three-axis acceleration), while the altimeter / barometer collects ambient air pressure data to calculate the altitude information of the mobile device. Both can be used individually or in combination. In the core process of determining position information based on the heading angle, the heading angle is first obtained by solving the TDOA and odometer data according to the original logic, and then the acquisition and fusion of target sensor data is initiated. For the IMU, its output angular velocity data can be used to assist in correcting the heading angle. Because the IMU can capture minute steering or attitude changes of the device in real time, the integral result of the angular velocity is compared with the heading angle solved by TDOA. The two information are then fused using a Kalman filter algorithm to offset the drift error of the heading angle over time (such as the slight directional deviation when a vehicle travels straight for a long time). In addition, IMU acceleration data can be used to calibrate mileage data. The speed obtained by integrating acceleration is checked for consistency with the speed converted from mileage data. If there is a deviation (such as the vehicle skidding causing the mileage data to be too large), the mileage value is corrected based on the acceleration data to ensure the accuracy of the "mileage-location" mapping relationship.

[0064] For altimeters, their core function is to supplement altitude information in 3D positioning scenarios. In 2D planar positioning (such as for ground vehicles), barometric pressure data can be used to determine whether the device is on the same altitude plane (such as different floors in an underground parking lot), avoiding altitude interference during planar position calculations. In 3D positioning, the altitude calculated from barometric pressure data (which needs to be calibrated initially to eliminate barometric pressure reference errors) is directly combined with the planar coordinates (x, y axes) calculated from heading angle and odometer data to form complete 3D position coordinates (x, y, z). During data fusion, timestamp alignment technology is used to ensure time synchronization of heading angle, odometer data, and target sensor data. Extended Kalman filter (EKF) or unscented Kalman filter (UKF) is used to process the noise characteristics of different sensors (such as IMU integral drift and ambient pressure fluctuations from the altimeter), ultimately outputting accurate fused position information.

[0065] By fusing data from multiple sources to locate mobile devices, the accuracy of the location results can be further improved.

[0066] The positioning method provided in this application will be described below with reference to specific embodiments, taking a vehicle as an example of a mobile device.

[0067] This application's embodiment utilizes a TDOA positioning system comprising two fixed UWB base stations and a UWB transceiver on a vehicle. The signal arrival time difference between the two base stations and the vehicle, obtained using the following method, is termed a set of TDOA observations. The downlink arrival time difference is obtained at the positioning terminal by broadcasting signals from the base stations, showing the arrival time differences from each base station to the terminal. In the downlink broadcast architecture, the vehicle does not need to actively transmit signals; it only needs to receive broadcast information from the base stations for positioning. The downlink TDOA method does not require strict time synchronization, and its broadcast positioning method gives the system the characteristic of unlimited positioning capacity. In the downlink broadcast architecture, the base stations do not need to constantly be in listening mode to monitor signals, reducing system power consumption.

[0068] like Figure 3 As shown, assume the UWB base stations are M and S, and the UWB transceiver on the vehicle is T. In the downlink TDOA measurement method, strict time synchronization is not required between the UWB base stations. After base station M broadcasts a signal, S and T record the time of receiving the signal; after a set time, base station S broadcasts a signal, and T records the time of receiving the signal. Calculated based on the known deployment locations of base stations. The time difference between the two received signals. Let S be the time difference between sending and receiving the signal. From these three times, the arrival time difference can be obtained. : Figure 2 The specific implementation and deployment method of this measurement method is shown. The signal arrival time difference from the two base stations (i.e., the first base station and the second base station) to the vehicle is obtained by the above two methods. Multiply by the electromagnetic wave propagation speed constant The distance difference between the two base stations and the vehicle can be obtained. In the embodiments of this application, distance differences are used. The time difference will not be used directly. Therefore, the TDOA observations mentioned below all refer to the distance difference between the two base stations and the vehicle. .

[0069] The “single downlink TDOA observation” in the embodiments of this application refers to a set of TDOA measurements obtained at the same time using only two ultra-wideband base stations in the system.

[0070] In the symbols used in the embodiments of this application, unbold lowercase letters represent scalars, bold lowercase letters represent vectors, and bold uppercase letters represent matrices.

[0071] In one embodiment, a method for estimating vehicle heading angle based on TDOA measurement using a geometric model is provided, and the specific implementation is as follows: This method is applicable to vehicles moving in a straight line on a two-dimensional plane, such as... Figure 4 As shown, base stations are A and B. Along the vehicle's travel path, a set of TDOA measurements was obtained at points B, C, and D. The baseline length is known. This can be obtained through an odometer. , TDOA measurement value , Choose a coordinate system with A as the origin and AB as the positive x-axis: The vehicle moves in a straight line, and the angle between its direction and the baseline (AB) is denoted as . That is, the heading angle. Let... Define the Euclidean distance to A and B: Similarly, define AD, BD, AE, and BE. Let the function... : in: Solve the system of nonlinear equations Numerical methods are typically used to solve this problem: First, initial parameter values ​​can be estimated. . The estimated position of the vehicle at that moment can be used. Can first A grid search is performed within the range at certain angular intervals (e.g., 0.5° or 1°). Then, the Levenberg-Marquardt and Newton algorithms can be used to find the roots, and the solution is obtained. This is the desired vehicle heading.

[0072] In another embodiment, a TDOA differential implementation method for estimating vehicle heading angle based on an ultra-wideband time difference of arrival gradient field model is provided, the flowchart of which is shown below. Figure 5 As shown.

[0073] A local coordinate system is established with the east, north, and sky directions as the positive x, y, and z axes, respectively. Within this local coordinate system, the known locations of the two ultra-wideband base stations are... Then in position The TDoA measurement value at the location is recorded as . This represents the field of TDoA value in space.

[0074] TDoA measurements at location The first derivative Jacobian matrix at the point ,Right now The gradient field is In smaller adjacent moments Within this range, using a first-order Taylor expansion approximation, the change in the TDoA measurement is as follows: Right now ,in The unit vector in the heading direction. The distance traveled by the vehicle is obtained from the wheel speed encoder.

[0075] Remember the moment The position is TDoA measurement value The vehicle's travel distance obtained by the wheel speed encoder is The difference between adjacent TDoA measurements is , As a constant, the above approximation is: ,Will Recorded as Initial position Given that, after observing multiple sets of data (N sets of data), the system of equations can be solved: The following is an example of a numerical optimization method for solving this problem in two dimensions: For a two-dimensional unit vector, use angles express: Step 1: Define the cost function (least squares form) For each (i), calculate the residual (error): The total cost function is then defined as: The problem is transformed into: Step 2: Optimize the initial point using grid search because There may be multiple local minima. To avoid getting trapped in a local minimum: First, you can start at A grid search is performed within the range at certain angular intervals (e.g., 0.5° or 1°). Secondly, the cost for all candidate angles can be calculated. Then we can find the angles with the lowest cost as the initial values ​​for optimization.

[0076] Step 3: Local Optimization Using the angle found in step 2 as the initial point, numerical optimization methods are used to further refine the solution: The gradient can be calculated, and the optimal angle can be obtained using the Gauss-Newton gradient method. Solving for the results This is the desired vehicle heading.

[0077] A covariance estimation algorithm for the TDOA differential heading angle estimation algorithm is as follows: If N sets of data are observed during the uniform speed and uniform time observation of the trolley traveling in a two-dimensional plane, the Cramé-Rao boundary of the above TDOA differential heading angle estimation is used as... : in, For the heading angle covariance, For heading angle estimation, The standard deviation of TDoA measurement for ultra-wideband networks. The number of equations. This represents the average vehicle distance traveled per segment of the wheel speed encoder. Let be the first-order derivative Jacobian matrix of the car's position. Note that when applying the cross product here, the vector is expanded to three dimensions, with the third dimension being 0: , Then apply. This is a covariance estimate for the TDOA differential heading angle estimation. The number of data observation sets N used in each estimation can be a fixed value, or it can be based on the vehicle mileage obtained from each wheel speed encoder segment. Adaptive adjustment.

[0078] Furthermore, in one embodiment, an error suppression algorithm for TDOA differential heading angle estimation is provided, as follows: The covariance of the heading angle estimate can be calculated. If the covariance is greater than the set threshold, the estimate is considered invalid.

[0079] Alternatively, the difference between the estimated heading angle and the current heading angle state can be calculated. If the difference is greater than the set threshold, the estimate is considered invalid. Alternatively, the residuals of the aforementioned cost equation can be calculated: If all residuals are greater than the set threshold, the estimate is considered invalid.

[0080] In one embodiment, a vehicle positioning system based on downlink TDOA, IMU, wheel speed encoder, and altimeter / barometer is provided.

[0081] like Figure 6 As shown, the system includes a positioning vehicle and two UWB base stations for TDOA measurement. Within the positioning area, high-precision, unbiased positioning of the vehicle is achieved through ultra-wideband downlink TDOA measurement and sensor measurements on the vehicle.

[0082] The UWB base station includes: a UWB transceiver, which communicates with the UWB base station and the positioning terminal to perform TDOA (Transmission Time Orientation Measurement).

[0083] The positioning vehicle includes: a UWB signal receiver for communicating with a UWB base station group to measure distance difference; a wheel speed encoder for measuring vehicle mileage; an inertial measurement unit (IMU) for measuring angular velocity and acceleration; an altimeter / barometer for measuring the absolute altitude of the vehicle; and an information processing unit for multi-source fusion positioning and other functions.

[0084] This positioning system uses a high-precision multi-source fusion positioning error state extended Kalman filter processing algorithm in the vehicle's information processing unit to achieve high-precision vehicle positioning. For example... Figure 7 As shown, the positioning algorithm includes an ultra-wideband TDOA measurement module, an IMU module, a wheel speed encoder module, an altimeter / barometer module, a ranging preprocessing module, a linearity check module, a heading angle estimation and error suppression module, and an extended Kalman filter module.

[0085] The ultra-wideband TDOA measurement module obtains TDOA measurement values ​​using the aforementioned uplink or downlink TDOA measurement methods. .

[0086] IMU module obtains vehicle angular velocity and acceleration .

[0087] The wheel speed encoder module obtains the distance traveled by the wheels. .

[0088] The altimeter module obtains the vehicle's altitude. .

[0089] The ultra-wideband smoothing module performs gross error removal and smoothing on ultra-wideband measurements. The following is an example: Median filtering is used to remove outliers: time points are recorded. The obtained TDOA measurement value In the window length is Measured values The median is If the difference between the current measured value and the median is If the value exceeds the threshold, the current measurement is considered an outlier and is not considered a valid measurement.

[0090] Smoothing distance measurement using moving average filtering: with the window length after outlier removal as... Measured values average This is the current measurement value.

[0091] The data queue stores smoothed ultrawideband downlink TDOA measurements. 1. Mileage traveled by the wheels from the wheel speed encoder module Initial position estimation from the error state Kalman filter . It is a time series.

[0092] The linearity check module detects whether the vehicle is traveling in a straight line. Here are two examples: 1. Based on the angular velocity input from the IMU module, when the angular velocity is less than the threshold, the vehicle is considered to be traveling in a straight line. 2. Based on the left and right wheel mileage input from the wheel speed encoder, when the difference in mileage between the left and right wheels is less than the threshold, the vehicle is considered to be traveling in a straight line. If the linearity check passes and the vehicle travels in a straight line, then subsequent heading angle estimation can be performed; otherwise, heading angle estimation is not performed.

[0093] The heading angle estimation and error suppression module estimates the vehicle's heading angle. The specific process can be found in the description of the aforementioned embodiments.

[0094] Error-state extended Kalman filtering uses the above-mentioned measurement and estimation data to complete the localization. The following is an example: For attitude Euler angles, For velocity vector, This is a three-dimensional position vector in Earth coordinate system. Longitude As a dimension, For height, For zero bias of the gyroscope and accelerometer, This indicates the error between the true physical value and the predicted state. (Superscript) This represents the projection of the physical quantity into the navigation coordinate system. The angular velocity of Earth's rotation. This represents the angular velocity of the navigation coordinate system relative to the Earth coordinate system. This is the rotation matrix from the vehicle coordinate system to the navigation coordinate system. The points above the variables represent the derivatives of those variables. This is state noise. To indicate the noise level during observation, the subscripts represent different noise measurements.

[0095] The pose state of the located vehicle is described by the following fifteen-dimensional error vector: Based on the measurement principle of the inertial sensing unit, the state transition equation is: ,in, The system contains four types of observations: (1) Ultra-wideband TDoA distance observation: Let the latitude, longitude and altitude coordinates of two ultra-wideband a and b in the geographic coordinate system be denoted. , The coordinates have been determined. Calculate the radii of the local meridian and the west-southeast circle. , When the range of motion of the carrier is relatively small, the radii of the local meridian and the zonal circle are approximately assumed to remain unchanged: in For the Earth's long radius, For the first eccentricity, but in, , , The observation equation is: .

[0096] (2) The vehicle heading observation matrix obtained by the aforementioned heading angle calculation is as follows: The observation equation is: (3) Observation of carrier velocity Body velocity measurement: Find the form of the error: Therefore, the observation matrix is Representing vectors The corresponding antisymmetric matrix.

[0097] The observation equation is: (4) Altimeter for absolute altitude The observations were conducted, and the observation matrix is ​​as follows: The observation equation is: .

[0098] Based on the above state transition equations and observation equations, extended Kalman filtering of the error state can be performed.

[0099] In one embodiment, an observability analysis theory for a positioning system based on ultra-wideband time difference of arrival measurement and inertial sensing units, wheel speed encoders, and altimeters is provided.

[0100] As mentioned earlier, the pose state of an agent is represented by the following fifteen-dimensional error vector. Describe: To simplify the analysis of the system's observability, the three-dimensional position of the navigation coordinate system (northeast to sky) is replaced by the three-dimensional position of latitude, longitude, and altitude in the Earth coordinate system. Under the premise of neglecting the previous condition, attitude, velocity, and position coordinates are decoupled, so the transformation of position coordinates does not affect other parts of the matrix.

[0101] Assuming the agent's movement is at a low speed and within a small range, and The magnitude of the error is very small, and its related terms are all non-major errors. The non-major error terms in the error equation also contribute very little to the observability of the system. In the qualitative analysis of observability, in order to avoid the interference of minor factors on major factors, the error terms related to the state matrix should be discarded. Based on the above assumptions and the state transition equation described in the third aspect, the state transition equation based on IMU measurement is... Its state transition matrix It can be simplified to: Under the above assumptions, the TDoA base station observes the distance: The TDoA error observations are obtained by performing a first-order linearization expansion: remember Let be the unit vector from base station a to the agent. In the three-dimensional case, For a 1*3 configuration, consider a two-dimensional case. The third dimension is 0.

[0102] As mentioned above, the observation matrix for heading angle estimation is: As mentioned above, the observation matrix for velocity estimation is: As mentioned above, the altimeter observation matrix is ​​as follows: Therefore, the observation equations can be combined and written as: Based on the observability analysis theory of piecewise linear time-invariant systems (PWCS) proposed by Goshen-Meskin and Bar-Itzhack, the observability matrix for the j-th time interval is... for: Non-zero behaviors include: The cumulative observation matrix for adjacent time points j and k If the rank is full (rank is 15), then the system is fully observable.

[0103] The results of the analysis of the applicability conditions of this system are as follows: (1) If That is, when the velocity direction is parallel to the gradient direction, the system is not completely observable. For example... Figure 8 As shown in the isogradient map, the velocity direction at this time is... With gradient direction Parallelism, the system is not fully observable, where observable means that the location information of movable devices determined by the scheme of this application is valid (i.e., relatively accurate), and unobservable means that the location information of movable devices determined by the scheme of this application is invalid (i.e., inaccurate). Additionally, when the gradient field strength... hour It will also approach 0, therefore in Within smaller areas, observability is poor. Figure 9 A heatmap showing the distribution of the TDOA gradient field strength is shown. The red area represents the region with a large gradient field strength, which is well observable. The blue area has a small gradient field strength, making accurate positioning difficult. (2) If the vehicle is moving at a constant speed in a straight line or is stationary, i.e. and (3) When the system remains unchanged, it is not fully observable; (4) When there is acceleration, deceleration and direction change, and the velocity direction is not parallel to the gradient direction, the system is fully observable.

[0104] The method and system described in the embodiments of this application for vehicle positioning have the following gain effect: Figure 10 As shown.

[0105] Under the same set of simulation data, the left figure shows the integrated navigation system without using the heading estimation method, and the right figure shows the integrated navigation system using the heading estimation method described in this patent. The root mean square error (MRSE) of positioning using the method described in this application's embodiments decreased from 0.522m to 0.269m, and the maximum heading angle error decreased from 4.9... Decreased to 2.8 This effectively improved positioning accuracy.

[0106] The solution in this application embodiment can achieve the following beneficial effects: (1) This application designs two methods for estimating the vehicle heading angle of TDOA: one is a geometric method for estimating the heading angle, and the other is a gradient field method for estimating the heading angle. These two estimation methods can effectively eliminate systematic errors in ultra-wideband measurements and realize vehicle heading angle constraints, which helps to improve positioning accuracy and reduce and then eliminate positioning drift; (2) Single TDOA observation refers to the system using only two ultra-wideband base stations and obtaining only one set of TDOA measurements at the same time. This application designs an ultra-wideband positioning system using only two base stations, reducing the need for base station deployment and providing a robust positioning method in scenarios such as sparse base stations and indoor obstruction; (3) The multi-sensor vehicle fusion extended Kalman filter positioning system with single TDOA observation, IMU, wheel speed encoder, and altimeter can provide higher accuracy and low drift navigation; In this system, the deployment and installation cost of UWB base stations is low, no high-precision calibration is required, and the measurement errors caused by initial base station calibration accuracy, ultra-wideband antenna delay, and ultra-wideband clock deviation are tolerated, achieving anti-interference, high reliability, and robust positioning; (4) This application theoretically demonstrates that an observable positioning system can be realized using TDOA, IMU, wheel speed encoder, and altimeter, and clarifies the applicable conditions of the system.

[0107] The solutions in the above embodiments can be freely combined to obtain new solutions when there is no conflict. Due to space limitations, they will not be listed one by one here.

[0108] Furthermore, this application also provides a computer product, which includes a computer program that, when executed, implements the method described in any of the above embodiments.

[0109] Furthermore, embodiments of this application also provide a mobile device, such as... Figure 11 As shown, the mobile device 10 includes a processing unit 11, a signal receiver 12 and a wheeled odometer 13 communicatively connected to the processing unit 11. The processing unit 11 includes a processor 111, a memory 112, and computer instructions stored in the memory 112 that can be executed by the processor 11. When the processor 111 executes the computer instructions, it implements the method described in any of the above embodiments. The mobile device can also be configured with an inertial measurement unit 14, a barometric altimeter 15, etc., based on actual needs.

[0110] Signal receiver 12 is used to receive signals synchronously transmitted by at least two base stations to obtain TDOA measurement data. A wheeled odometer is used to collect mileage data of the mobile device. An inertial measurement unit is used to collect motion state data of the mobile device. An altimeter / barometer is used to collect ambient air pressure data. The location information of the mobile device can be determined based on a comprehensive analysis of the determined heading angle, mileage data, motion state data, and air pressure data. Accordingly, this application also provides a computer storage medium storing a program that, when executed by a processor, implements the methods in any of the above embodiments.

[0111] The embodiments of this application may take the form of a computer program product implemented on one or more storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing program code. Computer-usable storage media include permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information may be computer-readable instructions, data structures, program modules, or other data.

[0112] The methods and apparatus provided in the embodiments of this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are only for the purpose of helping to understand the methods and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this application should not be construed as a limitation of this application.

Claims

1. A positioning method for a mobile device, characterized in that, The heading angle of the mobile device is consistent with its own velocity direction, and the method includes: During the movement of the mobile device, the mileage data of the mobile device at multiple time points and the TDOA measurement data at each of the multiple time points are acquired. The TDOA measurement data includes the time difference between the arrival of signals transmitted synchronously by at least two base stations to the signal receiver of the mobile device. An optimization model is constructed with the heading angle of the mobile device as the optimization variable and minimizing the sum of the target residuals corresponding to each of the multiple time points as the optimization objective. The target residual corresponding to each time point is the deviation between the measured value and the predicted value of the target parameter related to TDOA. The measured value of the target parameter is determined based on the TDOA measurement data at that time point, and the predicted value of the target parameter is determined based on the heading angle, the mileage data, and the location information of the at least two base stations. Solve the optimization model to obtain the heading angle of the mobile device, and determine the position information of the mobile device based on the heading angle and TDOA measurement data.

2. The method according to claim 1, characterized in that, The at least two base stations include a first base station and a second base station, and the target parameters include the distance difference and / or the rate of change of the distance difference per unit mileage; wherein the distance difference is the difference between the distance between the first base station and the second base station and the mobile device.

3. The method according to claim 2, characterized in that, The target parameter is the distance difference. The predicted value of the distance difference at any of the plurality of time points is determined based on the predicted location information of the mobile device at that time point and the location information of the at least two base stations. The predicted location information is determined based on the heading angle of the mobile device, the distance traveled by the mobile device between the reference time point and the current time point, and the initial location information of the mobile device at the reference time point. and / or The target parameter is the rate of change of the distance difference per unit mileage. The measured value of the rate of change of the distance difference per unit mileage at any of the plurality of time points is determined based on the following method: acquiring the TDOA measurement data of the mobile device at that time point, the mileage data at that time point, the TDOA measurement data of the mobile device at the previous time point, and the mileage data at the previous time point; determining the travel mileage of the mobile device during the time period between the previous time point and the current time point based on the mileage data of the mobile device at that time point and the mileage data of the previous time point; determining the change in the distance difference during the time period between the previous time point and the current time point based on the TDOA measurement data of the mobile device at that time point and the TDOA measurement data of the previous time point; and using the ratio of the change in the distance difference to the travel mileage as the measured value of the rate of change of the distance difference per unit mileage.

4. The method according to claim 2, characterized in that, The target parameter is the rate of change of the distance difference per unit mileage, and the predicted value of the rate of change of the distance difference per unit mileage at any of the plurality of time points is determined based on the following method: Determine the gradient field at that time point, wherein the gradient field is used to characterize the rate of change and direction of the distance difference at that time point; Determine the dot product of the gradient field and the unit vector of the heading angle, and use the dot product as the predicted value of the rate of change of the distance difference per unit mileage at the time point.

5. The method according to claim 4, characterized in that, Determining the gradient field at this time point includes: based on the predicted location information of the mobile device at this time point, the predicted location information of the mobile device at the previous time point, and the location information of the at least two base stations, determining the predicted change value of the distance difference during the time period between the previous time point and this time point; Based on the mileage data at this time point and the mileage data at the previous time point, the movement mileage of the mobile device during the time period between the previous time point and this time point is determined. The ratio of the predicted change value to the distance traveled is used as the gradient field; wherein, the predicted position information at any time point is determined based on the heading angle of the mobile device, the distance traveled by the mobile device between the reference time point and the reference time point, and the initial position information of the mobile device at the reference time point.

6. The method according to claim 2, characterized in that, After determining the heading angle, the method further includes: Determine the target accuracy of the heading angle; If the target accuracy is greater than a preset accuracy threshold, the position information of the mobile device is determined based on the heading angle; The target accuracy is determined based on one or more of the following parameters: the standard deviation of the TDOA measurement data at each of the plurality of time points, the number of the plurality of time points, the average mileage of the mobile device at the plurality of time points, and the difference between the angle between the gradient field direction and the heading angle direction at each of the plurality of time points and a right angle; the gradient field at each time point is used to characterize the rate of change of the distance difference at that time point; The target accuracy is negatively correlated with the standard deviation and the difference, and positively correlated with the number of the plurality of time points and the average mileage.

7. The method according to any one of claims 1-6, characterized in that, Before acquiring the location information of at least two preset base stations, the mileage data of the mobile device at multiple time points, and the TDOA measurement data at each of the multiple time points, the method further includes: Determine whether the movement route of the mobile device is a straight line; if so, execute the steps of acquiring the location information of at least two preset base stations, the mileage data of the mobile device at multiple time points, and the TDOA measurement data at each of the multiple time points; and / or The mobile device is also equipped with a target sensor. Determining the position information of the mobile device based on the heading angle includes: acquiring data collected by the target sensor; determining the position information of the mobile device based on the data collected by the target sensor and the heading angle; wherein, the target sensor includes one or more of the following: an inertial measurement unit and an altimeter.

8. The method according to claim 7, characterized in that, Determining whether the movement path of the mobile device is a straight line includes: The mobile device is equipped with an inertial measurement unit to acquire the angular velocity collected by the inertial measurement unit. If the angular velocity is less than a preset angular velocity threshold, the movement path of the mobile device is determined to be a straight line. and / or The mobile device is equipped with a wheeled odometer to obtain the mileage of the left wheel and the right wheel of the mobile device collected by the wheeled odometer. If the mileage of the left wheel and the mileage of the right wheel are less than a preset threshold, the movement path of the mobile device is determined to be a straight line.

9. A computer program product, characterized in that, The computer program product includes a computer program that, when executed, implements the method as described in any one of claims 1-8.

10. A mobile device, characterized in that, The mobile device includes a processing unit, a signal receiver, a wheeled odometer, an inertial measurement unit, and an altimeter, all communicatively connected to the processing unit. The signal receiver is used to receive signals transmitted synchronously by at least two base stations; The wheeled odometer is used to collect mileage data of the mobile device; The inertial measurement unit is used to collect motion state data of the mobile device; The altimeter is used to collect ambient air pressure data; The processing device includes a processor, a memory, and computer instructions stored in the memory that can be executed by the processor. When the processor executes the computer instructions, it implements the method as described in any one of claims 1-8, wherein the location information of the mobile device is determined based on a comprehensive analysis of the determined heading angle, the mileage data, the motion state data, and the air pressure data.

Images