Position prediction method and device based on UWB digital key

Abstract

The invention provides a position prediction method and device based on an ultra wide band (UWB) digital key, and a position prediction system applied to the UWB digital key, the position prediction system comprises a UWB tag and a plurality of UWB anchor points arranged on a vehicle, and the position prediction method comprises the following steps: determining whether UWB positioning information of the UWB tag is invalid; when the UWB positioning information is invalid, acquiring a velocity vector and a historical track set of the UWB tag and orientation information of the vehicle; and inputting the velocity vector, the historical trajectory set and the orientation information into a position prediction calculation model to calculate a predicted position of the UWB tag. The technical problem that when an existing automobile digital key is blocked by an obstacle and the digital key cannot be connected to an automobile UWB anchor point, the function of the digital key is in a failure or invalid state, and the user experience is affected is solved.

Description

Technical Field

[0001] The present application relates to the field of automobile technology, and in particular to a method and device for predicting a position of a digital key based on an ultra-wideband (UWB) digital key. Background Art

[0002] The implementation of the Ultra Wide Band (UWB) digital key for cars mainly relies on the distance measurement between the control terminal equipped with the digital key and the vehicle's UWB anchor point to achieve precise positioning. When there are obstacles between the control terminal equipped with the digital key and the UWB anchor point, such as pillars, this will block the distance measurement signal between the control terminal and the vehicle's UWB anchor point, resulting in positioning failure. When positioning fails, the digital key function will be in an invalid or ineffective state, which will seriously affect the user experience.

[0003] The problem is that when the current car digital key is blocked by an obstacle and cannot connect to the car's UWB anchor point, the digital key function will become invalid or ineffective, affecting the user experience. Summary of the invention

[0004] This application solves the technical problem that when the current digital key of a car is blocked by obstacles and cannot connect to the UWB anchor point of the car, the digital key function will be in an invalid or invalid state, which affects the user experience. When UWB positioning fails, this application obtains the speed vector, historical trajectory set and vehicle orientation of the control terminal as input parameters, inputs the predicted position calculation model, and calculates the predicted position, ensuring that the digital key function remains effective and improving the user experience.

[0005] To solve the above problems, the present application provides a position prediction method based on an ultra-wideband UWB digital key, which is applied to a position prediction system of a UWB digital key. The position prediction system includes a UWB tag and multiple UWB anchor points arranged on a vehicle. The position prediction method includes: determining whether the UWB positioning information of the UWB tag is invalid; when the UWB positioning information is invalid, obtaining a speed vector and a historical trajectory set of the UWB tag and the orientation information of the vehicle; and inputting the speed vector, the historical trajectory set and the orientation information into a position prediction calculation model to calculate the predicted position of the UWB tag.

[0006] Compared with the existing technology, the technical effect achieved by adopting this technical solution is: by predicting the UWB tag position when the UWB digital key positioning fails, the connection stability and reliability between the digital key and the car are improved. By combining the predicted position calculation model with parameters such as speed vector, historical trajectory set and vehicle orientation, the position of the digital key can be accurately predicted, so that users do not need to wait for UWB positioning to recover, which improves the user experience.

[0007] In one possible design, a velocity vector, a historical trajectory set, and a direction information are input into a position prediction calculation model to calculate a predicted position of the UWB tag, including: inputting a velocity vector, a historical trajectory set, and a direction information into a position prediction calculation model to calculate a predicted initial position; and calculating a predicted position by combining a Kalman filter algorithm with the predicted initial position.

[0008] Compared with the existing technology, the technical effect achieved by adopting this technical solution is as follows: by calculating the predicted initial position and then optimizing the predicted initial position using the Kalman filter algorithm, the formed UWB tag trajectory can be smoothed and closer to the actual walking trajectory of pedestrians, further improving the accuracy and stability of position prediction.

[0009] In a possible design, determining whether UWB positioning information of a UWB tag is invalid includes: measuring current distance information between the UWB tag and multiple UWB anchor points; if valid current distance information cannot be obtained, or the displacement of the obtained current distance information exceeds a set range threshold compared to the historical distance information obtained last time, then determining that the UWB positioning information of the UWB tag is invalid.

[0010] Compared with the existing technology, the technical effect achieved by adopting this technical solution is: by measuring the current distance information between the UWB tag and multiple UWB anchor points, the accuracy of the UWB positioning information can be monitored in real time, and the failure of the positioning information can be discovered in time. By setting the range threshold, and then comparing the displacement between the current position and the previous position with the range threshold, the failure of the UWB positioning information can be effectively judged, and the system's ability to judge the reliability of the UWB tag position status is improved.

[0011] In one possible design, the history trace set includes: in, is a collection of historical trajectories, is the historical trajectory point, the historical trajectory set Save at least the last five valid track positions of the UWB tag; The predicted initial position is calculated, including: The predicted initial position is calculated by the prediction formula, where the prediction formula is as follows: in, To predict the initial position; For the UWB tag in the UWB tag coordinate system, The value of the direction; For the UWB tag in the UWB tag coordinate system, The value of the direction; It is the UWB tag orientation information; is the moving speed of the UWB tag; is the direction information, and the direction information is the last valid direction information of the vehicle before parking; Corresponding historical trajectory points degree of relationship; A historical track point In the UWB tag coordinate system The value of the direction; A historical track point In the UWB tag coordinate system The value of the direction.

[0012] Compared with the prior art, the technical effect achieved by adopting this technical solution is as follows: the historical trajectory set stores at least the last five valid trajectory positions of the UWB tag, which enables the predicted position calculation model to more accurately analyze the movement trend and trajectory changes of the UWB tag, which improves the accuracy and stability of the predicted position. The historical trajectory set helps the model to better understand the movement law of the UWB tag, which enables the predicted position calculation model to more accurately predict the position of the UWB tag installed with the digital key, thereby improving the reliability and user experience of the position prediction system. By considering factors such as the velocity vector, historical trajectory and vehicle orientation, the prediction formula can more accurately calculate the predicted initial position, thereby improving the accuracy of the position prediction. The prediction formula includes the relationship degree parameter of the historical trajectory points, and the predicted position can be adjusted according to the importance of the historical trajectory points, making the prediction model more flexible and intelligent. The prediction formula takes into account the influence of the vehicle orientation, which can better adapt to the position prediction requirements under different vehicle motion states, thereby improving the stability and reliability of the system. The vehicle orientation is one of the important reference information for vehicle motion. By providing the information of the last valid orientation before the vehicle stops, it helps the predicted position calculation model to better understand the vehicle's motion state and direction, thereby increasing the accuracy and reliability of the predicted position of the UWB tag.

[0013] In a possible design, the predicted position is obtained by combining the Kalman filter algorithm with the predicted initial position calculation, including: calculating the Kalman predicted position according to the historical trajectory set, the initialization formula and the Kalman prediction formula; correcting the Kalman predicted position according to the update formula and the predicted initial position to obtain the predicted position.

[0014] Compared with the existing technology, the technical effect achieved by adopting this technical solution is as follows: the Kalman filter algorithm can correct the predicted position more stably, reduce the error and fluctuation of the position prediction, and improve the accuracy and stability of the position prediction. According to the historical trajectory set, the initialization formula and the Kalman prediction formula, it can adapt to the dynamic changes of the UWB tag more quickly, accelerate the convergence process, make the predicted position approach the real position more quickly, and improve the response speed of the system. The update formula can automatically adjust the predicted position according to the difference between the actual observation value and the predicted initial position, making the correction more accurate and effective, and improving the accuracy and stability of the position prediction.

[0015] In one possible design, the initialization formula includes a Kalman initialization formula, which is as follows: in, is the state error covariance matrix: is the process noise covariance; is the measurement noise covariance; is the measurement matrix.

[0016] Compared with the prior art, the technical effect achieved by adopting this technical solution is as follows: through parameters such as the state error covariance matrix, process noise covariance and measurement noise covariance, the Kalman initialization formula can more accurately estimate the state of the UWB tag, improving the accuracy and stability of the position prediction. By setting the process noise covariance and the measurement noise covariance, the Kalman initialization formula can more effectively handle noise interference, reduce the impact of noise on position prediction, and improve the robustness of the system. The introduction of the measurement matrix enables the Kalman initialization formula to more flexibly handle different types of measurement data, adapt to different system environments and sensor configurations, and improve the versatility and adaptability of the position prediction system.

[0017] In one possible design, the Kalman prediction formula includes: .

[0018] Compared with the existing technology, the technical effect achieved by adopting this technical solution is: when predicting the state through the Kalman prediction formula, the previous state prediction value is retained, and the state is updated according to the process noise covariance, making the state prediction more stable and continuous, reducing the accumulation of prediction errors, and improving the accuracy of position prediction. The simplicity and efficiency of the Kalman prediction formula enable the system to predict and update the state more quickly, improving the system's response speed and real-time performance.

[0019] In one possible design, the update formula includes a Kalman update formula, which is as follows: .

[0020] Compared with the prior art, the technical effect achieved by adopting this technical solution is as follows: when the state is corrected by the Kalman update formula, factors such as the measurement noise covariance, the state error covariance and the measurement matrix are taken into account, so that the predicted position can be corrected more accurately, the correction error is reduced, and the accuracy of the position prediction is improved. The Kalman update formula can be dynamically adjusted according to the measurement noise and the state error covariance, realizing the effective fusion of the measurement data and the prediction data, and improving the accuracy and stability of the correction. Through the state update method of the Kalman update formula, the system can correct the state more robustly, avoiding the position prediction deviation caused by noise interference or uncertainty, and improving the robustness of the position prediction system.

[0021] In one possible design, the location prediction calculation model will periodically calculate the predicted location of the UWB tag.

[0022] Compared with the existing technology, the technical effect achieved by adopting this technical solution is: by periodically calculating the predicted position, the system can timely update the location information of the UWB tag and maintain the real-time nature of the location prediction. Periodically calculating the predicted position can continuously monitor and update the location of the UWB tag, ensuring the continuity and stability of the location prediction, and avoiding the location deviation problem caused by long-term non-update.

[0023] In one possible design, the predicted position of the UWB tag is continuously calculated based on the position prediction calculation model and the historical trajectory set is updated until the UWB positioning information of the UWB tag is restored to validity.

[0024] Compared with the existing technology, the technical effect achieved by adopting this technical solution is: through continuous calculation based on the location prediction calculation model, the predicted location and historical trajectory set of the UWB tag can be updated in real time. This means that even if the UWB positioning information fails, the system can still make location predictions based on other information and continuously update the prediction results. In the case of failure of UWB positioning information, reliable location prediction results are still provided, which solves the problem that the digital key cannot connect to the car anchor point and improves the user experience.

[0025] The present application also provides a position prediction device based on an ultra-wideband UWB digital key, and the position prediction device based on an ultra-wideband UWB digital key implements a position prediction method based on an ultra-wideband UWB digital key.

[0026] Compared with the prior art, the technical effect achieved by adopting this technical solution is as follows: the position prediction device based on ultra-wideband UWB digital key of the present application includes the position prediction method based on ultra-wideband UWB digital key of any technical solution of the present application. Therefore, it has all the beneficial effects of the position prediction method based on ultra-wideband UWB digital key of any technical solution of the present application, which will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 A flowchart of a location prediction method based on an ultra-wideband (UWB) digital key provided in an embodiment of the present application; Figure 2 A schematic diagram of an embodiment of the present application in which there is an obstacle between the UWB tag and the UWB anchor point, rendering positioning ineffective; Figure 3 A schematic diagram of UWB tag trajectory prediction provided in an embodiment of the present application; Figure 4 A schematic diagram of the calculation update mechanism of the predicted position calculation model provided in an embodiment of the present application. DETAILED DESCRIPTION

[0028] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0029] See also Figures 1 to 4 This embodiment provides a location prediction method based on an ultra-wideband UWB digital key, which is applied to a location prediction system for a UWB digital key. The location prediction system includes a UWB tag and a plurality of UWB anchor points arranged on a vehicle. The location prediction method includes: Determine whether the UWB positioning information of the UWB tag is invalid; When the UWB positioning information fails, the velocity vector and historical trajectory set of the UWB tag and the vehicle's orientation information are obtained; The velocity vector, historical trajectory set and orientation information are input into the position prediction calculation model to calculate the predicted position of the UWB tag.

[0030] Specifically, Figure 2 The schematic diagram shows that there is an obstacle between the UWB tag and the UWB anchor point, which makes the positioning invalid. Figure 3 The UWB tag trajectory prediction diagram is as follows: when the UWB tag is blocked by an obstacle column, the connection between the UWB tag and each UWB anchor point of the vehicle is interrupted, and the UWB positioning becomes invalid. At this time, the prediction position calculation model is started to calculate the predicted trajectory of the UWB tag until the UWB positioning becomes restored. Then, the prediction position calculation model is exited. In this embodiment, the UWB tag equipped with the digital key software includes a smart phone, and the velocity vector is ,in, For the phone's orientation, It can be accessed through the smartphone operating system; , is the mobile phone moving speed, It can be obtained through the smartphone operating system. By combining the predicted position calculation model with parameters such as the velocity vector, historical trajectory set, and vehicle orientation, the predicted position can be made more accurate. Of course, the above-mentioned UWB tag equipped with the digital key software can also be other mobile terminals, etc., and this application is not limited to this.

[0031] Furthermore, in one example, the velocity vector, the historical trajectory set, and the orientation information are input into a position prediction calculation model to calculate the predicted position of the UWB tag, including: The velocity vector, the historical trajectory set and the orientation information are input into the position prediction calculation model to calculate the predicted initial position; The predicted position is obtained by combining the Kalman filter algorithm with the predicted initial position calculation.

[0032] Specifically, the Kalman filter algorithm is an algorithm used to estimate the state of the system. By fusing sensor measurements and system models, the system state can be estimated more accurately, and a more reliable estimation result can be provided even in the presence of noise. The Kalman filter algorithm can effectively suppress the noise of sensor measurements during the calculation of the predicted position, improve the accuracy and stability of the state estimation, and thus reduce the occurrence of failure or invalidity of the digital key function.

[0033] Further, in one example, determining whether the UWB positioning information of the UWB tag is invalid includes: Measure the current distance information between the UWB tag and multiple UWB anchor points; If the valid current distance information cannot be obtained, or the current distance information obtained is displaced beyond the set range threshold compared with the historical distance information obtained last time, it is determined that the UWB positioning information of the UWB tag is invalid.

[0034] Specifically, the current distance information between the UWB tag and multiple UWB anchor points of the vehicle is measured, that is, the distance between the UWB tag and multiple UWB anchor points of the vehicle is measured using ultra-wideband signals. UWB technology has the characteristics of large bandwidth, strong anti-interference ability, and strong penetration ability. If the effective current distance information cannot be obtained, it means that there is an obstacle between the UWB tag and each UWB anchor point of the vehicle, and the UBW positioning fails. If the displacement of the current distance information obtained compared with the historical distance information obtained last time exceeds the set range threshold, it means that the distance change between the UWB tag and the vehicle exceeds the expected range of the system, and the UBW positioning fails.

[0035] Furthermore, in one example, the historical track set includes: in, is a collection of historical trajectories, is the historical trajectory point, the historical trajectory set At least the last five valid track positions of the UWB tag are saved; The predicted initial position is calculated, including: The predicted initial position is calculated by the prediction formula, where the prediction formula is as follows: in, To predict the initial position; For the UWB tag in the UWB tag coordinate system, The value of the direction; For the UWB tag in the UWB tag coordinate system, The value of the direction; It is the UWB tag orientation information; is the moving speed of the UWB tag; is the direction information, and the direction information is the last valid direction information of the vehicle before parking; Corresponding historical trajectory points degree of relationship; A historical track point In the UWB tag coordinate system The value of the direction; A historical track point In the UWB tag coordinate system The value of the direction.

[0036] Specifically, in this embodiment, Figure 3 The UWB tag trajectory prediction diagram shown in the figure, the walking position of the UWB tag will form a trajectory, the historical trajectory collection The last five valid trajectory positions of the UWB tag are saved. When the predicted position calculation model calculates the predicted position, the historical trajectory set will be updated and the new trajectory position will replace , the original will be deleted, the original , , Will be updated to , , . Historical track collection The prediction position calculation model can more accurately analyze the movement trend and trajectory changes of the UWB tag. The UWB anchor points of the vehicle are installed at different positions of the vehicle. By understanding the last effective orientation of the vehicle before parking, the prediction position calculation model can better understand the movement state and direction between the UWB tag and the vehicle. Because the UWB tag eventually enters the UWB coordinate system to participate in the calculation, the orientation of the UWB tag is used when calculating the movement deviation of the UWB tag. The absolute value of the vehicle direction needs to be subtracted , so that the UWB tag faces The deviation is relative to the UWB coordinate system and is calculated as follows Axis and The axis deviation can be added or subtracted from the historical trajectory. Represents the corresponding historical point The relationship degree, where the coefficient Expressed as a percentage; coefficient The coefficient can be determined based on empirical values ​​or actual calibration, for example: for , which means calculating the current and The point relationship is ,and The point relationship is ,and The point relationship is ,and The point relationship is ,and The point relationship is , The sum is .

[0037] Furthermore, the predicted position is obtained by combining the predicted initial position with the Kalman filter algorithm, including: The Kalman prediction position is calculated based on the historical trajectory set, the initialization formula and the Kalman prediction formula; The Kalman predicted position is corrected according to the update formula and the predicted initial position to obtain the predicted position.

[0038] Specifically, in the Kalman filter algorithm, the initialization phase is used to set the initial state and covariance matrix of the system. The initialization formula can provide a starting point for the subsequent prediction and update steps, ensuring that the algorithm can start state estimation from a known state. In the Kalman prediction formula, according to the dynamic model and control input of the system, the state estimate and covariance matrix of the previous moment are used to predict the state and covariance of the system at the next moment. The Kalman prediction formula is a process of state estimation based on the system model, which can help the algorithm predict the state change trend of the system. In the update formula, the predicted state estimate is fused with the actual sensor measurement value, and the weight between the predicted state and the measurement value is adjusted by the Kalman gain to obtain the final state estimate and covariance matrix. The update formula is the process of correcting the state estimate by fusing the sensor measurement value and the predicted state to improve the accuracy and stability of the state estimate. These three formulas work together to form the core process of the Kalman filter algorithm. Initialization provides a starting point, prediction is based on the system model for state estimation, and update corrects the estimation result by fusing the measurement value and the predicted state, so that the algorithm can dynamically estimate the state of the system and adapt to the changing environment, thereby improving the accuracy and stability of the state estimate.

[0039] Furthermore, the initialization formula includes a Kalman initialization formula, and the Kalman initialization formula is as follows: in, is the state error covariance matrix: is the process noise covariance; is the measurement noise covariance; is the measurement matrix.

[0040] Specifically, by initializing parameters such as the state error covariance matrix, process noise covariance, and measurement noise covariance, the Kalman initialization formula can more accurately estimate the state of the UWB tag, thereby improving the accuracy and stability of position prediction.

[0041] Furthermore, the Kalman prediction formula includes: .

[0042] Specifically, when the state is predicted by the Kalman prediction formula, the previous state prediction value is retained, and the state is updated according to the process noise covariance, making the state prediction more stable and continuous, and reducing the accumulation of prediction errors.

[0043] Furthermore, the update formula includes a Kalman update formula, which is as follows: .

[0044] Specifically, the Kalman update formula can be dynamically adjusted according to the measurement noise and state error covariance, achieving effective fusion of measurement data and prediction data.

[0045] Furthermore, the location prediction calculation model will periodically calculate the predicted location of the UWB tag.

[0046] Specifically, in this embodiment, after entering the predicted position calculation model, it is necessary to periodically calculate ,Default cycle for , which can be adjusted according to the actual deployment. represents one cycle per second. Here, ' ' is the unit of frequency, which represents the number of cycles that occur per second.

[0047] Furthermore, the predicted position of the UWB tag is continuously calculated based on the position prediction calculation model and the historical trajectory set is updated until the UWB positioning information of the UWB tag is restored to validity.

[0048] Specifically, Figure 4 The calculation update mechanism diagram of the predicted position calculation model shown in the figure, the predicted position calculation model will continue to calculate the predicted position of the UWB tag until the UWB positioning information becomes valid again.

[0049] Specifically, the algorithm formula in this embodiment can be extended and applied to the BLE models and other positioning models in other embodiments.

[0050] The present application also provides a position prediction device based on an ultra-wideband UWB digital key, and the position prediction device based on an ultra-wideband UWB digital key implements a position prediction method based on an ultra-wideband UWB digital key.

[0051] Specifically, the position prediction device based on ultra-wideband UWB digital key of the present application includes the position prediction method based on ultra-wideband UWB digital key of any technical solution of the present application. Therefore, it has all the beneficial effects of the position prediction method based on ultra-wideband UWB digital key of any technical solution of the present application, which will not be repeated here.

[0052] Although the present application is disclosed as above, the present application is not limited thereto. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the present application, so the protection scope of the present application shall be subject to the scope defined by the claims.

Claims

1. A location prediction method based on an ultra-wideband (UWB) digital key, applied to a location prediction system of the UWB digital key, wherein the location prediction system comprises a UWB tag and a plurality of UWB anchor points arranged on a vehicle, characterized in that: The position prediction method comprises: Determining whether the UWB positioning information of the UWB tag is invalid; When the UWB positioning information fails, obtaining the velocity vector and historical trajectory set of the UWB tag and the orientation information of the vehicle; The velocity vector, the historical trajectory set, and the orientation information are input into a position prediction calculation model to calculate the predicted position of the UWB tag.

2. The location prediction method based on ultra-wideband UWB digital key according to claim 1 is characterized in that: The step of inputting the velocity vector, the historical trajectory set, and the orientation information into a position prediction calculation model to calculate the predicted position of the UWB tag includes: Inputting the velocity vector, the historical trajectory set and the orientation information into the position prediction calculation model to calculate and obtain a predicted initial position; The predicted position is calculated by combining the predicted initial position with the Kalman filter algorithm.

3. The location prediction method based on ultra-wideband UWB digital key according to claim 1 is characterized in that: The determining whether the UWB positioning information of the UWB tag is invalid includes: Measuring current distance information between the UWB tag and the plurality of UWB anchor points; If the valid current distance information cannot be obtained, or the current distance information obtained is displaced beyond a set range threshold compared with the historical distance information obtained last time, it is determined that the UWB positioning information of the UWB tag is invalid.

4. The location prediction method based on ultra-wideband UWB digital key according to claim 2 is characterized in that: The historical track set includes: in, is a collection of historical trajectories, is a historical trajectory point, the historical trajectory set Saving at least the last five valid track positions of the UWB tag; The calculation to obtain the predicted initial position includes: The predicted initial position is calculated by a prediction formula, wherein the prediction formula is as follows: in, is the predicted initial position; is the UWB tag in the UWB tag coordinate system, The value of the direction; is the UWB tag in the UWB tag coordinate system, The value of the direction; It is the UWB tag orientation information; is the moving speed of the UWB tag; is the orientation information, and the orientation information is the last valid orientation information of the vehicle before parking; Corresponding to the historical trajectory point degree of relationship; A historical track point In the UWB tag coordinate system The value of the direction; A historical track point In the UWB tag coordinate system The value of the direction.

5. The location prediction method based on ultra-wideband UWB digital key according to claim 4 is characterized in that: The step of calculating the predicted position by combining the predicted initial position with the Kalman filter algorithm includes: The Kalman predicted position is calculated according to the historical trajectory set, the initialization formula and the Kalman prediction formula; The Kalman predicted position is corrected according to an update formula and the predicted initial position to obtain the predicted position.

6. The location prediction method based on ultra-wideband UWB digital key according to claim 5 is characterized in that: The initialization formula includes a Kalman initialization formula, and the Kalman initialization formula is as follows: in, is the state error covariance matrix: is the process noise covariance; is the measurement noise covariance; is the measurement matrix.

7. The location prediction method based on ultra-wideband UWB digital key according to claim 6 is characterized in that: The Kalman prediction formula includes: 。 8. The location prediction method based on ultra-wideband UWB digital key according to claim 7 is characterized in that: The update formula includes a Kalman update formula, which is as follows: 。 9. The location prediction method based on ultra-wideband (UWB) digital key according to any one of claims 1 to 8, characterized in that: The position prediction calculation model will periodically calculate the predicted position of the UWB tag.

10. A location prediction device based on ultra-wideband UWB digital key, characterized in that: The position prediction device based on ultra-wideband UWB digital key implements the position prediction method based on ultra-wideband UWB digital key as described in claims 1-9.

Images