An inertial-assisted GNSS positioning method suitable for vehicle-mounted smartphone platforms

Abstract

The present invention discloses an inertial assisted GNSS positioning method suitable for a vehicle-mounted smartphone platform. The MEMS-IMU module provides celestial gyroscope data and forward accelerometer data, performs single-gyro and single-accelerometer dead reckoning, and then outputs it to the EKF calibration solution filter as filter state estimation information; the GNSS module provides GNSS speed and GNSS position, combines the relative altitude data provided by the mobile phone MEMS-barometer to perform GNSS confidence assessment, and then outputs it to the EKF calibration solution filter as the first set of observation data of the EKF filter; the wheel speed encoder module obtains wheel speed data through the car and mobile phone interconnection technology, and then outputs it to the EKF calibration solution filter as the second set of observation data of the EKF filter; finally, the EKF calibration solution filter outputs the fusion positioning result. This method first proposes a single-gyroscope and single-accelerometer dead reckoning arrangement algorithm suitable for the vehicle-mounted bracket scenario, which can effectively solve the problem of mobile phone bracket posture adjustment.

Description

Technical Field

[0001] The present invention relates to technologies related to GNSS positioning of smart phones, and in particular to an inertial assisted GNSS positioning method suitable for a vehicle-mounted smart phone platform. Background Art

[0002] Currently, smartphone users are accustomed to using mobile navigation functions in car scenarios. However, GNSS signals are easily blocked by tall buildings, tree shadows, overpasses, etc., and GNSS signals cannot be received at all in tunnels. Therefore, mobile phone GNSS positioning alone often cannot output accurate positioning information in the above scenarios, seriously affecting the user experience.

[0003] When users use the navigation and positioning function in a car, the smartphone is generally fixed on the car's center console using a car phone holder. Since the car phone holder can adjust the posture angle to a certain extent according to the driver's user habits, the built-in IMU module in the smartphone is not a completely fixed base. In the traditional six-degree-of-freedom IMU / GNSS fusion algorithm, if the user adjusts the holder's posture while the vehicle is driving, the fusion solution attitude angle will take a long time to converge, thus affecting the overall fusion positioning effect. Summary of the Invention

[0004] In view of the problem that when smartphones are used for in-vehicle navigation, GNSS signals are easily blocked by tall buildings, tree shades, overpasses, tunnels, etc., resulting in inaccurate positioning and affecting user experience, the present invention provides an inertial-assisted GNSS positioning method suitable for in-vehicle smartphone platforms.

[0005] An inertial-aided GNSS positioning method suitable for an in-vehicle smartphone platform. The MEMS-IMU module provides azimuth gyroscope data and forward accelerometer data for single-gyro and single-accelerometer dead reckoning, which is then output to an EKF calibration filter as filter state estimation information. The GNSS module provides GNSS velocity and GNSS position, which are combined with relative altitude data provided by the smartphone's MEMS-barometer for GNSS confidence assessment. The data is then output to the EKF calibration filter as the first set of observation data for the EKF filter. The wheel speed encoder module obtains wheel speed data through vehicle-mobile phone interconnection technology, which is then output to the EKF calibration filter as the second set of observation data for the EKF filter. Finally, the EKF calibration filter outputs a fused positioning result.

[0006] The MEMS-IMU module uses the forward MEMS accelerometer data and the azimuth MEMS gyroscope data after the mobile phone is installed and fixed on the bracket. Based on the forward MEMS accelerometer data and the azimuth MEMS gyroscope data, the attitude dead reckoning equation is compiled as the state prediction equation of the EKF filter. When the installation method of the mobile phone on the bracket is switched between landscape and portrait modes, the accelerometer is used to measure the direction of gravity and determine the azimuth gyroscope among the three axes.

[0007] The GNSS module provides position and velocity data as the first set of observation data for EKF filtering. For smartphones with built-in MEMS barometers, the weight of GNSS data in fusion positioning is adaptively adjusted by comparing the altitude data output by the smartphone's built-in MEMS barometer with the GNSS altitude data, thereby improving the final positioning effect. For smartphones without MEMS barometers, the weight in EKF fusion is determined by the number of satellites and precision coefficient information output by the GNSS module.

[0008] The wheel speed encoder module provides vehicle wheel speed data as the second set of observation data for the EKF filter, and the vehicle wheel speed data is obtained through mobile phone and vehicle interconnection technology.

[0009] The method described, the steps are as follows:

[0010] Step (1) The smartphone is fixedly mounted on the vehicle bracket in an arbitrary initial posture, the navigation and positioning function is turned on, and the GNSS module of the smartphone enters the GNSS positioning solution function and outputs the GNSS solution position and speed;

[0011] Step (2) Initialize the single gyroscope / single accelerometer dead reckoning state according to the GNSS output positioning result;

[0012] Step (3) determines whether the IMU data has arrived. If the IMU data has arrived, the celestial gyroscope and the forward accelerometer in the three axes are determined by comparing the accelerometer data with the gravity data.

[0013] Step (4) performing dead reckoning based on the gyroscopic data and the accelerometer data;

[0014] Step (5) determines whether the GNSS data has arrived. If the GNSS data has arrived, the GNSS position and velocity are used as observations, and optimal fusion is performed based on the EKF filter and the single gyro / single accelerometer dead reckoning results. The confidence of the GNSS data fusion is calculated by comparing with the barometer or by calculating the number of GNSS satellites and the precision factor information.

[0015] Step (6) determines whether the wheel speed data has arrived. If the wheel speed data has arrived, the wheel speed is used as the observation and optimal fusion is performed based on the EKF filter and the single gyro / single accelerometer dead reckoning result;

[0016] Step (7) If the mobile phone navigation and positioning function continues to be used, repeat steps (3) to (6). If the mobile phone navigation and positioning function is turned off, end.

[0017] Beneficial effects of the present invention:

[0018] This method proposes an inertial-aided GNSS positioning method suitable for in-vehicle smartphone platforms. Through the targeted design of a single-gyroscope and single-accelerometer dead reckoning algorithm arrangement, the EKF algorithm can effectively fuse the mobile phone GNSS data, mobile phone MEMS-IMU data, vehicle wheel speed data, and mobile phone MEMS barometer data. The obtained fused positioning result is far superior to pure GNSS positioning, effectively improving the user experience of mobile phone navigation and positioning functions in in-vehicle scenarios.

[0019] This method first proposes a single-gyroscope and single-accelerometer dead reckoning arrangement algorithm suitable for vehicle-mounted bracket scenarios, which can effectively solve the problem of mobile phone bracket posture adjustment.

[0020] With the maturity of car-mobile connectivity, it's now possible to obtain wheel speed data through connectivity technologies like Bluetooth and Wi-Fi. This method, building on the fusion of single-gyro and single-accelerometer / GNSS, further proposes a fusion solution that uses wheel speed data acquired through vehicle-connected technologies to assist smartphone positioning performance. This solution effectively improves overall fusion positioning, especially in scenarios where GNSS signals are completely blocked, such as tunnels. Furthermore, for some smartphones equipped with MEMS barometers, this method designs a solution that uses the barometer to assist in GNSS positioning. By comparing the altitude data output by the barometric altimeter with the GNSS altitude data, the weight of the GNSS data in the fusion positioning is adaptively adjusted, thereby improving the final positioning effect.

[0021] It is worth pointing out that not all technical solutions of the present invention can achieve all the above technical effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a diagram of an algorithm scheme for inertial-assisted GNSS positioning suitable for vehicle-mounted smartphone platforms.

[0023] Figure 2 A flowchart of the steps for inertial-aided GNSS positioning suitable for an in-vehicle smartphone platform. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention is further described in detail below based on the figures, combined with specific embodiments and related formula derivations.

[0025] like Figure 1As shown in the figure, an inertial-aided GNSS positioning method suitable for an in-vehicle smartphone platform is shown. The MEMS-IMU module provides azimuth gyroscope data and forward accelerometer data for single-gyroscope and single-accelerometer dead reckoning, which is then output to the EKF calibration and solution filter as filter state estimation information; the GNSS module provides GNSS velocity and GNSS position, which are combined with the relative altitude data provided by the mobile phone MEMS-barometer for GNSS confidence assessment, and then output to the EKF calibration and solution filter as the first set of observation data of the EKF filter; the wheel speed encoder module obtains wheel speed data through the car and mobile phone interconnection technology, and then outputs it to the EKF calibration and solution filter as the second set of observation data of the EKF filter; finally, the EKF calibration and solution filter outputs the fused positioning result.

[0026] In this example, the MEMS-IMU module uses only the forward MEMS accelerometer data and the azimuth MEMS gyroscope data after the phone is mounted on the bracket. When the phone is mounted on the bracket in landscape and portrait modes, the accelerometer measures the direction of gravity and determines the azimuth gyroscope in the three axes. Based on the forward accelerometer data and the azimuth gyroscope data, the attitude dead reckoning equation is compiled as the state estimation equation for the EKF filter. The specific dead reckoning state equation is designed as follows:

[0027]

[0028] in f is the heading angle, V is the vehicle's forward speed, X, Y For the east position and the north position, W z is the celestial gyroscope data, f y is the forward accelerometer data, D w is the gyro noise, D f is the accelerometer noise, E abs is the wheel speed scaling factor.

[0029] State A reflects the combined effect of gyro installation tilt and scale factor error and is a constant to be estimated;

[0030] State B reflects the combined effects of gyro bias, installation tilt, and scale factor, and is a constant to be estimated.

[0031] State C reflects the combined effect of the accelerometer installation tilt and scale factor, which is a constant to be estimated;

[0032] State D reflects the combined effects of the accelerometer bias, installation tilt, and scale factor, and is a constant to be estimated.

[0033] In the single-gyro / single-accelerometer dead reckoning equation design involved in this method, the IMU installation tilt, zero bias, scale factor, and other estimated constants are comprehensively expressed using four constant states: A, B, C, and D. Compared with the six-degree-of-freedom strapdown solution equation, it has a faster convergence speed during the EKF fusion process of GNSS position and velocity observations, so it can quickly estimate and eliminate the impact of installation tilt after the mobile phone holder is moved.

[0034] In this example, the GNSS module provides position and velocity data as the first set of observation data for EKF filtering. For some smart phones with built-in MEMS barometers, the altitude data output by the smartphone's built-in MEMS barometer is compared with the GNSS altitude data to calculate the information difference, and the weight of the GNSS data in the fusion positioning is adaptively adjusted, thereby improving the final positioning effect. If the smartphone does not have a MEMS barometer, the weight in the EKF fusion is determined by the number of satellites, precision factor, and other information output by the GNSS module. In the EKF-based single gyro and single accelerometer / GNSS fusion algorithm, the GNSS observation equation is as follows:

[0035]

[0036] Among them, X gnss Y gnss is the GNSS easting position and northing position observation, V x_gnss Y y_gnss is the GNSS easting speed and northing speed, Δx , Δy , Δv x , Δv y is the GNSS position noise and velocity noise.

[0037] In this example, the wheel speed encoder module provides vehicle wheel speed data as the second set of observation data for the EKF filter. Given the current maturity of mobile phone and car interconnection technology, it is now possible for mobile phones to obtain vehicle wheel speed data through car networking. This method proposes using the wheel speed data obtained from mobile phone and car networking as one of the effective information for auxiliary GNSS positioning. Based on the EKF algorithm, it is effectively integrated with IMU data and GNSS data to improve the final positioning effect. In the EKF fusion algorithm, the wheel speed observation equation is as follows:

[0038]

[0039] in, V abs is the wheel speed data, E absis the wheel speed scaling factor, Δv abs The noise is measured for wheel speed.

[0040] The principle diagram of the single gyro single accelerometer / GNSS / wheel speed / air pressure data fusion positioning solution based on the EKF algorithm framework is as follows: Figure 1 As shown, the EKF algorithm process involved is as follows:

[0041] Assume that the state equation and measurement equation of the combined positioning system after linear discretization are:

[0042] Equation of state:

[0043] Measurement equation:

[0044] State noise variance matrix and measurement noise variance matrix:

[0045] After the system is discretized, the system matrix can be obtained: System noise driving matrix: , measurement matrix: , combined with the state noise matrix and the measurement noise matrix, the Kalman filter algorithm process is as follows:

[0046] Status estimation:

[0047] Covariance estimation:

[0048] Optimal gain calculation:

[0049] Optimal state estimate:

[0050] Optimal covariance estimate: .

[0051] This example involves specific steps such as Figure 2 As shown, the steps are as follows:

[0052] Step (1) The smartphone is fixedly mounted on the vehicle bracket in an arbitrary initial posture, the navigation and positioning function is turned on, and the GNSS module of the smartphone enters the GNSS positioning solution function and outputs the GNSS solution position and speed;

[0053] Step (2) Initialize the single gyroscope / single accelerometer dead reckoning state according to the GNSS output positioning result;

[0054] Step (3) determines whether the IMU data has arrived. If the IMU data has arrived, the celestial gyroscope and the forward accelerometer in the three axes are determined by comparing the accelerometer data with the gravity data.

[0055] Step (4) performing dead reckoning based on the gyroscopic data and the accelerometer data;

[0056] Step (5) determines whether the GNSS data has arrived. If the GNSS data has arrived, the GNSS position and velocity are used as observations, and optimal fusion is performed based on the EKF filter and the single gyro / single accelerometer dead reckoning results. The confidence of the GNSS data fusion is calculated by comparing with the barometer or by calculating the number of GNSS satellites, precision factor, and other information.

[0057] Step (6) determines whether the wheel speed data has arrived. If the wheel speed data has arrived, the wheel speed is used as the observation and optimal fusion is performed based on the EKF filter and the single gyro / single accelerometer dead reckoning result;

[0058] Step (7) If the mobile phone navigation and positioning function continues to be used, repeat steps (3) to (6). If the mobile phone navigation and positioning function is turned off, end.

[0059] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by any technician familiar with this technical field within the technical scope disclosed by the present invention should be covered by the scope of protection of the present invention.

Claims

1. An inertial-assisted GNSS positioning method suitable for a vehicle-mounted smartphone platform, characterized by: The MEMS-IMU module provides gyroscope data and accelerometer data to perform single-gyro and single-accelerometer dead reckoning, which is then output to the EKF calibration filter as filter state estimation information. The GNSS module provides GNSS velocity and GNSS position, which are combined with the relative altitude data provided by the phone's MEMS-barometer to perform GNSS confidence assessment. The data is then output to the EKF calibration filter as the first set of observation data for the EKF filter. The wheel speed encoder module obtains wheel speed data through the car and mobile phone interconnection technology, and then outputs it to the EKF calibration and solution filter as the second set of observation data of the EKF filter; finally, the EKF calibration and solution filter outputs the fusion positioning result; The MEMS-IMU module uses the forward MEMS accelerometer data and the azimuth MEMS gyroscope data of the mobile phone after it is mounted on the bracket. Based on the forward MEMS accelerometer data and the azimuth MEMS gyroscope data, the attitude dead reckoning equation is compiled as the state prediction equation of the EKF filter. When the mounting method of the mobile phone on the bracket is switched between landscape and portrait modes, the accelerometer measures the direction of gravity to determine the azimuth gyroscope among the three axes. The GNSS module provides position and velocity data as the first set of observation data for EKF filtering. For smartphones with built-in MEMS barometers, the weight of GNSS data in fusion positioning is adaptively adjusted by comparing the altitude data output by the smartphone's MEMS barometer with the GNSS altitude data, thereby improving the final positioning effect. For smartphones without MEMS barometers, the weight in EKF fusion is determined by the number of satellites and precision coefficient information output by the GNSS module. The wheel speed encoder module provides vehicle wheel speed data as the second set of observation data for the EKF filter, and the vehicle wheel speed data is obtained through mobile phone and vehicle interconnection technology; Here are the steps: Step (1) The smartphone is fixedly mounted on the vehicle bracket in an arbitrary initial posture, the navigation and positioning function is turned on, and the GNSS module of the smartphone enters the GNSS positioning solution function and outputs the GNSS solution position and speed; Step (2) Initialize the single gyroscope / single accelerometer dead reckoning state according to the GNSS output positioning result; Step (3) determines whether the IMU data has arrived. If the IMU data has arrived, the celestial gyroscope and the forward accelerometer in the three axes are determined by comparing the accelerometer data with the gravity data. Step (4) performing dead reckoning based on the azimuth gyroscope data and the forward accelerometer data; Step (5) determines whether the GNSS data has arrived. If the GNSS data has arrived, the GNSS position and velocity are used as observations, and optimal fusion is performed based on the EKF filter and the single gyro / single accelerometer dead reckoning results. The confidence of the GNSS data fusion is calculated by comparing with the barometer or by calculating the number of GNSS satellites and the precision factor information. Step (6) determines whether the wheel speed data has arrived. If the wheel speed data has arrived, the wheel speed is used as the observation and optimal fusion is performed based on the EKF filter and the single gyro / single accelerometer dead reckoning result; Step (7) If the mobile phone navigation and positioning function continues to be used, repeat steps (3) to (6). If the mobile phone navigation and positioning function is turned off, end.

Images