Magnetic-inertial fusion carrier speed estimation method, device and system

Abstract

The invention discloses a magnetic-inertial fusion carrier speed estimation method. The method comprises the following steps: extracting sensor data of a front sensor and a rear sensor in a preset time window; the sensor data comprises a magnetic induction intensity sequence and acceleration information; carrying out traversal search on the carrier speed in a preset speed range, and calculating a plurality of displacement sequences of front and rear sensors by combining acceleration information; the preset speed range comprises a plurality of candidate carrier speeds; combining each displacement sequence with the corresponding magnetic induction intensity sequence to obtain a plurality of original displacement-magnetic field sequence pairs of the front sensor and the rear sensor; performing equidistant resampling on each original displacement-magnetic field sequence pair to obtain a plurality of new displacement-magnetic field sequence pairs of the front sensor and the rear sensor; and performing similarity calculation on the two new displacement-magnetic field sequence pairs corresponding to the same candidate carrier speed, and determining the candidate carrier speed corresponding to the displacement-magnetic field sequence pair with the maximum similarity as the carrier speed.

Description

Technical Field

[0001] The present invention relates to the field of speed measurement, and particularly to a method, device and system for estimating the speed of a carrier by magnetic-inertial fusion. Background Art

[0002] The Global Navigation Satellite System (GNSS) is the main means to obtain information such as speed and position. However, in scenarios where GNSS signals are missing, such as tunnels, underground parking lots, and urban canyons, it is difficult for a carrier to achieve continuous and reliable speed measurement through satellite signals. At this time, inertial sensors are usually used as an alternative solution. However, low-cost inertial sensors are difficult to independently provide long-term stable speed estimation due to large cumulative errors and significant drifts. Especially in complex dynamic environments, the limitations of a single sensor are further highlighted, and there is an urgent need for a robust speed measurement method based on low-cost sensors to fill the data gap when GNSS fails.

[0003] In existing odometer technologies, the information of wheel odometers cannot be accessed by mobile phones or other mobile terminals, and thus cannot be used by mass navigation and positioning devices. Although visual odometers and lidar odometers can estimate speed by matching environmental features, they require high computing resources, and the performance of visual odometers drops significantly under insufficient light, sparse features, or bad weather conditions. These methods are difficult to adapt to low-cost and low-power sensors.

[0004] In man-made buildings, a large number of steel structures will cause the distortion of the geomagnetic field, forming unique magnetic field characteristics at each location. Based on this characteristic, there are currently two methods for measuring the speed of a carrier using magnetic sensors. The first is the speed measurement method based on a multi-magnetic sensor array. According to Maxwell's equations, the speed is estimated through indoor magnetic field characteristics without the need to pre-establish magnetic map information. However, this method requires a sensor array that is precisely installed and calibrated, with complex operations and difficult to achieve mass application. The second is the dual-magnetic sensor measurement method, in which two sensors are respectively placed at the head and tail of the carrier, and the speed is estimated by matching the magnetic field time series. Although this method avoids the problem of sensor calibration, it has extremely high requirements for the matching accuracy of the magnetic field time series. Existing methods are only applicable to low-speed moving carriers such as pedestrians or robots. For high-speed moving carriers such as vehicles, the existing matching accuracy of the magnetic field time series drops significantly, and high-precision carrier speed estimation cannot be obtained. Summary of the Invention

[0005] To overcome the deficiencies of the above-mentioned prior art, based on an inertial measurement unit and a magnetometer, the present invention proposes a method, device and system for estimating the speed of a carrier by magnetic-inertial fusion, which are low-cost, do not require prior calibration, and are applicable to high-speed motion scenarios.

[0006] According to one aspect of the specification of the present invention, the present invention provides a method for estimating the carrier speed by magnetic-inertial fusion, including: extracting the sensor data of the front sensor and the rear sensor within a preset time window; wherein, the front sensor and the rear sensor are respectively installed at the front and rear of the carrier and are both nine-axis sensors, and the sensor data includes a magnetic induction intensity sequence and acceleration information; performing a traversal search on the carrier speed within a preset speed range, and calculating a plurality of displacement sequences of the front sensor and the rear sensor respectively in combination with the acceleration information; wherein, the preset speed range includes a plurality of candidate carrier speeds; combining each displacement sequence with the corresponding magnetic induction intensity sequence to obtain a plurality of original displacement-magnetic field sequence pairs of the front sensor and the rear sensor respectively; performing equidistant resampling on each original displacement-magnetic field sequence pair to obtain a plurality of new displacement-magnetic field sequence pairs of the front sensor and the rear sensor respectively; calculating the similarity between the new displacement-magnetic field sequence pairs of the front sensor and the new displacement-magnetic field sequence pairs of the rear sensor corresponding to the same candidate carrier speed, and determining the candidate carrier speed corresponding to the displacement-magnetic field sequence pair with the maximum similarity as the carrier speed.

[0007] Further, it also includes: projecting the magnetic induction intensity sequence and the acceleration information into the carrier coordinate system.

[0008] Further, performing a traversal search on the carrier speed within a preset speed range and calculating a plurality of displacement sequences of the front sensor and the rear sensor respectively in combination with the acceleration information includes: calculating the speed change sequence of the carrier within the preset time window based on the acceleration information collected by the front sensor; calculating the absolute speed sequence corresponding to each candidate carrier speed in combination with the speed change sequence; wherein, a plurality of candidate carrier speeds correspond to a plurality of absolute speed sequences; calculating a plurality of displacement sequences of the front sensor based on the plurality of absolute speed sequences; calculating a plurality of displacement sequences of the rear sensor based on the plurality of displacement sequences of the front sensor and the distance between the two sensors.

[0009] Further, calculating the speed change sequence of the carrier within the preset time window based on the acceleration information collected by the front sensor includes: calculating the acceleration of the carrier at each moment within the preset time window based on the acceleration information collected by the front sensor; calculating the speed change amount between each moment and the next moment of the carrier based on the acceleration; integrating the speed change amount to obtain the speed change amount from each moment to the last moment of the carrier; jointly forming the speed change sequence by the speed change amounts from each moment to the last moment within the preset time window.

[0010] Further, an absolute velocity sequence corresponding to each candidate carrier velocity is calculated in combination with the velocity change sequence; wherein, several candidate carrier velocities correspond to several absolute velocity sequences, including: combining each candidate carrier velocity with the velocity change sequence to calculate the absolute velocity corresponding to each moment within a preset time window; jointly forming the absolute velocity sequence with the absolute velocities at each moment within the preset time window; wherein, the several candidate carrier velocities correspond to several absolute velocity sequences.

[0011] Further, several displacement sequences of the front sensor are calculated based on the several absolute velocity sequences, including;

[0012] The displacement amount between each moment and the next moment of the carrier is calculated based on each absolute velocity sequence; the displacement amount is integrated to obtain the displacement amount between each moment and the last moment of the carrier; the displacement amounts between each moment and the last moment within the preset time window are jointly formed into the displacement sequence; wherein, several absolute velocity sequences correspond to several displacement sequences.

[0013] Further, each original displacement-magnetic field sequence pair is resampled at equal intervals to obtain several new displacement-magnetic field sequence pairs for the front sensor and the rear sensor respectively, including: reconstructing a set of target sampling points at equal intervals; aligning each original displacement-magnetic field sequence pair at the target sampling points; performing linear interpolation on the aligned original displacement-magnetic field sequence pairs at the target sampling points, so as to obtain several new displacement-magnetic field matching sequence pairs for the front sensor and the rear sensor respectively.

[0014] Further, the similarity between the new displacement-magnetic field sequence pair of the front sensor and the new displacement-magnetic field sequence pair of the rear sensor corresponding to the same candidate carrier velocity is calculated, including: comparing the similarity between the new displacement-magnetic field sequence pair of the front sensor and the new displacement-magnetic field sequence pair of the rear sensor corresponding to each candidate carrier velocity by using Pearson correlation coefficient, Euclidean distance, Manhattan distance, Chebyshev distance, Mahalanobis distance, cosine similarity or dynamic time warping.

[0015] According to one aspect of the specification of the present invention, the present invention provides an estimation device for the speed of a magneto-inertial fusion carrier, and the estimation device includes: a data extraction module for extracting sensor data of a front sensor and a rear sensor within a preset time window; wherein, the front sensor and the rear sensor are respectively installed at the front and rear of the carrier and are both nine-axis sensors, and the sensor data includes a magnetic induction intensity sequence and acceleration information; a displacement calculation module for performing a traversal search on the carrier speed within a preset speed range, and calculating a plurality of displacement sequences of the front sensor and the rear sensor respectively in combination with the acceleration information; wherein, the preset speed range includes a plurality of candidate carrier speeds; a first magnetic moment matching module for combining each displacement sequence with the corresponding magnetic induction intensity sequence to obtain a plurality of original displacement-magnetic field sequence pairs of the front sensor and the rear sensor respectively; a second magnetic moment matching module for performing equidistant resampling on each original displacement-magnetic field sequence pair to obtain a plurality of new displacement-magnetic field sequence pairs of the front sensor and the rear sensor respectively; a carrier speed determination module for calculating the similarity between the new displacement-magnetic field sequence pair of the front sensor and the new displacement-magnetic field sequence pair of the rear sensor corresponding to the same candidate carrier speed, and determining the candidate carrier speed corresponding to the displacement-magnetic field sequence pair with the maximum similarity as the carrier speed.

[0016] According to one aspect of the specification of the present invention, the present invention provides an estimation system for the speed of a magneto-inertial fusion carrier, characterized in that the estimation system includes a front sensor, a rear sensor and a processing module, and the front sensor and the rear sensor are respectively installed at the front and rear of the carrier and are both nine-axis sensors; the processing module is used to execute the method for estimating the speed of a magneto-inertial fusion carrier according to the sensor data collected by the front and rear sensors.

[0017] The above technical solution installs nine-axis sensors at the front and rear of the carrier respectively, utilizes the characteristic that the magnetic sequences collected by the front and rear nine-axis sensors are consistent at the same position, combines the lever arm compensation and the sequence matching optimization algorithm, and searches for the optimal carrier speed solution, so as to obtain an accurate carrier speed measurement result.

[0018] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0019] (1) The present invention only calculates the speed through the internal nine-axis sensors and does not rely on external signals (such as GNSS signals). Therefore, the present invention can still obtain accurate carrier speed measurement results in GNSS signal denied environments such as tunnels, urban canyons, and underground parking lots, thus effectively overcoming the problem of reduced speed measurement accuracy caused by signal loss in traditional satellite navigation.

[0020] (2) The present invention combines the acceleration information measured by the inertial measurement unit with the magnetic field information measured by the magnetometer, and can timely reflect the change of the carrier speed even in the case of high-speed movement. Therefore, the present invention can be used for measuring the carrier speed in high-speed movement scenarios.

[0021] (3) Compared with the solutions relying on external infrastructures, the present invention does not rely on a sensor array that requires precise installation and calibration, and only two low-cost nine-axis sensors are needed to achieve autonomous speed measurement. It has the advantages of convenient deployment, low cost, strong robustness, etc., and significantly improves the navigation reliability in complex environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings used in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0023] Figure 1 It is a flowchart of the carrier speed estimation method provided by the embodiment of the present invention.

[0024] Figure 2 It is the first sub-flowchart of the carrier speed estimation method provided by the embodiment of the present invention.

[0025] Figure 3 It is the second sub-flowchart of the carrier speed estimation method provided by the embodiment of the present invention.

[0026] Figure 4 It is a schematic diagram of sensor installation in vehicle applications provided by the embodiment of the present invention.

[0027] Figure 5 It is a schematic diagram of sensor installation in UAV applications provided by the embodiment of the present invention.

[0028] Figure 6 It is a schematic diagram of sensor installation in pedestrian applications provided by the embodiment of the present invention.

[0029] Figure 7 It is a schematic diagram of the structure of the estimation device provided by the embodiment of the present invention.

[0030] Figure 8 It is a schematic diagram of the structure of the estimation system provided by the embodiment of the present invention.

[0031] In the figure, 1 is the front sensor; 2 is the rear sensor; 3 is the carrier; 4 is the estimation device; 41 is the data extraction module, 42 is the displacement calculation module; 43 is the first magnetic moment matching module; 44 is the second magnetic moment matching module; 45 is the carrier speed determination module; 100 is the processing module; 1000 is the estimation system. Detailed implementation manner

[0032] The terms "including" and "having" and any variations thereof in the specification and claims of the present invention and the above-mentioned drawings are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0033] The block diagrams shown in the drawings are only functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor devices and / or microcontroller devices. The flowcharts shown in the drawings are only exemplary illustrations and do not necessarily include all the content and operations / steps, nor do they have to be executed in the described order. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to the actual situation.

[0034] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the protection scope of the present invention. In addition, the technical features in each embodiment or a single embodiment provided by the present invention can be arbitrarily combined with each other to form a new technical solution. This combination is not restricted by the order of steps and / or the structure composition mode, but must be based on the fact that those of ordinary skill in the art can implement it. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the protection scope required by the present invention.

[0035] Please refer to the attached Figure 1 , the present invention provides a method for estimating the speed of a carrier by magnetic-inertial fusion. The method for estimating the speed of a carrier includes steps S101-S109.

[0036] Step S101: Extract the sensor data of the front sensor and the rear sensor within a preset time window. Herein, the front sensor and the rear sensor are respectively installed at the front and rear of the carrier, and both are nine-axis sensors. The sensor data includes the magnetic induction intensity sequence and the acceleration information.

[0037] In step S101, both the front sensor 1 and the rear sensor 2 are nine-axis sensors. The nine-axis sensor includes a magnetometer and an inertial measurement unit. The magnetometer is used to collect the magnetic induction intensity sequence, and the inertial measurement unit includes an accelerometer which is used to collect the acceleration information. The nine-axis sensor can be in the hardware forms such as intelligent sensors, intelligent terminals, customized devices, etc. In other words, the nine-axis sensor is not limited to a single hardware form, and it can be carried on various different types of hardware devices. For example, common consumer electronic products such as smart phones and smart watches, or it can also be applied to professional fields such as customized devices according to actual needs. The types of magnetometers include but are not limited to vector magnetometers, scalar magnetometers, and gradient magnetometers. The types of carriers 3 include but are not limited to: vehicles, wheeled robots, humanoid robots, unmanned aerial vehicles, underwater vehicles, pedestrians (as shown in the appendix). Figures 4 - 6 In addition, the distance between the front sensor 1 and the rear sensor 2 is fixed. In this embodiment, the geometric center of one of the nine-axis sensors can be used as the origin, and the distance between the centers of the two nine-axis sensors can be measured and projected onto the forward direction of the carrier 3, so as to obtain the distance between the two nine-axis sensors in the carrier coordinate system, which is convenient for subsequent displacement calculation of the rear sensor 2. Among them, the distance between the centers of the two nine-axis sensors can be measured by tools such as meter rulers, tape measures, total stations, etc., and no limitation is made here.

[0038] Furthermore, the front sensor 1 is installed at the front of the carrier 3, and the rear sensor 2 is installed at the rear of the carrier 3. It can be understood that since the present invention needs to collect magnetic field data at different positions through the front-back position difference and calculate the similarity between the displacement and the magnetic field sequence to measure the carrier speed. Therefore, it is necessary to ensure that the installation positions of the two nine-axis sensors are respectively at the front and rear of the carrier 3, and no specific requirements are made for the specific installation positions of the nine-axis sensors. In this embodiment, when installing the two nine-axis sensors, it is necessary to ensure that the nine-axis sensors are tightly and fixedly connected to the installation positions to avoid the phenomenon of insecure installation. The installation methods include but are not limited to straps and tapes.

[0039] Step S102: Project the sensor data into the vehicle coordinate system. Specifically, construct a direction cosine matrix based on the installation angles of the nine-axis sensors, and realize the conversion of the sensor data from the sensor coordinate system to the vehicle coordinate system through coordinate transformation. It can be understood that the two nine-axis sensors collect sensor data in their respective sensor coordinate systems. Therefore, after extracting the sensor data of the two nine-axis sensors within a preset time window, it is necessary to project the sensor data collected by the two nine-axis sensors into the vehicle coordinate system uniformly to facilitate subsequent comprehensive processing of the sensor data of the two nine-axis sensors. Please refer to the appendix for details. Figure 2 Next, step S102 (including steps S1021 - S1023) will be further introduced.

[0040] Step S1021: Project the magnetic induction intensity sequence into the vehicle coordinate system. In this embodiment, the magnetic induction intensity sequences synchronously collected by the front and rear magnetometers within the preset time window in step S101 are respectively and and , where f represents the front sensor 1 and r represents the rear sensor 2. The formula for projecting the magnetic induction intensity sequence into the vehicle coordinate system is: , where represents the sensor coordinate system, and represents the vehicle coordinate system; and respectively represent the magnetic induction intensity sequences in the vehicle coordinate system and the sensor coordinate system, and represents the direction cosine matrix from the sensor coordinate system to the vehicle coordinate system.

[0041] Step S1023: Project the acceleration information into the vehicle coordinate system. In this embodiment, the specific force (i.e., acceleration information) synchronously collected by the front and rear accelerometers within the preset time window in step S101 is used. The formula for projecting the specific force in the sensor coordinate system into the vehicle coordinate system is: . In the formula, . Where is the specific force in the vehicle coordinate system at time , and is the accelerometer zero bias.

[0042] Step S103: Traverse and search the vehicle speed within a preset speed range, and calculate the respective displacement sequences of the front sensor and the rear sensor in combination with the acceleration information. The preset speed range includes several candidate vehicle speeds.

[0043] In step S103, based on the distance between the two nine-axis sensors in the carrier coordinate system obtained in step S101 and the acceleration information in the carrier coordinate system obtained in step S102, several displacement sequences of the front sensor and the rear sensor are calculated respectively. Please refer to the appendix for details. Figure 3 Next, step S103 (including steps S1031 - S1037) will be further introduced.

[0044] Step S1031: Calculate the speed change sequence of the carrier within a preset time window based on the acceleration information collected by the front sensor. Specifically, calculate the acceleration of the carrier at each moment within the preset time window based on the acceleration information collected by the front sensor. Calculate the speed change amount between each moment and the next moment of the carrier based on the acceleration. Integrate the speed change amounts to obtain the speed change amount from each moment of the carrier to the last moment. The speed change amounts from each moment to the last moment within the preset time window together form the speed change sequence.

[0045] In this embodiment, the acceleration of the carrier 3 at each moment within the preset time window is calculated based on the acceleration information collected by the front sensor, and the formula is: . In the formula, represents the navigation coordinate system, represents the acceleration vector in the carrier coordinate system at moment , and is the gravity vector. represents the direction cosine matrix from the navigation coordinate system to the sensor coordinate system at moment . The direction cosine matrix is obtained through the attitude solution of the inertial navigation system and is known information in the present invention. Further, the speed change amount between each moment and the next moment of the carrier is calculated based on the acceleration, and the formula is: . In the formula, is the speed change amount between moment and moment , and is the forward acceleration of the carrier at moment . Then, integrate the speed change amounts to obtain the speed change amount from each moment of the carrier 3 to the last moment, and the formula is , that is, it represents the total speed change amount from moment to the last moment of the time window. It can be understood that the total speed change amounts from each moment within the preset time window

[0046] to the last moment of the time window together constitute the speed change sequence of the carrier 3.Step S1033: Calculate the absolute velocity sequence corresponding to each candidate carrier velocity in combination with the velocity change sequence; among them, several candidate carrier velocities correspond to several absolute velocity sequences. Specifically, combine each candidate carrier velocity with the velocity change sequence, and calculate the absolute velocity corresponding to each moment within a preset time window. The absolute velocities at each moment within the preset time window together form the absolute velocity sequence. Among them, several candidate carrier velocities correspond to several absolute velocity sequences.

[0047] In this embodiment, the carrier velocity refers to the velocity of carrier 3 at the last moment of the preset time window Initialize a group of candidate carrier velocities , for each candidate carrier velocity , calculate the absolute velocity sequence of carrier 3 within the preset time window starting from the current moment. The formula is: . In the formula, and are respectively the velocity at the moment and the velocity at the last moment of the time window , is the velocity change amount between the moment and the moment. It can be understood that knowing the total velocity change amount between the moment and the last moment of the time window , and also knowing the velocity at the last moment , the velocity at any moment can be obtained. The velocities at each moment within the preset time window together constitute an absolute velocity sequence of carrier 3. Since the velocity at the last moment of the present invention is a group of candidate carrier velocities , the several absolute velocity sequences calculated in step S1033 are those of the front sensor.

[0048] Step S1035: Calculate several displacement sequences of the front sensor based on several absolute velocity sequences. Specifically, calculate the displacement amount between each moment and the next moment of the carrier based on each absolute velocity sequence. Integrate the displacement amount to obtain the displacement amount between each moment and the last moment of the carrier. The displacement amounts between each moment and the last moment within the preset time window together form the displacement sequence. Among them, several absolute velocity sequences correspond to several displacement sequences.

[0049] In this embodiment, calculate the displacement amount between each moment and the next moment of the carrier based on each absolute velocity sequence, and integrate the displacement amount to obtain the displacement amount between each moment and the last moment of the carrier. The formula is: . In the formula, is the front sensor The displacement at a certain moment compared to the last moment of the time window. Understandably, by from a certain moment to the last moment accumulating the displacements at each moment in between, the displacement amount of a certain moment compared to the last moment can be obtained, and the displacement amounts of each moment within the preset time window compared to the last moment together constitute a displacement sequence of the carrier 3. Since the step S1033 calculates several absolute velocity sequences, the step S1035 calculates several displacement sequences. It should be noted that the methods for integrating the forward acceleration and the absolute velocity sequences include, but are not limited to, left rectangular integration, right rectangular integration, mid-rectangular integration, trapezoidal integration, and Simpson integration, which are not limited here.

[0050] Step S1037, based on several displacement sequences of the front sensor and the distance between the two sensors, calculate several displacement sequences of the rear sensor.

[0051] In this embodiment, based on several displacement sequences of the front sensor and the distance between the two sensors, calculate several displacement sequences of the rear sensor, and the formula is: . In the formula, is the distance between the two nine-axis sensors, is the rear sensor the displacement amount at a certain moment compared to the last moment of the time window.

[0052] Step S105, combine each displacement sequence with the corresponding magnetic induction intensity sequence to obtain several original displacement-magnetic field sequence pairs for the front sensor and the rear sensor respectively.

[0053] In step S105, the two nine-axis sensors collect sensor data synchronously within the time window At each sampling moment, combine the displacement at this moment with the magnetic induction intensity to form an original displacement-magnetic field data pair. That is, the displacement sequences and the magnetic induction intensity sequences corresponding to all moments within the time window are combined into an original displacement-magnetic field sequence pair. Combining several displacement sequences with the corresponding magnetic induction intensity sequences forms several original displacement-magnetic field sequence pairs.

[0054] Step S107, perform equidistant resampling on each original displacement-magnetic field sequence pair to obtain several new displacement-magnetic field sequence pairs for the front sensor and the rear sensor respectively.

[0055] In step S107, a new set of equally spaced target sampling points is reconstructed. Each original displacement-magnetic field sequence is aligned at the target sampling points. The aligned original displacement-magnetic field sequence pairs are linearly interpolated at the target sampling points, thereby obtaining new displacement-magnetic field matching sequence pairs for the front sensor and the rear sensor respectively. The new displacement-magnetic field matching sequence pairs can be directly used to compare the magnetic field similarities of two nine-axis sensors at the same position. It can be understood that the original displacement-magnetic field sequences obtained in step S105 are data collected at equal time intervals, and their sampling points are evenly distributed in the time dimension. In order to calculate the similarity of the displacement-magnetic field sequence pairs of two nine-axis sensors corresponding to the same candidate carrier speed by using the characteristic that the magnetic sequences collected by the front and rear magnetometers at the same position are consistent. Therefore, it is necessary to resample the original displacement-magnetic field sequences at equal distances to convert the original displacement-magnetic field sequences from "evenly sampled in the time dimension" to "evenly sampled in the distance dimension".

[0056] Step S109: Calculate the similarity between the new displacement-magnetic field sequence pair of the front sensor and the new displacement-magnetic field sequence pair of the rear sensor corresponding to the same candidate carrier speed, and determine the candidate carrier speed corresponding to the displacement-magnetic field sequence pair with the maximum similarity as the carrier speed.

[0057] In step S109, in this embodiment, the Pearson correlation coefficient is used to compare each candidate carrier speed corresponding displacement-magnetic field sequences and for similarity. The Pearson correlation coefficient calculation formula is as follows: , where is the Pearson correlation coefficient,[[]] is the number of samples of the sequence (i.e., the number of target sampling points), and are respectively and sample values of,[[]] and are respectively and means of. It can be understood that because the magnetic field at the same position is unique, the displacement-magnetic field sequence pair with the maximum similarity indicates that the magnetic field data of the front and rear magnetometers at the corresponding position are the most consistent, and the corresponding candidate carrier speed most conforms to the actual motion of the carrier. Therefore, it is determined as the carrier speed. In some other feasible embodiments, similarity measurement methods such as Euclidean distance, Manhattan distance, Chebyshev distance, Mahalanobis distance, cosine similarity, and dynamic time warping (DTW) can also be used to calculate the similarity of the displacement-magnetic field sequence pairs, which is not limited here.

[0058] Please refer to the attached Figure 7, the present invention also provides an estimation device 4 for the carrier speed of magnetic-inertial fusion. The estimation device 4 includes a data extraction module 41, a displacement calculation module 42, a first magnetic moment matching module 43, a second magnetic moment matching module 44, and a carrier speed determination module 45. The data extraction module 41 is used to extract the sensor data of the front sensor 1 and the rear sensor 2 within a preset time window; wherein, the front sensor 1 and the rear sensor 2 are respectively installed at the front and rear of the carrier 3 and are both nine-axis sensors, and the sensor data includes the magnetic induction intensity sequence and the acceleration information. The displacement calculation module 42 is used to perform a traversal search for the carrier speed within a preset speed range, and calculate a plurality of displacement sequences of the front sensor and the rear sensor respectively in combination with the acceleration information; wherein, the preset speed range includes a plurality of candidate carrier speeds. The first magnetic moment matching module 43 is used to combine each displacement sequence with the corresponding magnetic induction intensity sequence to obtain a plurality of original displacement-magnetic field sequence pairs of the front sensor 1 and the rear sensor 2 respectively. The second magnetic moment matching module 44 is used to perform equidistant resampling on each original displacement-magnetic field sequence pair to obtain a plurality of new displacement-magnetic field sequence pairs of the front sensor 1 and the rear sensor 2 respectively. The carrier speed determination module 45 is used to calculate the similarity between the new displacement-magnetic field sequence pairs of the front sensor 1 and the new displacement-magnetic field sequence pairs of the rear sensor 2 corresponding to the same candidate carrier speed, and determine the candidate carrier speed corresponding to the displacement-magnetic field sequence pair with the maximum similarity as the carrier speed.

[0059] Please refer to the attached Figure 8 , the present invention also provides an estimation system 1000 for the carrier speed of magnetic-inertial fusion. The estimation system 1000 includes a front sensor 1, a rear sensor 2, and a processing module 100. The front sensor 1 and the rear sensor 2 are respectively installed at the front and rear of the carrier 3 and are both nine-axis sensors. The processing module 100 is used to execute the magnetic-inertial fusion carrier speed estimation method according to the sensor data collected by the front and rear sensors.

[0060] In summary, in the embodiments of the present invention, nine-axis sensors are respectively installed at the front and rear of the carrier. By utilizing the characteristic that the magnetic sequences collected by the front and rear nine-axis sensors are consistent at the same position, combined with the lever-arm compensation and sequence matching optimization algorithms, the optimal carrier speed solution is searched and calculated, so as to obtain an accurate carrier speed measurement result. That is to say, the present invention only calculates the speed through the internal nine-axis sensors and does not rely on external signals (such as GNSS signals). Therefore, the present invention can still obtain an accurate carrier speed measurement result in GNSS signal rejection environments such as tunnels, urban canyons, and underground parking lots, thus effectively overcoming the problem of the decline in speed measurement accuracy caused by signal loss in traditional satellite navigation. Secondly, the present invention combines the acceleration information measured by the inertial measurement unit with the magnetic field information measured by the magnetometer, and can timely reflect the change of the carrier speed even in the case of high-speed movement. Therefore, the present invention can be used for the carrier speed measurement in high-speed movement scenarios. In addition, compared with the solutions that rely on external infrastructure, the present invention does not rely on a sensor array that requires precise installation and calibration, and only two low-cost nine-axis sensors are needed to achieve autonomous speed measurement, which has the advantages of convenient deployment, low cost, and strong robustness, and significantly improves the navigation reliability in complex environments.

[0061] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the technical solutions of the embodiments of the present invention.

Claims

1. A method for estimating the speed of a carrier by magnetic-inertial fusion, characterized in that Including: Extracting sensor data of the front sensor and the rear sensor within a preset time window; wherein, the front sensor and the rear sensor are respectively installed at the front and rear of the carrier and are both nine-axis sensors, and the sensor data includes a magnetic induction intensity sequence and acceleration information; Performing a traversal search on the carrier speed within a preset speed range, and calculating a plurality of displacement sequences of the front sensor and the rear sensor respectively in combination with the acceleration information; wherein, the preset speed range includes a plurality of candidate carrier speeds; Combining each displacement sequence with the corresponding magnetic induction intensity sequence to obtain a plurality of original displacement-magnetic field sequence pairs of the front sensor and the rear sensor respectively; Performing equidistant resampling on each original displacement-magnetic field sequence pair to obtain a plurality of new displacement-magnetic field sequence pairs of the front sensor and the rear sensor respectively; Calculating the similarity between the new displacement-magnetic field sequence pairs of the front sensor and the new displacement-magnetic field sequence pairs of the rear sensor corresponding to the same candidate carrier speed, and determining the candidate carrier speed corresponding to the displacement-magnetic field sequence pair with the maximum similarity as the carrier speed.

2. The method for estimating the carrier speed by magnetic-inertial fusion according to claim 1, characterized in that, Also including: Projecting the magnetic induction intensity sequence and the acceleration information into the carrier coordinate system.

3. A method for estimating the carrier speed by magnetic-inertial fusion according to claim 1, characterized in that Performing a traversal search on the carrier speed within a preset speed range, and calculating a plurality of displacement sequences of the front sensor and the rear sensor respectively in combination with the acceleration information, including: Calculating a speed change sequence of the carrier within a preset time window based on the acceleration information collected by the front sensor; Calculating an absolute speed sequence corresponding to each candidate carrier speed in combination with the speed change sequence; wherein, a plurality of candidate carrier speeds correspond to a plurality of absolute speed sequences; Calculating a plurality of displacement sequences of the front sensor based on the plurality of absolute speed sequences; Calculating a plurality of displacement sequences of the rear sensor based on the plurality of displacement sequences of the front sensor and the distance between the two sensors.

4. The method for estimating the carrier velocity by magnetic-inertial fusion according to claim 3, wherein Calculating a speed change sequence of the carrier within a preset time window based on the acceleration information collected by the front sensor, including: Calculating the acceleration of the carrier at each moment within a preset time window based on the acceleration information collected by the front sensor; Calculating the speed change amount between each moment and the next moment of the carrier based on the acceleration; Integrating the speed change amount to obtain the speed change amount from each moment to the last moment of the carrier; Jointly forming the speed change sequence by the speed change amounts from each moment to the last moment within the preset time window.

5. The method for estimating the carrier speed by magnetic-inertial fusion according to claim 3, characterized in that, Calculating an absolute speed sequence corresponding to each candidate carrier speed in combination with the speed change sequence; wherein, a plurality of candidate carrier speeds correspond to a plurality of absolute speed sequences, including: Combining each candidate carrier speed with the speed change sequence to calculate the absolute speed corresponding to each moment within a preset time window; Jointly forming the absolute speed sequence by the absolute speeds at each moment within the preset time window; wherein, the plurality of candidate carrier speeds correspond to a plurality of absolute speed sequences.

6. The method for estimating the carrier velocity of magnetic-inertial fusion according to claim 3, wherein Calculating a plurality of displacement sequences of the front sensor based on the plurality of absolute speed sequences, including; Calculate the displacement of the carrier between each moment and the next moment based on each absolute velocity sequence; Integrate the displacement to obtain the displacement of the carrier between each moment and the last moment; Combine the displacements between each moment and the last moment within a preset time window to form the displacement sequence; wherein, a plurality of absolute velocity sequences correspond to a plurality of displacement sequences.

7. A magnetic-inertial fusion-based carrier velocity estimation method according to claim 1, characterized in that, Resample each original displacement-magnetic field sequence pair at equal intervals to obtain a plurality of new displacement-magnetic field sequence pairs for the front sensor and the rear sensor respectively, including: Reconstruct a set of equally spaced target sampling points; Align each original displacement-magnetic field sequence pair at the target sampling points; Perform linear interpolation on the aligned original displacement-magnetic field sequence pairs at the target sampling points to obtain a plurality of new displacement-magnetic field matching sequence pairs for the front sensor and the rear sensor respectively.

8. A method for estimating the carrier velocity by magnetic-inertial fusion according to claim 1, characterized in that, Calculate the similarity between the new displacement-magnetic field sequence pairs of the front sensor and the new displacement-magnetic field sequence pairs of the rear sensor corresponding to the same candidate carrier velocity, including: Compare the similarity between the new displacement-magnetic field sequence pairs of the front sensor and the new displacement-magnetic field sequence pairs of the rear sensor corresponding to each candidate carrier velocity by using Pearson correlation coefficient, Euclidean distance, Manhattan distance, Chebyshev distance, Mahalanobis distance, cosine similarity or dynamic time warping.

9. An estimation device for the speed of a magneto-inertial fusion carrier, characterized in that, The estimation device includes: A data extraction module for extracting sensor data of the front sensor and the rear sensor within a preset time window; wherein, the front sensor and the rear sensor are respectively installed at the front and rear of the carrier and are both nine-axis sensors, and the sensor data includes a magnetic induction intensity sequence and acceleration information; A displacement calculation module for traversing and searching the carrier velocity within a preset velocity range, and calculating a plurality of displacement sequences for the front sensor and the rear sensor respectively in combination with the acceleration information; wherein, the preset velocity range includes a plurality of candidate carrier velocities; A first magnetic moment matching module for combining each displacement sequence with the corresponding magnetic induction intensity sequence to obtain a plurality of original displacement-magnetic field sequence pairs for the front sensor and the rear sensor respectively; A second magnetic moment matching module for resampling each original displacement-magnetic field sequence pair at equal intervals to obtain a plurality of new displacement-magnetic field sequence pairs for the front sensor and the rear sensor respectively; A carrier velocity determination module for calculating the similarity between the new displacement-magnetic field sequence pairs of the front sensor and the new displacement-magnetic field sequence pairs of the rear sensor corresponding to the same candidate carrier velocity, and determining the candidate carrier velocity corresponding to the displacement-magnetic field sequence pair with the maximum similarity as the carrier velocity.

10. An estimation system for the speed of a magneto-inertial fusion carrier, characterized in that, The estimation system includes: A front sensor and a rear sensor, which are respectively installed at the front and rear of the carrier and are both nine-axis sensors; and A processing module, which is configured to execute a method for estimating the carrier velocity by magnetic-inertial fusion according to any one of claims 1-8 based on the sensor data collected by the front and rear sensors.

Images