CN111323009A - Magnetic suspension train positioning method and system - Google Patents
Magnetic suspension train positioning method and system Download PDFInfo
- Publication number
- CN111323009A CN111323009A CN202010157760.9A CN202010157760A CN111323009A CN 111323009 A CN111323009 A CN 111323009A CN 202010157760 A CN202010157760 A CN 202010157760A CN 111323009 A CN111323009 A CN 111323009A
- Authority
- CN
- China
- Prior art keywords
- train
- measurement unit
- camera
- map
- positioning
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/10—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration
- G01C21/12—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning
- G01C21/16—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning by integrating acceleration or speed, i.e. inertial navigation
- G01C21/165—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning by integrating acceleration or speed, i.e. inertial navigation combined with non-inertial navigation instruments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
Abstract
The invention relates to the technical field of magnetic suspension train positioning, in particular to a magnetic suspension train positioning method and system. Acquiring image frames on a train motion path in real time; acquiring three-axis acceleration and three-axis angular velocity of the train in real time; extracting feature point information in the image frame, and tracking the extracted feature points in the image frame; preprocessing the measured value of the inertial measurement unit; fusing data of the camera and the inertia measurement unit to obtain an intermediate variable for solving a pre-integral formula, and updating a pre-integral value of the inertia measurement unit; and combining the updated pre-integral value with an adjacent interframe motion formula to obtain the current position and attitude information of the magnetic suspension train. The invention provides a high-temperature superconducting magnetic levitation test vehicle test positioning method based on a visual inertial odometer, which combines a visual sensor and an inertial sensor, meets the requirement of acquiring position and speed information of a magnetic levitation vehicle in real time, and is an ideal expansion of the traditional sensor mode.
Description
Technical Field
The invention relates to the technical field of magnetic suspension train positioning, in particular to a magnetic suspension train positioning method and system.
Background
Along with the development of social economy and the increasing population scale, the demand of people on public transportation is increasing. As a new traffic technology in recent years, the maglev train has the advantages of low noise, no friction between wheel tracks, stability, comfort and environmental protection, and has great application prospect in the field of urban ground traffic in the future. In the running of the magnetic suspension train, the control system is the key for the normal running of the train, and the reliability of the train control system can be enhanced by accurately estimating the position and the speed of the train.
Since the high-temperature superconducting magnetic suspension annular experimental line of southwest university of transportation is successfully developed, the problem of system optimization around the annular line becomes a great research direction. The ring line is positioned by adopting the detection technology of identifying the marker by an infrared sensor, the infrared sensor is positioned at the bottom of the magnetic suspension train, and black and white stripes with equal intervals are laid on the magnetic track of the ring line and the guard rail. The infrared sensor can detect the general position of the train by identifying the black and white stripes.
The conventional speed measurement and positioning methods for magnetic suspension trains at home and abroad are mainly divided into two categories according to the reference of position information: absolute positioning methods and relative positioning methods. Absolute positioning relies on a marker of known location alongside the track to obtain absolute train position information. The relative positioning method can measure and calculate the relative position information of the train by using the initial position and the train displacement on the set route.
The absolute positioning method mainly comprises the methods of inquiry-responder, pulse width coding, GNSS and the like. The transponder method is that the absolute mileage value of the current position is stored in the transponder, when a train runs to the position of the ground transponder, the transponder sends the stored information to the vehicle-mounted inquiry unit in an electromagnetic induction mode, and the received information is processed by the vehicle-mounted computer to achieve the positioning purpose. The absolute positioning technology based on pulse width coding is similar to the realization form of a query-responder method, and the method identifies the mark plates which are paved at fixed positions along the track through a vehicle-mounted reader. The GNSS-based positioning method is a high-precision radio navigation positioning system based on 24 in-orbit satellites, and the Doppler effect is utilized to measure the information of a moving object.
The relative positioning method mainly comprises a method based on an induction loop, a sleeper technology and long stator tooth slot detection. The induction loop method is based on the electromagnetic induction phenomenon of an induction coil at the bottom of a vehicle and an induction loop laid on a track, so that positioning and speed measurement are realized. The speed measurement positioning method based on sleeper counting is characterized in that counting and calculating are carried out on metal sleepers through eddy current sensors, and under the condition that the distance between the sleepers is fixed, relative displacement is obtained through measuring position pulse frequency, so that speed information is obtained. The method based on the long stator tooth space detection detects the long stator tooth space by means of a sensor at a train rotor, and obtains train speed information by measuring the number of the passing tooth spaces in unit time.
The prior art objective generally has the following problems:
1) the current infrared sensor technique that adopts carries out the not enough of location that tests the speed:
(1) the problem of positioning information loss and jump caused by the suspension drift phenomenon when the magnetic suspension vehicle runs;
(2) when the magnetic suspension vehicle is in an un-suspended state, the vehicle bottom is in contact with the magnetic track, so that the black and white stripes adhered to the position of the magnetic suspension vehicle are greatly abraded;
(3) the measurement accuracy is low due to the insufficient information contained in the black and white stripes.
2) The defects of the traditional magnetic levitation technology
The method for inquiring the transponder in absolute positioning, pulse width coding and induction loop in relative positioning has strong anti-interference capability and high precision, but has high equipment maintenance cost.
Satellite signals in the GNSS system are easily shielded by terrain and buildings, and factors such as atmosphere, multipath effect, electromagnetic interference and the like can influence data accuracy, so that the method cannot perform speed measurement and positioning in an indoor environment.
The speed measurement positioning method based on sleeper counting is mainly applied to medium and low speed maglev trains such as a long sand maglev express line and the like, and the used disc-shaped eddy current sensor requires that the diameter of a measured object is more than or equal to 2X of the diameter of a probe coil, is limited by the width of a metal sleeper between ring-shaped line guard rails and is not suitable for the method.
The method based on the long stator tooth space is generally used in high-speed magnetic levitation, but the method is not suitable for the environment to be measured of a ring line because the method depends on the tooth space structure of the long stator.
By combining the analysis, different magnetic levitation systems and line characteristics all around the world have respective corresponding speed measurement positioning modes, and the speed measurement positioning modes have certain limitations for the high-temperature superconducting magnetic levitation annular experimental line which is different from the structural modes of normal conduction and low-temperature superconducting. The high-temperature superconducting magnetic levitation is still in an experimental stage and is limited by inherent environments (such as a linear motor laying section, magnetic field interference, abrasion and the like), and a new speed measurement positioning scheme is needed to acquire real-time speed and position information of a train.
Disclosure of Invention
The present invention aims to provide a method and a system for positioning a maglev train to improve the above problems. In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
in one aspect, the present application provides a method for positioning a magnetic levitation train, the method comprising:
arranging a camera and an inertia measurement unit on a target magnetic suspension train; the camera collects image frames on a train motion path in real time; the inertia measurement unit acquires the three-axis acceleration and the three-axis angular velocity of the train in real time;
extracting feature point information in the image frame, and tracking the feature points extracted from the image frame by a feature matching method;
preprocessing the measurement value of the inertia measurement unit by using a pre-integral formula;
fusing data of the camera and the inertia measurement unit to obtain an intermediate variable for solving a pre-integral formula, and updating a pre-integral value of the inertia measurement unit;
and combining the updated pre-integral value with the motion relation of the adjacent frames to obtain the current position and attitude information of the magnetic suspension train.
Optionally, the fusing the data of the camera and the inertial measurement unit to obtain an intermediate variable for solving a pre-integration formula includes:
calculating the relative rotation from the inertial measurement unit coordinate system to the camera coordinate system;
initializing a camera, and solving the poses of all frames and the coordinates of the observation landmark points;
and aligning the initialized result of the camera with the pre-integration result of the inertial measurement unit, and solving the bias of the gyroscope, the absolute scale, the gravity acceleration, the acceleration bias and the speed of each frame.
Optionally, the combining the updated pre-integration value with the adjacent inter-frame motion formula to obtain the current position and posture information of the maglev train includes:
and calculating the position, the speed and the offset angle according to the motion relation of adjacent frames by combining the obtained values of the gyroscope offset, the absolute scale, the gravity acceleration and the acceleration offset with the obtained triaxial acceleration value and triaxial angular velocity value of the inertial measurement unit.
Optionally, the camera is arranged at the head position of the target maglev train, and the orientation is the train running direction.
Optionally, the method further comprises:
and selecting the collected image frames, and taking the frames with sufficient number of characteristic points and uniform distribution of the characteristic points as key frames. The purpose is to reduce the amount of computation.
Optionally, the method further comprises:
by using a sliding window method, the key frames are optimized through the fixed window size, the number of the optimized key frames in the process is ensured to be within a certain range, and the excessive calculation amount is avoided.
Since the VIO technique requires the calculation of the current image frame according to the image-to-image variation relationship. The current frame image we process is a frame of a series of images acquired by a camera, each image being closely related to the others. The fixed sliding window size is used for specifying how many image frames are processed at one time, and can be set manually in actual calculation. The number of key frames is limited by the size of the fixed window, so that the purpose of balancing the calculated amount is achieved. The key frames need to satisfy the characteristics of sufficient quantity of the feature points, uniform distribution of the feature points as much as possible and the like, and the selection of the key frames can be artificially limited according to different environments, for example, 10 frames are required to be separated between adjacent key frames, and the number of the feature points of the key frames is up to 30. After the primary estimation of the magnetic suspension train pose is completed, the pose needs to be further optimized, and the optimization function comprises a visual re-projection residual error, an inertial measurement unit measurement residual error and an marginalized prior residual error.
Optionally, the method further comprises:
performing loop detection on the currently acquired new key frame, wherein the loop detection comprises: carrying out similarity detection on the new key frame and the old key frame stored in the key frame database;
if the similarity degree of the new key frame and the old key frame is detected to reach a preset value, loop detection is successful, and the attitude map is adjusted globally to obtain a better motion result. The loop detection is successful, namely the position of the train passing through the old key frame is collected; the preset value can be set manually. The objective function to be optimized comprises a visual re-projection residual, an IMU measurement residual, an marginalized prior residual and a closed-loop frame re-projection error. The pose graph is as follows: only the connections between all camera poses are concerned and the graph formed by the trajectories of the keyframes is retained.
Optionally, the method further comprises:
and when the image frame features are lost, recording a generated map after the loss as a new map, recording a map before the loss as an old map, storing key frames in the new map and the old map in a key frame database, and aligning and fusing the new map and the old map after loop detection is successful. When the situation that the feature points cannot be extracted due to image blurring, unobvious texture, dynamic object influence and the like occurs, the image frame feature tracking loss is judged.
Optionally, the method further comprises:
when the train stops running, the state quantity of each key frame in the map and the relation of adjacent frames are stored, and in the next measurement, the speed measurement and the positioning of the target magnetic suspension train in the environment to be measured can be directly realized by loading the stored map.
In another aspect, the present invention provides a magnetic levitation train positioning system, comprising:
the camera is arranged on the target magnetic suspension train and used for acquiring image frames on a train motion path in real time;
the system comprises an inertia measurement unit, a three-axis acceleration sensor, a three-axis angular velocity sensor and a control unit, wherein the inertia measurement unit is arranged on a target magnetic suspension train and is used for acquiring the three-axis acceleration and the three-axis angular velocity of the train in real time;
the visual characteristic tracking module is used for extracting characteristic point information in the image frame and tracking the extracted characteristic points in the image frame by a characteristic matching method;
the pre-integration module is used for preprocessing the measurement value of the inertia measurement unit by using a pre-integration formula;
the initialization module is used for fusing data of the camera and the inertia measurement unit to obtain an intermediate variable for solving a pre-integral formula and updating a pre-integral value of the inertia measurement unit;
and the pose estimation module is used for combining the updated pre-integral value with the motion formula of the adjacent frame to obtain the current position and attitude information of the magnetic suspension train.
The invention has the beneficial effects that:
the invention provides a high-temperature superconducting magnetic levitation test vehicle test positioning method based on a visual inertial odometer, which combines a visual sensor and an inertial sensor, meets the requirement of acquiring position and speed information of a magnetic levitation vehicle in real time, and is an ideal expansion of the traditional sensor mode.
The invention adopts a method based on a visual inertial odometer to realize the positioning and map building work of the magnetic suspension vehicle. The camera can catch abundant scene information, and the inertial measurement unit can obtain the acceleration and the angular velocity information of magnetic suspension car in a short time high frequency to alleviate the influence of the dynamic fuzzy problem that the quick motion caused to camera observed value. The camera data can also effectively correct the drift of the inertial measurement unit in the reading, and the two sensors complement each other to a certain extent, so that the camera data has good robustness and the potential of providing accurate motion estimation.
The camera and the inertia measurement unit have small volume, high precision, strong portability, low power consumption and low cost, and are widely applied to equipment needing motion control.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by the practice of the embodiments of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
Fig. 1 is a schematic flow chart of a magnetic levitation train positioning method according to an embodiment of the present invention;
fig. 2 is a block diagram of a positioning system of a magnetic levitation train according to an embodiment of the present invention.
Detailed Description
In order to make the objects, 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 with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures. Meanwhile, in the description of the present invention, the terms "first", "second", and the like are used only for distinguishing the description, and are not to be construed as indicating or implying relative importance.
In one aspect, as shown in fig. 1, the present embodiment provides a magnetic levitation train positioning method, which includes step S10, step S20, step S30, step S40, and step S50.
S10, arranging a camera and an inertia measurement unit on a target magnetic suspension train; the camera collects image frames on a train motion path in real time; the inertia measurement unit acquires the three-axis acceleration and the three-axis angular velocity of the train in real time;
s20, extracting feature point information in the image frame, and tracking the extracted feature points in the image frame by a feature matching method;
s30, preprocessing the measurement value of the inertia measurement unit by using a pre-integral formula;
s40, fusing data of the camera and the inertial measurement unit to obtain an intermediate variable for solving a pre-integral formula, and updating a pre-integral value of the inertial measurement unit;
and S50, combining the updated pre-integral value with the motion formula of the adjacent frame to obtain the current position and attitude information of the magnetic suspension train and obtain the current position and attitude information of the magnetic suspension train.
The pre-integration formula is as follows:
in the above-mentioned formula,
is b iskThe time of day inertial measurement unit coordinate system rotation matrix,
is a quaternion of the rotation matrix and,
in order to be a measure of the acceleration,
in order to be biased by the acceleration,
in order to take the measurements of the gyroscope,
in order to bias the gyroscope,
for the purpose of the successive inter-frame position increments,
for the purpose of the adjacent inter-frame speed increment,
for adjacent interframe yaw angle increments, the three increments are only related to the current state of the inertial measurement unit and are not related to other state quantities.
The motion formula between the adjacent frames is as follows:
in the above-mentioned formula,
is world coordinate tokMoment inertial measurement unit coordinate system rotation matrix, gwIs the gravity acceleration under the world coordinate system,
each represents bk+1Position, velocity, and yaw angle of the frame.
Optionally, step S40 may further include step S401, step S402, and step S403.
S401, obtaining the relative rotation from an inertial measurement unit coordinate system to a camera coordinate system;
s402, initializing a camera, and solving the poses of all frames and the coordinates of the observation landmark points;
and S403, aligning the initialized result of the camera with the pre-integration result of the inertial measurement unit, and solving the bias of the gyroscope, the absolute scale, the gravity acceleration, the acceleration bias and the speed of each frame.
Optionally, in step S50, the method may further include:
and calculating the position, the speed and the offset angle by combining the obtained values of the gyroscope offset, the absolute scale, the gravity acceleration and the acceleration offset with the obtained triaxial acceleration value and triaxial angular velocity value of the inertial measurement unit through an adjacent interframe motion formula.
Optionally, the camera is arranged at the head position of the target maglev train, and the orientation is the train running direction.
Optionally, the method for positioning a magnetic levitation train may further include step S60.
And S60, selecting the collected image frames, and taking the frames with sufficient number of characteristic points and uniform distribution of the characteristic points as key frames. The purpose is to reduce the amount of computation.
Optionally, the method for positioning a magnetic levitation train may further include step S70.
And S70, optimizing the key frames by using a sliding window method through a fixed window size, ensuring the number of the optimized key frames in the process to be within a certain range, and avoiding overlarge calculation amount. The objective function to be optimized comprises a weight sensing projection residual error, an inertia measurement unit measurement residual error and a marginalization prior residual error.
Since the VIO technique requires the calculation of the current image frame according to the image-to-image variation relationship. The current frame image we process is a frame of a series of images acquired by a camera, each image being closely related to the others. The fixed sliding window size is used for specifying how many image frames are processed at one time, and can be set manually in actual calculation. The number of key frames is limited by the fixed window size, so that the purpose of balancing the calculated amount is achieved. The key frame to be optimized comprises a visual re-projection residual, an inertial measurement unit measurement residual and a marginalized prior residual. The optimized key frame has the characteristics of sufficient quantity of characteristic points, uniform distribution of the characteristic points and the like, the selection of the key frame can be artificially limited according to different environments, for example, 10 frames are needed to be separated between adjacent key frames, the number of the characteristic points of the key frame per se is up to 30, and the like
Optionally, the method for positioning a magnetic levitation train may further include step S80.
S80, carrying out loop detection on the currently acquired new key frame, wherein the loop detection comprises the following steps: carrying out similarity detection on the new key frame and the old key frame stored in the key frame database; if the similarity degree of the new key frame and the old key frame is detected to reach a preset value, loop detection is successful, and the attitude map is adjusted globally to obtain a better motion result. The loop detection is successful, namely the position of the train passing through the old key frame is collected; the preset value can be set manually. The objective function to be optimized comprises a visual re-projection residual, an IMU measurement residual, an marginalized prior residual and a closed-loop frame re-projection error. The pose graph is as follows: only the connections between all camera poses are concerned and the graph formed by the trajectories of the keyframes is retained.
Optionally, the method for positioning a magnetic levitation train may further include step S90.
And S90, when the image frame features are lost, recording a generated map after the loss as a new map, recording the map before the loss as an old map, storing key frames in the new map and the old map in a key frame database, and fusing the new map and the old map after loop detection is successful. When the situation that the feature points cannot be extracted due to image blurring, unobvious texture, dynamic object influence and the like occurs, the image frame feature tracking loss is judged.
Optionally, the magnetic levitation train positioning method may further include step S100.
And S100, when the train stops running, storing the state quantity of each key frame in the map and the relation between adjacent frames, and directly realizing speed measurement and positioning of the target magnetic suspension train in the environment to be measured by loading the stored map in the next measurement.
In another aspect, the present invention provides a magnetic levitation train positioning system, comprising:
the camera is arranged on the target magnetic suspension train and used for acquiring image frames on a train motion path in real time;
the system comprises an inertia measurement unit, a three-axis acceleration sensor, a three-axis angular velocity sensor and a control unit, wherein the inertia measurement unit is arranged on a target magnetic suspension train and is used for acquiring the three-axis acceleration and the three-axis angular velocity of the train in real time;
the visual characteristic tracking module is used for extracting characteristic point information in the image frame and tracking the extracted characteristic points in the image frame by a characteristic matching method;
the pre-integration module is used for preprocessing the measurement value of the inertia measurement unit by using a pre-integration formula;
the initialization module is used for fusing data of the camera and the inertia measurement unit to obtain an intermediate variable for solving a pre-integral formula and updating a pre-integral value of the inertia measurement unit;
and the pose estimation module is used for combining the updated pre-integration value with the motion formula of the adjacent frame to obtain the current position and attitude information of the magnetic suspension train.
Optionally, the system may further comprise a key frame retrieval module.
And the key frame retrieval module is used for selecting the collected image frames and taking the frames with sufficient number of characteristic points and uniform distribution of the characteristic points as a key frame.
Optionally, the system may further comprise a sliding window non-linear optimization module.
And the sliding window nonlinear optimization module is used for optimizing the key frames by using a sliding window method and through a relatively fixed window size, so that the number of the optimized key frames in the process is ensured to be within a certain range, and the excessive calculation amount is avoided. The objective function to be optimized comprises a visual re-projection residual, an IMU measurement residual and an marginalized prior residual.
Optionally, the system may further include a loop detection module and a global pose close-coupling optimization module.
A loop detection module, configured to perform loop detection on a currently acquired new key frame, where the loop detection includes: carrying out similarity detection on the new key frame and the old key frame stored in the key frame database;
and the global pose close coupling optimization module is used for successfully performing loop detection and globally adjusting the pose graph to obtain a better motion result if the similarity degree of the new key frame and the old key frame reaches a preset value.
Optionally, the system may further include a repositioning and pose graph reuse module.
And the repositioning and pose graph reusing module is used for recording a map generated after the image frame characteristics are lost as a new map, recording a map before the loss as an old map, storing key frames in the new map and the old map in a key frame database, and fusing the new map and the old map after loop detection is successful.
Optionally, the repositioning and pose graph reusing module may be further configured to:
the method is used for storing the state quantity of each key frame and the relation between adjacent frames in the map when the train stops running, and in the next measurement, the speed measurement and the positioning of the target magnetic suspension train in the environment to be measured can be directly realized by loading the stored map.
The implementation principle and the technical effects of the positioning system provided in the embodiment of the present invention are the same as those of the aforementioned embodiment of the positioning method, and for the sake of brief description, reference may be made to corresponding contents in the aforementioned embodiment of the positioning method where no part of the embodiment of the positioning system is mentioned.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and all the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. A method for positioning a magnetic levitation train, comprising:
arranging a camera and an inertia measurement unit on a target magnetic suspension train; the camera collects image frames on a train motion path in real time; the inertia measurement unit acquires the three-axis acceleration and the three-axis angular velocity of the train in real time;
extracting feature point information in the image frame, and tracking the extracted feature points in the image frame;
preprocessing the measurement value of the inertia measurement unit by using a pre-integral formula;
fusing data of the camera and the inertia measurement unit to obtain an intermediate variable for solving a pre-integral formula, and updating a pre-integral value of the inertia measurement unit;
and combining the updated pre-integral value with an adjacent interframe motion formula to obtain the current position and attitude information of the magnetic suspension train.
2. The method for positioning a maglev train according to claim 1, wherein the fusing the data of the camera and the inertial measurement unit to obtain the intermediate variable for solving the pre-integration formula comprises:
calculating the relative rotation from the inertial measurement unit coordinate system to the camera coordinate system;
initializing a camera, and solving the poses of all frames and the coordinates of the observation landmark points;
and aligning the initialized result of the camera with the pre-integration result of the inertial measurement unit, and solving the bias of the gyroscope, the absolute scale, the gravity acceleration, the acceleration bias and the speed of each frame.
3. The method for positioning a maglev train according to claim 2, wherein the step of combining the updated pre-integration value with the motion formula between adjacent frames to obtain the current position and attitude information of the maglev train comprises:
and calculating the position, the speed and the offset angle by combining the obtained values of the gyroscope offset, the absolute scale, the gravity acceleration and the acceleration offset with the obtained triaxial acceleration value and triaxial angular velocity value of the inertial measurement unit through an adjacent interframe motion formula.
4. The method for positioning a magnetic levitation train as recited in claim 1, wherein: the camera is arranged at the head position of the target magnetic suspension train and faces to the running direction of the train.
5. The method for locating a magnetic levitation train as recited in claim 1, further comprising:
and selecting the collected image frames, and taking the frames with sufficient number of characteristic points and uniform distribution of the characteristic points as key frames.
6. The method for locating a magnetic levitation train as recited in claim 5, further comprising:
and optimizing the key frame by a fixed window size by using a sliding window method.
7. The method for locating a magnetic levitation train as recited in claim 5, further comprising:
performing loop detection on the currently acquired new key frame, wherein the loop detection comprises: carrying out similarity detection on the new key frame and the old key frame stored in the key frame database;
if the similarity degree of the new key frame and the old key frame is detected to reach a preset value, loop detection is successful, and the attitude map is adjusted globally to obtain a better motion result.
8. Method for positioning a magnetic levitation train according to claim 7, further comprising:
and when the image frame characteristics are lost, recording a generated map after the loss as a new map, recording a map before the loss as an old map, storing key frames in the new map and the old map in a key frame database, and fusing the new map and the old map after loop detection is successful.
9. Method for positioning a magnetic levitation train according to claim 7, further comprising:
when the train stops running, the state quantity of each key frame in the map and the relation of adjacent frames are stored, and in the next measurement, the speed measurement and the positioning of the target magnetic suspension train in the environment to be measured can be directly realized by loading the stored map.
10. A magnetic levitation train positioning system, the system comprising:
the camera is arranged on the target magnetic suspension train and used for acquiring image frames on a train motion path in real time;
the system comprises an inertia measurement unit, a three-axis acceleration sensor, a three-axis angular velocity sensor and a control unit, wherein the inertia measurement unit is arranged on a target magnetic suspension train and is used for acquiring the three-axis acceleration and the three-axis angular velocity of the train in real time;
the visual characteristic tracking module is used for extracting characteristic point information in the image frame and tracking the extracted characteristic points in the image frame;
the pre-integration module is used for preprocessing the measurement value of the inertia measurement unit by using a pre-integration formula;
the initialization module is used for fusing data of the camera and the inertia measurement unit to obtain an intermediate variable for solving a pre-integral formula and updating a pre-integral value of the inertia measurement unit;
and the pose estimation module is used for combining the updated pre-integration value with the motion formula of the adjacent frame to obtain the current position and attitude information of the magnetic suspension train.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010157760.9A CN111323009A (en) | 2020-03-09 | 2020-03-09 | Magnetic suspension train positioning method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010157760.9A CN111323009A (en) | 2020-03-09 | 2020-03-09 | Magnetic suspension train positioning method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111323009A true CN111323009A (en) | 2020-06-23 |
Family
ID=71171441
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010157760.9A Pending CN111323009A (en) | 2020-03-09 | 2020-03-09 | Magnetic suspension train positioning method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111323009A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113390435A (en) * | 2021-05-13 | 2021-09-14 | 中铁二院工程集团有限责任公司 | High-speed railway multi-element auxiliary positioning system |
| CN114585879A (en) * | 2020-09-27 | 2022-06-03 | 深圳市大疆创新科技有限公司 | Pose estimation method and device |
| CN115597535A (en) * | 2022-11-29 | 2023-01-13 | 中国铁路设计集团有限公司（Cn） | High-speed magnetic suspension track irregularity detection system and method based on inertial navigation |
| CN117050760A (en) * | 2023-10-13 | 2023-11-14 | 山西中科冶金建设有限公司 | Intelligent coal charging and coke discharging system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107869989A (en) * | 2017-11-06 | 2018-04-03 | 东北大学 | A kind of localization method and system of the fusion of view-based access control model inertial navigation information |
| CN108665540A (en) * | 2018-03-16 | 2018-10-16 | 浙江工业大学 | Robot localization based on binocular vision feature and IMU information and map structuring system |
| CN109059930A (en) * | 2018-08-31 | 2018-12-21 | 西南交通大学 | A kind of method for positioning mobile robot of view-based access control model odometer |
| CN110044354A (en) * | 2019-03-28 | 2019-07-23 | 东南大学 | A kind of binocular vision indoor positioning and build drawing method and device |
-
2020
- 2020-03-09 CN CN202010157760.9A patent/CN111323009A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107869989A (en) * | 2017-11-06 | 2018-04-03 | 东北大学 | A kind of localization method and system of the fusion of view-based access control model inertial navigation information |
| CN108665540A (en) * | 2018-03-16 | 2018-10-16 | 浙江工业大学 | Robot localization based on binocular vision feature and IMU information and map structuring system |
| CN109059930A (en) * | 2018-08-31 | 2018-12-21 | 西南交通大学 | A kind of method for positioning mobile robot of view-based access control model odometer |
| CN110044354A (en) * | 2019-03-28 | 2019-07-23 | 东南大学 | A kind of binocular vision indoor positioning and build drawing method and device |
Non-Patent Citations (2)
| Title |
|---|
| 李建禹: "基于单目视觉与IMU结合的SLAM技术研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
| 杜义龙: "基于视觉和惯性的移动机器人室内定位算法研究和实现", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114585879A (en) * | 2020-09-27 | 2022-06-03 | 深圳市大疆创新科技有限公司 | Pose estimation method and device |
| CN113390435A (en) * | 2021-05-13 | 2021-09-14 | 中铁二院工程集团有限责任公司 | High-speed railway multi-element auxiliary positioning system |
| CN113390435B (en) * | 2021-05-13 | 2022-08-26 | 中铁二院工程集团有限责任公司 | High-speed railway multi-element auxiliary positioning system |
| CN115597535A (en) * | 2022-11-29 | 2023-01-13 | 中国铁路设计集团有限公司（Cn） | High-speed magnetic suspension track irregularity detection system and method based on inertial navigation |
| CN117050760A (en) * | 2023-10-13 | 2023-11-14 | 山西中科冶金建设有限公司 | Intelligent coal charging and coke discharging system |
| CN117050760B (en) * | 2023-10-13 | 2023-12-15 | 山西中科冶金建设有限公司 | Intelligent coal charging and coke discharging system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111323009A (en) | Magnetic suspension train positioning method and system | |
| CN106092056B (en) | A kind of vehicle-mounted dynamic monitoring method of high-speed railway bridge foundation settlement deformation | |
| Kao | Integration of GPS and dead-reckoning navigation systems | |
| CN101907714B (en) | GPS aided positioning system and method based on multi-sensor data fusion | |
| CN108731670A (en) | Inertia/visual odometry combined navigation locating method based on measurement model optimization | |
| CN107402006B (en) | Based on the matched train precision positioning method of track geometry characteristic information and system | |
| EP0496508B1 (en) | Heading detecting method | |
| CN102128625B (en) | Initial matching method for use in gravimetric map matching in gravity-aided inertial navigation system | |
| CN104819716A (en) | Indoor and outdoor personal navigation algorithm based on INS/GPS (inertial navigation system/global position system) integration of MEMS (micro-electromechanical system) | |
| JP2016180980A (en) | Information processing device, program, and map data updating system | |
| CN201269758Y (en) | Vehicle mounted full automatic detection recording system for traffic signs | |
| CN107247275B (en) | Urban GNSS vulnerability monitoring system and method based on bus | |
| CN108931244A (en) | Ins error suppressing method and system based on train kinematic constraint | |
| CN105865450A (en) | Zero-speed update method and system based on gait | |
| Zhou et al. | Kinematic measurement of the railway track centerline position by GNSS/INS/odometer integration | |
| CN110412596A (en) | A kind of robot localization method based on image information and laser point cloud | |
| CN103644888A (en) | Inertial reference measurement method for detecting bridge deformation | |
| Retscher et al. | Ubiquitous positioning technologies for modern intelligent navigation systems | |
| Zhou et al. | Onboard train localization based on railway track irregularity matching | |
| CN114136275A (en) | Track line state detection device and roadbed settlement detection method | |
| Chen et al. | Experimental study on the potential of vehicle’s attitude response to railway track irregularity in precise train localization | |
| CN110631573A (en) | Multi-information fusion method for inertia/mileometer/total station | |
| CN110658543B (en) | High-speed railway track geometric parameter detection method based on non-contact measurement | |
| CN114162170B (en) | Track measurement system and measurement method | |
| Sahli et al. | Small pipeline trajectory estimation using MEMS based IMU |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |