CN114022556A - Positioning initialization method, device and computer readable storage medium - Google Patents
Positioning initialization method, device and computer readable storage medium Download PDFInfo
- Publication number
- CN114022556A CN114022556A CN202111356493.9A CN202111356493A CN114022556A CN 114022556 A CN114022556 A CN 114022556A CN 202111356493 A CN202111356493 A CN 202111356493A CN 114022556 A CN114022556 A CN 114022556A
- Authority
- CN
- China
- Prior art keywords
- positioning
- state variable
- state
- initialization
- value
- 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
- 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
- G06T7/74—Determining position or orientation of objects or cameras using feature-based methods involving reference images or patches
-
- 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
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Theoretical Computer Science (AREA)
- Automation & Control Theory (AREA)
- Image Analysis (AREA)
- Navigation (AREA)
Abstract
The application discloses a positioning initialization method, a positioning initialization device and a computer readable storage medium, wherein the positioning initialization method comprises the following steps: acquiring prior information of state variables of the positioning equipment in a static state; determining an initial value of the state variable based on prior information of the state variable; wherein the state variables are used for positioning and comprise visual state variables and inertial state variables; updating the state variable in the process of operation positioning; and determining that the positioning initialization is completed under the condition that the updated state variable is determined to meet convergence. By the scheme, the initialization process has high stability.
Description
Technical Field
The present application relates to the field of positioning technologies, and in particular, to a positioning initialization method, an apparatus, and a computer-readable storage medium.
Background
The visual inertial tracking positioning system is an important underlying technology in the fields of computer vision, robots, unmanned vehicles, three-dimensional reconstruction, augmented reality and the like, and the quick initialization method of the visual inertial tracking positioning system has important practical value in the fields of augmented reality, virtual reality and the like.
However, the initialization method of the conventional visual inertial tracking positioning system is complex, and the user needs to be guided by more professional techniques to use the system normally. Generally, the initialization scheme separates visual initialization from inertial sensor initialization, and first, vision needs enough parallax to reconstruct an existing scene, and after the visual initialization is completed, the result of the visual initialization is aligned with the result of the inertial sensor, so that the initialization of an inertial system is completed. Such an initialization process has some drawbacks: on one hand, the device cannot provide enough parallax when being static or rotating, so that the visual initialization cannot be completed; on the other hand, the initialization of the inertial system is not successful with a certain probability under the influence of the visual initialization effect and the inertial sensor data; meanwhile, the success rate of initialization is low for scenes such as outdoor environments with poor textures or large scales.
Disclosure of Invention
The application provides at least one positioning initialization method, device and computer readable storage medium.
A first aspect of the present application provides a method for positioning initialization, where the method includes: acquiring prior information of state variables of the positioning equipment in a static state; determining an initial value of the state variable based on prior information of the state variable; wherein the state variables are used for positioning and comprise visual state variables and inertial state variables; updating the state variable in the process of operation positioning; and determining that the positioning initialization is completed under the condition that the updated state variable is determined to meet convergence.
Therefore, the initial value is set for the state variable by using the prior information of the state variable acquired in the static state, wherein the state variable is used for positioning and comprises the visual state variable and the inertial state variable, and in the running positioning process, the positioning initialization is determined to be completed under the condition that the updated state variable meets the convergence, namely, the visual and inertial systems are initialized at the same time by using the prior information when the equipment is close to the static state, so that the positioning initialization process is rapid and stable, and the user operation is easy.
Wherein the inertial state variables include at least one of gravity, position, angle, velocity, scale, inertial system offset; and/or the visual state variable comprises at least one of three-dimensional point inverse depth, position and angle.
Therefore, since the inertial state variable includes at least one of gravity, position, angle, speed, scale and inertial system offset, and the visual state variable includes at least one of three-dimensional point inverse depth, position and angle, the initial value setting can be performed for each state variable of the system by using the prior information of each state variable acquired in the static state, thereby realizing the initialization of the whole system.
Wherein the determining an initial value of the state variable based on the prior information of the state variable comprises: using the prior information of the inertia state variable as an initial value of the inertia state variable; and/or, the obtaining of the prior information of the state variable of the positioning device in the static state includes at least one of: under the condition that the state variable comprises the speed, determining that the speed is a static speed value, and taking the static speed value as prior information of the inertia state variable; under the condition that the state variable comprises an inertial system offset, acquiring the offset or a preset calibration value in the positioning process before the current positioning initialization, and taking the offset or the preset calibration value in the positioning process before the current positioning initialization as prior information of the inertial system offset; and acquiring the current value of the inertial state variable of the positioning equipment in the static state by using an inertial complementary filter, and taking the acquired current value of the inertial state variable as the prior information of the inertial state variable.
Therefore, the acquired prior information of the inertial state variable of the positioning equipment in the static state is used as the initial value of the inertial state variable, so that the accuracy of the initialization process of the system is higher.
Before the obtaining of the prior information of the state variable of the positioning device in the static state, the method further includes: acquiring two continuous frames of first scene images acquired by the positioning equipment; and under the condition that the parallax values between all the characteristic points in the first scene images of the two continuous frames are smaller than a first preset threshold value, determining that the positioning equipment is in a static state.
Therefore, two continuous frames of first scene images acquired by the positioning equipment are acquired; under the condition that the parallax values of all the feature points in the first scene images of two continuous frames are smaller than the first preset threshold, the positioning device is determined to be in a static state, that is, whether the positioning device is in the static state can be accurately judged, so that a proper initialization moment can be provided for the initialization process.
Wherein, in the process of operating positioning, updating the state variable includes: in the process of operation positioning, acquiring the current value of the state variable; and obtaining the updated value of the state variable based on the current value of the state variable.
Therefore, after the initial value is set for the state variable, in the operation positioning process before the system completes initialization, whether the updated state variable meets convergence can be accurately judged through obtaining the updated value of the state variable, and whether the positioning initialization is completed can be determined.
Wherein, prior to the updating the state variable, the method further comprises: and decoupling the updating of a plurality of state variables.
Therefore, the mutual influence among the state variables can be reduced by performing decoupling processing on the updating of the state variables, so that the convergence speed of a single variable can be accelerated in the operation positioning process after the initial values are set for the state variables and before the system completes initialization, and the system can complete initialization quickly and stably.
Wherein the decoupling the update of the plurality of state variables comprises: determining an initial value confidence coefficient of each state variable, wherein the initial confidence coefficient is used for representing the influence degree of the initial value on the updated state variable; the obtaining of the updated value of the state variable based on the current value of the state variable includes: and obtaining an updated value of the state variable based on the initial value, the initial value confidence and the current value of the state variable.
Therefore, by setting an initial value confidence coefficient for each state variable, wherein the initial confidence coefficient is used for representing the influence degree of the initial value on the updated state variable, the mutual influence among the state variables is reduced, each state variable has independent observability, so that the system can be converged towards the optimal value direction, the fluctuation in the convergence process is small, and the stability of the static initialization and the convergence speed of the system can be improved.
Wherein, in the process of operating positioning, obtaining the current value of the state variable comprises at least one of the following steps: under the condition that the state variable comprises scale information, carrying out motion initialization on the positioning equipment in the operation positioning process, and obtaining the current value of the scale information by using the scale information obtained after the motion initialization is completed; and under the condition that the state variable comprises the three-dimensional point inverse depth, in the running positioning process, acquiring a second scene image acquired by the positioning equipment, and determining a random inverse depth value of a corresponding inverse depth three-dimensional point for a two-dimensional point in the second scene image along the depth direction of the point to serve as a current value of the three-dimensional point inverse depth.
Therefore, in the process of operation positioning, the scale information obtained after the positioning equipment completes the motion initialization is used as the current value of the scale information, so that the scale information can be quickly converged after the static initialization is completed; in addition, the inverse depth value of the inverse depth three-dimensional point is observed in the operation positioning process, so that the stability of the system can be improved.
Wherein, prior to said determining random inverse depth values for corresponding inverse depth three-dimensional points for two-dimensional points in the second scene image along a depth direction of the points, the method further comprises: acquiring device orientation information corresponding to at least two frames of the second scene images; and eliminating the outer points in the second scene image according to the two-dimensional points in the at least two frames of second scene images and the equipment orientation information.
Therefore, according to the two-dimensional points in the at least two frames of second scene images and the equipment orientation information, the outliers in the second scene images can be eliminated, so that the scene images can be optimized, and the stability of system initialization is improved.
Wherein, in the process of operating positioning, updating the state variable further comprises: under the condition that the state variable comprises the speed, judging whether the positioning equipment is in a static state or not in the operation positioning process; taking a static speed value as a current value of the speed under the condition that the positioning equipment is in a static state; based on the current value of the speed, an updated value of the speed is obtained.
Therefore, in the process of positioning operation, under the condition that the positioning equipment is judged to be in a static state, the static speed value is used as the current value of the speed, the speed information of the system is directly restrained, the large offset generated by the system can be effectively restrained, and the stability of system initialization is improved.
Wherein, after updating the state variables during the run positioning, the method further comprises at least one of: acquiring uncertainty corresponding to the updated value of the state variable, and determining that the state variable meets convergence under the condition that the uncertainty of the state variable is lower than a second preset threshold; and determining the pose of the positioning equipment by using the updated value of the state variable.
Therefore, the lower the uncertainty of the state variable is, the better the convergence of the state variable is, so that by acquiring the uncertainty corresponding to the updated value of the state variable and judging whether the uncertainty of the state variable is lower than a second preset threshold, whether the state variable meets the convergence can be determined, and further whether the positioning initialization is completed can be determined; in addition, the pose information of the positioning equipment can be obtained according to the visual state variable and the inertia state variable so as to realize the positioning of the equipment.
In order to solve the above problem, a second aspect of the present application provides a positioning initialization apparatus, including: the acquisition module is used for acquiring prior information of state variables of the positioning equipment in a static state; the setting module is used for determining an initial value of the state variable based on the prior information of the state variable; wherein the state variables are used for positioning and comprise visual state variables and inertial state variables; the updating module is used for updating the state variable in the process of operating and positioning; and the determining module is used for determining that the positioning initialization is completed under the condition that the updated state variable is determined to meet the convergence.
In order to solve the above problem, a third aspect of the present application provides a positioning initialization apparatus, which includes a memory and a processor coupled to each other, where the processor is configured to execute program instructions stored in the memory to implement the positioning initialization method in the first aspect.
In order to solve the above problem, a fourth aspect of the present application provides a computer-readable storage medium having stored thereon program instructions that, when executed by a processor, implement the positioning initialization method in the first aspect described above.
According to the scheme, the priori information of the state variables acquired in the static state is utilized to set the initial values for the state variables, wherein the state variables are used for positioning and comprise the visual state variables and the inertial state variables, and in the operation positioning process, the positioning initialization is determined to be completed under the condition that the updated state variables meet the convergence, namely, the vision and inertial systems are initialized at the same time by utilizing the priori information in the static state approaching equipment, so that the positioning initialization process is rapid and stable, and meanwhile, the user operation is easy.
Drawings
FIG. 1 is a schematic flow chart diagram illustrating an embodiment of a positioning initialization method of the present application;
FIG. 2 is a flowchart illustrating an embodiment of step S13 in FIG. 1;
FIG. 3 is a schematic flow chart diagram illustrating another embodiment of a positioning initialization method according to the present application;
FIG. 4 is a schematic flow chart diagram illustrating an application scenario of the positioning initialization method of the present application;
FIG. 5 is a block diagram of an embodiment of a positioning initialization apparatus according to the present application;
FIG. 6 is a block diagram of another embodiment of the position initialization apparatus of the present application;
FIG. 7 is a block diagram of an embodiment of a computer-readable storage medium of the present application.
Detailed Description
The following describes in detail the embodiments of the present application with reference to the drawings attached hereto.
In the following description, for purposes of explanation and not limitation, specific details are set forth such as particular system structures, interfaces, techniques, etc. in order to provide a thorough understanding of the present application.
The terms "system" and "network" are often used interchangeably herein. The term "and/or" herein is merely an association describing an associated object, meaning that three relationships may exist, e.g., a and/or B, may mean: a exists alone, A and B exist simultaneously, and B exists alone. In addition, the character "/" herein generally indicates that the former and latter related objects are in an "or" relationship. Further, the term "plurality" herein means two or more than two.
Referring to fig. 1, fig. 1 is a schematic flowchart illustrating a positioning initialization method according to an embodiment of the present application. Specifically, the method may include the steps of:
step S11: and acquiring prior information of state variables of the positioning equipment in a static state.
Step S12: determining an initial value of the state variable based on prior information of the state variable. Wherein the state variables are used for positioning and include visual state variables and inertial state variables.
The visual inertial tracking positioning system is an algorithm for realizing SLAM (synchronous positioning and mapping) by fusing camera and IMU (inertial measurement unit) data, and due to the nonlinearity of the visual inertial tracking positioning system, the performance of a sensor of the visual inertial tracking positioning system depends on the accuracy of an initial value seriously no matter a method based on filtering or graph optimization. Once the initialization is poor, not only the convergence speed is reduced, but also an erroneous estimation is caused, and therefore, a robust initialization method is crucial. Generally, in the initialization process of a visual inertial tracking positioning system, firstly, initialization of only vision is carried out to solve the relative pose of a camera; and then, aligning the initial parameters with IMU pre-integration to solve the initialization parameters so as to complete the visual inertia joint initialization process. However, in an actual process, it is difficult for the visual inertial tracking positioning system to obtain an accurate initial state, on one hand, the scale information of the camera cannot be directly observed, and on the other hand, the non-zero acceleration motion is required to initialize the scale information, but when the positioning device is in a static state, the initialization method is ineffective.
The main body of the positioning initialization method may be a positioning initialization apparatus, for example, the positioning initialization method may be executed by a positioning device, a server, or other processing devices, where the positioning device may be a mobile device such as a robot, an unmanned vehicle, or an unmanned aerial vehicle, or may also be a User Equipment (UE), a User terminal, a cordless telephone, a Personal Digital Assistant (PDA), a handheld device, a computing device, a vehicle-mounted device, a wearable device, or the like. In some possible implementations, the location initialization method may be implemented by a processor calling computer readable instructions stored in a memory. The positioning initialization method initializes the positioning device in a static state, and prior information of a state variable of the positioning device in the static state needs to be acquired before initialization. For example, IMU data may be taken as prior information of the inertial state variables, and camera data may also be taken as prior information of the visual state variables. For example, IMU data may be taken as a priori information because both gyroscope and accelerometer bias may be estimated, and the gravity vector may also be observed, where the IMU is substantially free of errors except for some small measurement errors and random walk of the device, and since the time between two frames of images is short, the sensor bias therebetween may be considered as a fixed value, and thus the motion may be estimated a priori before a new frame of image by pre-integration of IMU measurement data. For another example, camera data can be used as prior information, and the present application initializes when the positioning device is in a stationary state, so that the previous frame of visual information can be used as the prior information of the current frame of visual information, and image matching is performed, so that the initialization of the system can be performed very stably under the condition of very slow motion or being stationary. Therefore, the static state in this application may be a state in which the movement of the positioning apparatus is very slow, or may be a state in which the positioning apparatus is completely static.
Specifically, the inertial state variables include at least one of gravity, position, angle, velocity, dimension, inertial system offset; the visual state variable comprises at least one of three-dimensional point inverse depth, position and angle.
In an embodiment, the step S12 specifically includes: and using the prior information of the inertia state variable as an initial value of the inertia state variable. It can be understood that, since the present application is used for initialization when the positioning apparatus is in a static state, the inertial state variable is not changed when the positioning apparatus is in the static state, and therefore, the prior information of the inertial state variable can be used as the initial value of the inertial state variable.
In one implementation scenario, the step S11 includes: and under the condition that the state variable comprises the speed, determining that the speed is a static speed value, and taking the static speed value as the prior information of the inertia state variable. It can be understood that the speed in the stationary state is a stationary speed value, and since the present embodiment is used for initializing the positioning apparatus in the stationary state, the speed of the positioning apparatus is the stationary speed value, and the stationary speed value is used as the prior information of the inertial state variable, and the speed variable has a great confidence, so that the accuracy of the initialization process of the system is higher.
In one implementation scenario, the step S11 includes: and under the condition that the state variable comprises an inertial system offset, acquiring the offset or a preset calibration value in the positioning process before the current positioning initialization, and taking the offset or the preset calibration value in the positioning process before the current positioning initialization as the prior information of the inertial system offset. It can be understood that the IMU has a certain offset, the offset is generally related to temperature, factory accuracy, and the like, and when the positioning device is used for the first time, the inertial system offset may be a preset calibration value. Alternatively, when the positioning system is reinitialized during use, the tracking positioning is already performed before the initialization of the current positioning, so that the offset during the positioning process before the initialization of the current positioning may be used as the inertial system offset for the initialization of the current positioning.
In one implementation scenario, the step S11 includes: and acquiring the current value of the inertial state variable of the positioning equipment in the static state by using an inertial complementary filter, and taking the acquired current value of the inertial state variable as the prior information of the inertial state variable. It can be understood that, in the IMU, the accelerometer has better low-frequency characteristics, because the angle of the acceleration can be directly calculated, and there is no accumulated error, so it is also more accurate after a long time, and the gyroscope can cause larger output error or even cannot be used due to the accumulation of the integrated error after a long time. Therefore, the inertial complementary filter is used, the gyroscope is taken as the main component in a short time, the accelerometer is used more accurately for a long time, the specific gravity of the accelerometer is increased at the moment, different weights are given to the gyroscope and the accelerometer, weighting and summing are carried out, and the real-time value of the inertial state variable is obtained; and when the real-time value of the inertia state variable acquired by the inertia complementary filter is converged and the positioning equipment is in a static state, the current value of the inertia state variable at the moment can be acquired, and the acquired current value of the inertia state variable is used as the prior information of the inertia state variable.
It can be understood that, when the positioning device itself can calculate its gravity direction, and the calculated gravity direction is sufficiently reliable, the gravity direction calculated by the positioning device itself can be used to replace the gravity direction obtained by the above inertial complementary filter, as the prior information of the gravity direction.
Therefore, the accuracy of the initialization process of the system can be higher by taking the acquired prior information of the inertial state variable of the positioning device in the static state, such as the speed, the inertial system offset and the like, as the initial value of the inertial state variable.
Step S13: and updating the state variable in the process of operating positioning.
Step S14: and determining that the positioning initialization is completed under the condition that the updated state variable is determined to meet convergence.
It can be understood that, from the time when the initial value setting is performed on the state variable of the visual inertial tracking positioning system to the time when the convergence of the visual inertial tracking positioning system occurs, the positioning apparatus may be in a stationary state or a moving state, that is, the positioning apparatus is in the operation positioning process. After initial values of state variables of the visual inertial tracking positioning system are set, each state variable updates data in real time and the convergence degree of the state variables changes continuously in the positioning process of the positioning equipment, so that the state variables can be updated, and positioning initialization is determined to be completed under the condition that the updated state variables meet convergence; it should be noted that in the case that all the state variables satisfy convergence, it means that the positioning initialization of the entire visual inertial tracking positioning system is completed.
According to the scheme, the priori information of the state variables acquired in the static state is utilized to set the initial values for the state variables, wherein the state variables are used for positioning and comprise the visual state variables and the inertial state variables, and in the operation positioning process, the positioning initialization is determined to be completed under the condition that the updated state variables meet the convergence, namely, the vision and inertial systems are initialized at the same time by utilizing the priori information in the static state approaching equipment, so that the positioning initialization process is rapid and stable, and meanwhile, the user operation is easy.
Referring to fig. 2, fig. 2 is a schematic flowchart illustrating an embodiment of step S13 in fig. 1. In this embodiment, the step S13 may specifically include the following steps:
step S131: and acquiring the current value of the state variable in the process of operation positioning.
Step S132: and obtaining the updated value of the state variable based on the current value of the state variable.
It can be understood that, during the operation positioning process of the positioning device, for each state variable, there exists a current value of the state variable at each time, the current value of the state variable is an observed value, and after the current value of the state variable is obtained, due to system accumulated errors or observation errors and the like, the obtained current value of the state variable may be inaccurate, so that an updated value of the state variable may be obtained by adopting a filtering optimizer or a non-linear optimizer to perform calculation based on the obtained current value of the state variable and a historical value of the state variable, and the updated value of the state variable is more accurate than the current value of the state variable.
According to the scheme, after the initial value is set for the state variable, in the operation positioning process before the system completes initialization, whether the updated state variable meets convergence can be accurately judged through obtaining the updated value of the state variable, and therefore whether the positioning initialization is completed can be determined.
In an implementation scenario, before the updating the state variable in step S13, the positioning initialization method further includes: and decoupling the updating of a plurality of state variables.
It will be appreciated that before all state variables in the visual inertial tracking positioning system converge, the state variables are coupled together, such as dimensions, gravity direction, inertial sensor offset, etc., which may cause the system to converge in a non-optimal direction or fluctuate significantly during convergence. Therefore, the mutual influence among the state variables can be reduced by performing decoupling processing on the updating of the state variables, so that the convergence speed of a single variable can be accelerated in the operation positioning process after the initial values are set for the state variables and before the system completes initialization, and the system can complete initialization quickly and stably.
In an implementation scenario, the step of performing a decoupling process on the updates of the state variables includes: and determining an initial value confidence of each state variable, wherein the initial confidence is used for representing the influence degree of the initial value on the updated state variable. In this case, the step S132 includes:
step S1321: and obtaining an updated value of the state variable based on the initial value, the initial value confidence and the current value of the state variable.
In practical applications, an initial value confidence may be set for each state variable empirically, and the initial confidence may represent: after initial value setting is carried out on state variables of the visual inertial tracking positioning system, how much weight to believe the initial values or believe current values in a tracking state in the subsequent tracking observation process; in the process of operation positioning, a filtering optimizer or a nonlinear optimizer is adopted, and an updated value corresponding to the current time can be obtained by calculation according to an initial value and an initial value confidence of a state variable and an observed value of the state variable obtained at each time, so that for the current time, a value and a confidence of the state variable at the previous time, namely a historical value and a corresponding confidence of the state variable are obtained, and after a current value of the state variable at the current time is obtained, the filtering optimizer or the nonlinear optimizer is adopted for calculation, and the updated value of the state variable at the current time can be obtained. Therefore, after the updated value of the state variable is obtained, the confidence of the updated value can be obtained, so that the influence of the accumulated error or the observation error of the system can be weakened.
According to the scheme, the initial value confidence coefficient is set for each state variable, wherein the initial confidence coefficient is used for representing the influence degree of the initial value on the updated state variable, so that the mutual influence among the state variables is reduced, each state variable has independent observability, the system can be converged towards the optimal value direction, the fluctuation in the convergence process is small, and the stability of static initialization and the convergence speed of the system can be improved.
In an implementation scenario, the step S13 may include: and under the condition that the state variable comprises scale information, performing motion initialization on the positioning equipment in the operation positioning process, and obtaining the current value of the scale information by using the scale information obtained after the motion initialization is completed. It can be understood that, since the pure visual positioning system cannot solve the problem of the scale information, the IMU can estimate the scale information through its own inertial measurement value, and thus the scale information is obtained by adding the IMU. In the existing motion initialization method, the scale information of the visual inertial tracking positioning system can be obtained more accurately. Therefore, in the process of positioning initialization, the method can be operated simultaneously in the background by combining the existing motion initialization technology, and the system information of the motion initialization, such as more accurate scale information, is applied to the current visual inertial tracking positioning system after the motion initialization is stable, so that the method is beneficial to the rapid convergence of the scale information in the positioning initialization.
In an implementation scenario, the step S13 may include: and under the condition that the state variable comprises the three-dimensional point inverse depth, in the running positioning process, acquiring a second scene image acquired by the positioning equipment, and determining a random inverse depth value of a corresponding inverse depth three-dimensional point for a two-dimensional point in the second scene image along the depth direction of the point to serve as a current value of the three-dimensional point inverse depth.
It can be understood that when the positioning device is in a static state, the three-dimensional information in the scene cannot be obtained correctly, and in the absence of the three-dimensional information of the scene, the visual inertial tracking positioning system has no constraint and is easy to generate offset. Thus, for feature points in the scene image, a random inverse depth value may be given along the depth direction of the point, forming an inverse depth three-dimensional point. Therefore, the current value of the three-dimensional point inverse depth of the inverse-depth three-dimensional point can be acquired at any time in the operation positioning process. When the positioning equipment normally moves, the inverse depth value converges towards the correct direction, and when the positioning equipment does not move or moves a small amount, the inverse depth value can be used as prior information to restrict the system to not generate large deviation, so that the stability of initializing the visual inertial tracking positioning system in an unknown depth scene can be improved. In addition, when the positioning equipment shakes rapidly, the discontinuous information integration of the inertial system can generate an accumulated error rapidly, and the reverse-depth three-dimensional point serving as the three-dimensional information of the environment can simultaneously restrict the current frame information and the historical frame information, so that the accumulated error is reduced; for example, the inverse depth three-dimensional points can be converted into global Euclidean space three-dimensional points, that is, the coordinates of the points and the pose information of the positioning equipment can be decoupled, so that the error accumulation is further reduced, and the stability of the positioning system in a small-amplitude quick shaking scene can be effectively improved.
In an implementation scenario, if the positioning apparatus includes a depth sensor, the random inverse depth value of the inverse depth three-dimensional point may be replaced by a depth value obtained by the depth sensor.
Further, before the determining the random inverse depth values of the corresponding inverse depth three-dimensional points for the two-dimensional points in the second scene image along the depth direction of the points, the localization initialization method further includes: acquiring device orientation information corresponding to at least two frames of the second scene images; and eliminating the outer points in the second scene image according to the two-dimensional points in the at least two frames of second scene images and the equipment orientation information.
It can be understood that, in order to obtain a relatively accurate scene image, so that accurate scene image information is obtained in the process of positioning initialization of the system, on the premise that the device orientation information of at least two frames of scene images is known, a two-point random sample consistency detection method can be adopted to eliminate obvious outliers in the scene images. The orientation information of the equipment can be obtained through calculation of the inertia complementary filter, so that outliers in the scene images can be removed according to the two-dimensional points in the at least two frames of scene images and the orientation information of the equipment, the scene images can be optimized, and the stability of positioning initialization of the positioning equipment in the dynamic scene is improved.
In an implementation scenario, the step S13 may further include: under the condition that the state variable comprises the speed, judging whether the positioning equipment is in a static state or not in the operation positioning process; taking a static speed value as a current value of the speed under the condition that the positioning equipment is in a static state; based on the current value of the speed, an updated value of the speed is obtained. Therefore, in the process of positioning operation, under the condition that the positioning equipment is judged to be in a static state, the static speed value is used as the current value of the speed, and the updated value of the speed is obtained by utilizing a filtering optimizer or a nonlinear optimizer for calculation, so that the static speed value is used as the current value of the speed under the condition that the positioning equipment is in the static state, the speed information of the system can be directly constrained, the large offset generated by the system can be effectively inhibited, and the stability of system initialization is improved.
In an implementation scenario, after step S13, the positioning initialization method may further include: and acquiring uncertainty corresponding to the updated value of the state variable, and determining that the state variable meets convergence under the condition that the uncertainty of the state variable is lower than a second preset threshold value. It can be understood that, in the process of positioning operation, while maintaining one state variable, the correlation between the state variable and other state variables is maintained, that is, the covariance matrix of the state variable is obtained from the covariance matrix and state variable information, the smaller the information entropy is, the lower the uncertainty of the state variable is, and the lower the uncertainty is, the better the convergence of the state variable is. Therefore, by obtaining the uncertainty corresponding to the updated value of the state variable and judging whether the uncertainty of the state variable is lower than a second preset threshold value, whether the state variable meets convergence can be determined, and whether positioning initialization is completed can be determined.
In an implementation scenario, after step S13, the positioning initialization method may further include: and determining the pose of the positioning equipment by using the updated value of the state variable. It can be understood that the pose information of the positioning device can be obtained according to the visual state variable and the inertial state variable, so that the positioning of the device can be realized.
Referring to fig. 3, fig. 3 is a schematic flowchart illustrating a positioning initialization method according to another embodiment of the present application. Specifically, the method may include the steps of:
step S31: and acquiring two continuous frames of first scene images acquired by the positioning equipment.
Step S32: and under the condition that the parallax values between all the characteristic points in the first scene images of the two continuous frames are smaller than a first preset threshold value, determining that the positioning equipment is in a static state.
In this embodiment, whether the positioning device is in a static state may be determined by using the visual information, and specifically, it may be determined that the positioning device is in the static state by obtaining two consecutive frames of first scene images acquired by the positioning device, under a condition that it is determined that disparity values between all feature points in the two consecutive frames of first scene images are smaller than a first preset threshold, that is, it may be accurately determined whether the positioning device is in the static state, so that a suitable initialization time may be provided for an initialization process.
Step S33: and acquiring prior information of state variables of the positioning equipment in a static state.
Step S34: determining an initial value of the state variable based on prior information of the state variable; wherein the state variables are used for positioning and include visual state variables and inertial state variables.
Step S35: and updating the state variable in the process of operating positioning.
Step S36: and determining that the positioning initialization is completed under the condition that the updated state variable is determined to meet convergence.
In this embodiment, steps S33-S36 are substantially similar to steps S11-S14 of the above embodiments of the present application, and are not repeated herein.
Referring to fig. 4, fig. 4 is a schematic flowchart of an application scenario of the positioning initialization method of the present application. In an application scenario, the positioning initialization method of the positioning device is divided into three stages, including an initialization preparation stage, an initialization stage and an initialization completion stage. In the initialization preparation stage, mainly for obtaining a priori information of positioning initialization and selecting a proper initialization time, for example, the inertial complementary filter may be used to estimate the gravity direction as the a priori information of the gravity direction, and then after the inertial filter converges, the visual information is used to determine whether the positioning device is in a stationary state, and if it is determined that the positioning device is in the stationary state, the initialization stage may be entered. In the initialization stage, system initialization is mainly carried out, namely initial values of state variables of the system are set by using prior information, and the state variables in the system are decoupled by setting confidence degrees of the state variables; for example, setting initial values of the state variables of the system includes: setting a gravity vector, wherein the gravity direction estimated in the initialization preparation stage can be used as the gravity direction in the gravity vector; setting a speed value, and initializing the speed value by using zero speed; setting an inertial system offset, wherein a stored value or a preset calibration value of the previous system in stable operation can be used; before the convergence of the whole system, state variables such as the dimension, the gravity direction, the bias of the inertial sensor and the like are coupled together, and the convergence of the system towards a non-optimal value direction or large fluctuation in the convergence process can be promoted. In the initialization completion stage, a plurality of strategies are mainly adopted to increase the system stability from the initialization stage to before the system convergence; for example, after initial value setting is performed on a state variable of a system, and in a process before convergence of the system, in order to obtain a relatively accurate scene image, so that accurate scene image information is obtained in a process of positioning initialization of the system, on the premise that device orientation information of two frames of scene images is known, an obvious outlier in the scene image can be removed by adopting a two-point random sample consistency detection method; for example, an inverse depth three-dimensional point is newly added and maintained, specifically, for a feature point in a scene image, a random inverse depth value can be given along the depth direction of the point to form the inverse depth three-dimensional point, through the inverse depth value of the inverse depth three-dimensional point, when a positioning device normally moves, a system can be converged towards the correct direction, when the positioning device does not move or moves a small amount of time, the system can be constrained as prior information, so that a large deviation can not be generated, when the positioning device rapidly shakes, the inverse depth three-dimensional point as three-dimensional information of an environment can simultaneously constrain current frame information and historical frame information, so that an accumulated error is reduced; for example, zero-speed updating is performed, specifically, before system convergence, by taking a stationary speed value as an updated value of speed under the condition that the positioning device is judged to be in a stationary state, the speed information of the system is directly constrained, a large offset generated by the system can be effectively inhibited, and the stability of system initialization is improved; finally, the pose information of the positioning equipment can be output according to the state variables, so that the positioning of the equipment can be realized.
Referring to fig. 5, fig. 5 is a schematic diagram of a positioning initialization apparatus according to an embodiment of the present application. The positioning initialization device 50 includes: an obtaining module 500, configured to obtain prior information of a state variable of a positioning device in a static state; a setting module 502, configured to determine an initial value of the state variable based on prior information of the state variable; wherein the state variables are used for positioning and comprise visual state variables and inertial state variables; an update module 504, configured to update the state variable in a positioning process; a determining module 506, configured to determine that the positioning initialization is completed when it is determined that the updated state variable satisfies convergence.
In the above solution, the setting module 502 sets an initial value for the state variable by using the prior information of the state variable acquired by the acquiring module 500 in the static state, where the state variable is used for positioning and includes a visual state variable and an inertial state variable, and in the operation positioning process, the determining module 506 determines that the positioning initialization is completed under the condition that it is determined that the updated state variable satisfies convergence, that is, it is implemented that the visual and inertial systems are initialized at the same time by using the prior information when the device approaches the static state, so that the positioning initialization process is fast and stable, and the user operation is easy.
In some embodiments, the setting module 502 is specifically configured to use the a priori information of the inertial state variable as the initial value of the inertial state variable. At this time, the obtaining module 500 may be configured to: determining the speed as a stationary speed value in case the state variable comprises a speed; and/or under the condition that the state variable comprises an inertial system offset, acquiring the offset or a preset calibration value in the positioning process before the current positioning initialization, and taking the offset or the preset calibration value in the positioning process before the current positioning initialization as the prior information of the inertial system offset; and/or acquiring the current value of the inertial state variable of the positioning equipment in the static state by using an inertial complementary filter, and taking the acquired current value of the inertial state variable as the prior information of the inertial state variable.
In some embodiments, the positioning initialization apparatus 50 further includes: the judging module is used for acquiring two continuous frames of first scene images acquired by the positioning equipment; and under the condition that the parallax values between all the characteristic points in the first scene images of the two continuous frames are smaller than a first preset threshold value, determining that the positioning equipment is in a static state.
In some embodiments, the updating module 504 is specifically configured to obtain a current value of the state variable in a positioning process; and obtaining the updated value of the state variable based on the current value of the state variable.
In some embodiments, the setting module 502 is further configured to perform a decoupling process on updates of a number of the state variables; specifically, the setting module 502 is configured to determine an initial value confidence of each state variable, where the initial confidence is used to indicate a degree of influence of the initial value on the updated state variable. At this time, the updating module 504 is configured to obtain an updated value of the state variable based on the initial value, the initial value confidence and the current value of the state variable.
In some embodiments, the updating module 504 executes the step of obtaining the current value of the state variable in the positioning running process, and may specifically include at least one of the following steps: under the condition that the state variable comprises scale information, carrying out motion initialization on the positioning equipment in the operation positioning process, and obtaining the current value of the scale information by using the scale information obtained after the motion initialization is completed; and under the condition that the state variable comprises the three-dimensional point inverse depth, in the running positioning process, acquiring a second scene image acquired by the positioning equipment, and determining a random inverse depth value of a corresponding inverse depth three-dimensional point for a two-dimensional point in the second scene image along the depth direction of the point to serve as a current value of the three-dimensional point inverse depth.
In some embodiments, the updating module 504 is further configured to obtain device orientation information corresponding to at least two frames of the second scene image; and eliminating the outer points in the second scene image according to the two-dimensional points in the at least two frames of second scene images and the equipment orientation information.
In some embodiments, the updating module 504 is further configured to determine whether the positioning device is in a stationary state during the positioning operation if the state variable includes a speed; in the case where the pointing device is in a stationary state, a stationary speed value is taken as an updated value of the speed.
In some embodiments, the determining module 506 may be further configured to obtain an uncertainty corresponding to the updated value of the state variable, and determine that the state variable satisfies convergence when the uncertainty of the state variable is lower than a second preset threshold; and/or determining the pose of the positioning equipment by using the updated value of the state variable.
Referring to fig. 6, fig. 6 is a schematic diagram of a positioning initialization apparatus according to another embodiment of the present application. The positioning initialization apparatus 60 includes a memory 61 and a processor 62 coupled to each other, and the processor 62 is configured to execute program instructions stored in the memory 61 to implement the steps of any of the above-mentioned embodiments of the positioning initialization method. In one specific implementation scenario, the positioning initialization device 60 may include, but is not limited to: microcomputer, server.
In particular, the processor 62 is configured to control itself and the memory 61 to implement the steps in any of the above-described embodiments of the positioning initialization method. The processor 62 may also be referred to as a CPU (Central Processing Unit). The processor 62 may be an integrated circuit chip having signal processing capabilities. The Processor 62 may also be a general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic, discrete hardware components. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. In addition, the processor 62 may be collectively implemented by an integrated circuit chip.
According to the scheme, the processor sets the initial value for the state variable by using the prior information of the state variable acquired in the static state, wherein the state variable is used for positioning and comprises the visual state variable and the inertial state variable, and in the operation positioning process, the positioning initialization is determined to be completed under the condition that the updated state variable meets the convergence requirement, namely, the visual and inertial systems are initialized at the same time by using the prior information in the static state approaching equipment, so that the positioning initialization process is rapid and stable, and meanwhile, the user operation is easy.
Referring to fig. 7, fig. 7 is a block diagram illustrating an embodiment of a computer-readable storage medium according to the present application. The computer readable storage medium 70 stores program instructions 700 capable of being executed by the processor, the program instructions 700 being for implementing the steps in any of the above-described embodiments of the location initialization method.
The disclosure relates to the field of augmented reality, and aims to detect or identify relevant features, states and attributes of a target object by means of various visual correlation algorithms by acquiring image information of the target object in a real environment, so as to obtain an AR effect combining virtual and reality matched with specific applications. For example, the target object may relate to a face, a limb, a gesture, an action, etc. associated with a human body, or a marker, a marker associated with an object, or a sand table, a display area, a display item, etc. associated with a venue or a place. The vision-related algorithms may involve visual localization, SLAM, three-dimensional reconstruction, image registration, background segmentation, key point extraction and tracking of objects, pose or depth detection of objects, and the like. The specific application can not only relate to interactive scenes such as navigation, explanation, reconstruction, virtual effect superposition display and the like related to real scenes or articles, but also relate to special effect treatment related to people, such as interactive scenes such as makeup beautification, limb beautification, special effect display, virtual model display and the like.
The detection or identification processing of the relevant characteristics, states and attributes of the target object can be realized through the convolutional neural network. The convolutional neural network is a network model obtained by performing model training based on a deep learning framework.
In the several embodiments provided in the present application, it should be understood that the disclosed method and apparatus may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, a division of a module or a unit is merely one type of logical division, and an actual implementation may have another division, for example, a unit or a component may be combined or integrated with another system, or some features may be omitted, or not implemented. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or units through some interfaces, and may be in an electrical, mechanical or other form.
Units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on network elements. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.
In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be substantially implemented or contributed to by the prior art, or all or part of the technical solution may be embodied in a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, a network device, or the like) or a processor (processor) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
Claims (14)
1. A method for location initialization, the method comprising:
acquiring prior information of state variables of the positioning equipment in a static state;
determining an initial value of the state variable based on prior information of the state variable; wherein the state variables are used for positioning and comprise visual state variables and inertial state variables;
updating the state variable in the process of operation positioning;
and determining that the positioning initialization is completed under the condition that the updated state variable is determined to meet convergence.
2. The method of claim 1, wherein the inertial state variables include at least one of gravity, position, angle, velocity, scale, inertial system bias;
and/or the visual state variable comprises at least one of three-dimensional point inverse depth, position and angle.
3. The method of claim 1 or 2, wherein the determining an initial value of the state variable based on the a priori information of the state variable comprises:
using the prior information of the inertia state variable as an initial value of the inertia state variable;
and/or, the obtaining of the prior information of the state variable of the positioning device in the static state includes at least one of:
under the condition that the state variable comprises the speed, determining that the speed is a static speed value, and taking the static speed value as prior information of the inertia state variable;
under the condition that the state variable comprises an inertial system offset, acquiring the offset or a preset calibration value in the positioning process before the current positioning initialization, and taking the offset or the preset calibration value in the positioning process before the current positioning initialization as prior information of the inertial system offset;
and acquiring the current value of the inertial state variable of the positioning equipment in the static state by using an inertial complementary filter, and taking the acquired current value of the inertial state variable as the prior information of the inertial state variable.
4. The method according to any one of claims 1 to 3, wherein before said obtaining a priori information of the state variables of the positioning device in a stationary state, the method further comprises:
acquiring two continuous frames of first scene images acquired by the positioning equipment;
and under the condition that the parallax values between all the characteristic points in the first scene images of the two continuous frames are smaller than a first preset threshold value, determining that the positioning equipment is in a static state.
5. The method according to any one of claims 1 to 4, wherein the updating the state variable during the running positioning comprises:
in the process of operation positioning, acquiring the current value of the state variable;
and obtaining the updated value of the state variable based on the current value of the state variable.
6. The method of claim 5, wherein prior to said updating said state variable, said method further comprises:
and decoupling the updating of a plurality of state variables.
7. The method of claim 6, wherein the decoupling the updates of the number of state variables comprises:
determining an initial value confidence coefficient of each state variable, wherein the initial confidence coefficient is used for representing the influence degree of the initial value on the updated state variable;
the obtaining of the updated value of the state variable based on the current value of the state variable includes:
and obtaining an updated value of the state variable based on the initial value, the initial value confidence and the current value of the state variable.
8. The method according to any one of claims 5 to 7, wherein said obtaining the current value of the state variable during the run positioning comprises at least one of the following steps:
under the condition that the state variable comprises scale information, carrying out motion initialization on the positioning equipment in the operation positioning process, and obtaining the current value of the scale information by using the scale information obtained after the motion initialization is completed;
and under the condition that the state variable comprises the three-dimensional point inverse depth, in the running positioning process, acquiring a second scene image acquired by the positioning equipment, and determining a random inverse depth value of a corresponding inverse depth three-dimensional point for a two-dimensional point in the second scene image along the depth direction of the point to serve as a current value of the three-dimensional point inverse depth.
9. The method of claim 8, wherein prior to determining the random inverse depth values for the corresponding inverse depth three-dimensional points for the two-dimensional points in the second scene image along the depth direction of the points, the method further comprises:
acquiring device orientation information corresponding to at least two frames of the second scene images;
and eliminating the outer points in the second scene image according to the two-dimensional points in the at least two frames of second scene images and the equipment orientation information.
10. The method of claim 5, wherein updating the state variable during the run positioning further comprises:
under the condition that the state variable comprises the speed, judging whether the positioning equipment is in a static state or not in the operation positioning process;
taking a static speed value as a current value of the speed under the condition that the positioning equipment is in a static state;
based on the current value of the speed, an updated value of the speed is obtained.
11. Method according to any of claims 1 to 10, characterized in that after updating the state variables during the running fix, the method further comprises at least one of the following steps:
acquiring uncertainty corresponding to the updated value of the state variable, and determining that the state variable meets convergence under the condition that the uncertainty of the state variable is lower than a second preset threshold;
and determining the pose of the positioning equipment by using the updated value of the state variable.
12. A position initialization apparatus, comprising:
the acquisition module is used for acquiring prior information of state variables of the positioning equipment in a static state;
the setting module is used for determining an initial value of the state variable based on the prior information of the state variable; wherein the state variables are used for positioning and comprise visual state variables and inertial state variables;
the updating module is used for updating the state variable in the process of operating and positioning;
and the determining module is used for determining that the positioning initialization is completed under the condition that the updated state variable is determined to meet the convergence.
13. A positioning initialization apparatus comprising a memory and a processor coupled to each other, the processor being configured to execute program instructions stored in the memory to implement the positioning initialization method according to any one of claims 1 to 11.
14. A computer-readable storage medium having stored thereon program instructions, which when executed by a processor implement the positioning initialization method of any one of claims 1 to 11.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111356493.9A CN114022556A (en) | 2021-11-16 | 2021-11-16 | Positioning initialization method, device and computer readable storage medium |
PCT/CN2022/098214 WO2023087681A1 (en) | 2021-11-16 | 2022-06-10 | Positioning initialization method and apparatus, and computer-readable storage medium and computer program product |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111356493.9A CN114022556A (en) | 2021-11-16 | 2021-11-16 | Positioning initialization method, device and computer readable storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114022556A true CN114022556A (en) | 2022-02-08 |
Family
ID=80064583
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111356493.9A Pending CN114022556A (en) | 2021-11-16 | 2021-11-16 | Positioning initialization method, device and computer readable storage medium |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN114022556A (en) |
WO (1) | WO2023087681A1 (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114993304A (en) * | 2022-06-29 | 2022-09-02 | 北京石头创新科技有限公司 | Method, device, medium and electronic equipment for determining position of robot |
WO2023087681A1 (en) * | 2021-11-16 | 2023-05-25 | 上海商汤智能科技有限公司 | Positioning initialization method and apparatus, and computer-readable storage medium and computer program product |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10043076B1 (en) * | 2016-08-29 | 2018-08-07 | PerceptIn, Inc. | Visual-inertial positional awareness for autonomous and non-autonomous tracking |
CN109238277B (en) * | 2018-08-01 | 2020-10-27 | 清华大学 | Positioning method and device for visual inertial data depth fusion |
CN112798010B (en) * | 2019-11-13 | 2023-05-09 | 北京三快在线科技有限公司 | Initializing method and device of VIO system of visual inertial odometer |
CN113052897A (en) * | 2021-03-25 | 2021-06-29 | 浙江商汤科技开发有限公司 | Positioning initialization method and related device, equipment and storage medium |
CN114022556A (en) * | 2021-11-16 | 2022-02-08 | 浙江商汤科技开发有限公司 | Positioning initialization method, device and computer readable storage medium |
-
2021
- 2021-11-16 CN CN202111356493.9A patent/CN114022556A/en active Pending
-
2022
- 2022-06-10 WO PCT/CN2022/098214 patent/WO2023087681A1/en unknown
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2023087681A1 (en) * | 2021-11-16 | 2023-05-25 | 上海商汤智能科技有限公司 | Positioning initialization method and apparatus, and computer-readable storage medium and computer program product |
CN114993304A (en) * | 2022-06-29 | 2022-09-02 | 北京石头创新科技有限公司 | Method, device, medium and electronic equipment for determining position of robot |
Also Published As
Publication number | Publication date |
---|---|
WO2023087681A1 (en) | 2023-05-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109029433B (en) | Method for calibrating external parameters and time sequence based on vision and inertial navigation fusion SLAM on mobile platform | |
CN111156984B (en) | Monocular vision inertia SLAM method oriented to dynamic scene | |
Rambach et al. | Learning to fuse: A deep learning approach to visual-inertial camera pose estimation | |
US20210183100A1 (en) | Data processing method and apparatus | |
US20220051031A1 (en) | Moving object tracking method and apparatus | |
JP7131994B2 (en) | Self-position estimation device, self-position estimation method, self-position estimation program, learning device, learning method and learning program | |
RU2662913C2 (en) | Method of robot localization in localization plane | |
US10636190B2 (en) | Methods and systems for exploiting per-pixel motion conflicts to extract primary and secondary motions in augmented reality systems | |
TWI795885B (en) | Visual positioning method, device and computer-readable storage medium | |
CN110533719B (en) | Augmented reality positioning method and device based on environment visual feature point identification technology | |
WO2023087681A1 (en) | Positioning initialization method and apparatus, and computer-readable storage medium and computer program product | |
US11922658B2 (en) | Pose tracking method, pose tracking device and electronic device | |
KR101985344B1 (en) | Sliding windows based structure-less localization method using inertial and single optical sensor, recording medium and device for performing the method | |
WO2023050634A1 (en) | Positioning method and apparatus, device, storage medium, and computer program product | |
CN113066127B (en) | Visual inertial odometer method and system for calibrating equipment parameters on line | |
CN110749308B (en) | SLAM-oriented outdoor positioning method using consumer-grade GPS and 2.5D building models | |
WO2022188334A1 (en) | Positioning initialization method and apparatus, device, storage medium, and program product | |
CN113984068A (en) | Positioning method, positioning apparatus, and computer-readable storage medium | |
CN114565728A (en) | Map construction method, pose determination method, related device and equipment | |
CN113052897A (en) | Positioning initialization method and related device, equipment and storage medium | |
CN113327270A (en) | Visual inertial navigation method, device, equipment and computer readable storage medium | |
CN113916223B (en) | Positioning method and device, equipment and storage medium | |
CN113847907A (en) | Positioning method and device, equipment and storage medium | |
US11158119B2 (en) | Systems and methods for reconstructing a three-dimensional object | |
Keshavan et al. | An optical flow-based solution to the problem of range identification in perspective vision systems |
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 |