A method and device for collecting inertial navigation data

Abstract

The present invention relates to a method and device for collecting inertial navigation data, including: obtaining initial navigation data of an object to be collected at a first moment; obtaining real-time navigation data of the object to be collected at the second moment and the error of the real-time navigation data, as well as the influence value of the change in the motion state of the object to be collected caused by the real-time navigation data and the error of the real-time navigation data; obtaining a correction value output after processing a first image through an image correction model; the first image is an image of the environment where the object to be collected is located at the second moment; based on the influence value and the correction value, obtaining updated navigation data for representing the updated motion state of the object to be collected; according to the initial navigation data and the updated navigation data, determining and collecting inertial navigation data for representing the motion state of the object to be collected at the target moment. The present invention can improve the accuracy of inertial navigation data collection.

Description

Technical Field

[0001] The present invention relates to the technical field of navigation, and particularly relates to a method, an apparatus, an electronic device, and a non-transitory computer-readable storage medium for collecting inertial navigation data. Background Art

[0002] Inertial navigation data is a series of information about the motion state of an object obtained through an inertial navigation system, such as the position, velocity, acceleration, angle, and angular velocity of the object. Currently, inertial navigation data is mainly collected through inertial navigation sensors (such as accelerometers and gyroscopes). The above sensors infer the motion state of an object by measuring its acceleration and angular velocity, providing real-time position and orientation information for the navigation system.

[0003] However, traditional inertial navigation systems have some defects. First, due to the noise and errors in the sensors themselves, traditional inertial navigation systems will accumulate errors during long-term use, reducing the accuracy of collecting inertial navigation data. Second, inertial navigation systems cannot overcome temporary interferences, such as sudden braking or sudden turning, which easily lead to system drift and then collect incorrect inertial navigation data. In addition, existing inertial navigation systems lack the perception of the surrounding environment during data collection, making the system vulnerable to changes in the external environment and thus equally likely to collect incorrect inertial navigation data.

[0004] Therefore, the existing inertial navigation systems have low accuracy in collecting inertial navigation data and cannot collect accurate inertial navigation data. Summary of the Invention

[0005] The present invention aims at the technical problems existing in the prior art and provides a method, an apparatus, an electronic device, and a non-transitory computer-readable storage medium for collecting inertial navigation data that can improve the accuracy and robustness of collecting inertial navigation data.

[0006] The technical solution of the present invention for solving the above technical problems is as follows:

[0007] The present invention provides a method for collecting inertial navigation data, the method comprising:

[0008] Obtaining initial navigation data of an object to be collected at a first moment;

[0009] Obtaining real-time navigation data of the object to be collected at a second moment, the error of the real-time navigation data, and the influence value of the real-time navigation data and the error of the real-time navigation data on the change of the motion state of the object to be collected;

[0010] Obtain the correction value output after processing the first image by the image correction model; the first image is an image of the environment where the object to be collected is located at the second moment;

[0011] Based on the influence value and the correction value, obtain updated navigation data for representing the updated motion state of the object to be collected;

[0012] According to the initial navigation data and the updated navigation data, determine and collect inertial navigation data for representing the motion state of the object to be collected at the target moment.

[0013] Optionally, the inertial navigation data includes speed update data, angle update data, and angular velocity update data;

[0014] The obtaining of the influence value of the change in the motion state of the object to be collected caused by the real-time navigation data and the error of the real-time navigation data includes:

[0015] For each type of the inertial navigation data, obtain the first data and the second data in the real-time navigation data;

[0016] Obtain the first error term corresponding to the first data and the second error term corresponding to the second data;

[0017] Perform fusion processing on the first data, the second data, the first error term, and the second error term, and then perform integration processing on the fusion feature to obtain the influence value.

[0018] Optionally, the performing of fusion processing on the first data, the second data, the first error term, and the second error term to obtain a fusion feature, and then performing integration processing on the fusion feature to obtain the influence value includes:

[0019] Perform square root processing on the second data to obtain second square root data;

[0020] Perform fusion processing on the second square root data and the second error term to obtain a first influence value representing the change in the motion state of the object to be collected caused by the acquisition error of the second data;

[0021] Perform fusion processing on the first data, the first error term, and the first influence value and then perform integration processing to obtain the influence value.

[0022] Optionally, the obtaining of the correction value output after processing the first image by the image correction model includes:

[0023] For each type of the inertial navigation data, obtain the image information corresponding to the inertial navigation data;

[0024] Integrate the image information to obtain the cumulative value corresponding to the image information;

[0025] Call the image correction model to correct the cumulative value to obtain the correction result corresponding to the image information; the correction result is the weight or offset term for correcting the image information;

[0026] Apply the correction result to the image information to obtain the corrected image information as the correction value.

[0027] Optionally, for each type of the inertial navigation data, obtaining the image information corresponding to the inertial navigation data includes:

[0028] Call an image sensor to obtain an initial image that meets the preset acquisition conditions; the initial image includes video frames and static images;

[0029] Preprocess the initial image to obtain a preprocessed image;

[0030] Extract the motion features related to the object to be collected and the inertial navigation data from the preprocessed image;

[0031] Perform a fusion process on the motion features to synthesize the image information corresponding to each type of the inertial navigation data.

[0032] Optionally, the inertial navigation data includes position update data, and the method further includes:

[0033] Obtain the horizontal motion state of the object to be collected at a second moment; the second moment is any moment within the range interval from the first moment to the target moment;

[0034] Based on the influence value, the correction value, and the horizontal motion state, obtain first updated navigation data for representing the updated motion state of the object to be collected;

[0035] Determine and collect the position update data of the object to be collected at the target moment according to the initial navigation data and the first updated navigation data.

[0036] Optionally, the position update data of the object to be collected at the target moment is represented as:

[0037]

[0038] where P(t + △t) is the position update data, P(t) is the position of the object to be collected at the first moment, V(τ)·cos(θ(τ) is the horizontal motion state, is the influence value, is the correction value

[0039] The present invention also provides a device for collecting inertial navigation data, the device comprising:

[0040] A first acquisition module, configured to acquire initial navigation data of an object to be collected at a first moment;

[0041] A second acquisition module, configured to acquire real-time navigation data of the object to be collected at the second moment and an error of the real-time navigation data, and an influence value of a change in the motion state of the object to be collected caused by the real-time navigation data and the error of the real-time navigation data;

[0042] A third acquisition module, configured to acquire a correction value output after processing a first image through an image correction model; the first image is an image of the environment where the object to be collected is located at the second moment;

[0043] An updated navigation data module, configured to obtain updated navigation data for representing an updated motion state of the object to be collected based on the influence value and the correction value;

[0044] A data determination module, configured to determine and collect inertial navigation data for representing the motion state of the object to be collected at the target moment according to the initial navigation data and the updated navigation data.

[0045] In addition, to achieve the above object, the present invention also provides an electronic device, comprising: a memory, configured to store a computer software program; a processor, configured to read and execute the computer software program, thereby implementing the method for collecting inertial navigation data as described above.

[0046] In addition, to achieve the above object, the present invention also provides a non-transitory computer-readable storage medium, in which a computer software program is stored, and when the computer software program is executed by a processor, the method for collecting inertial navigation data as described above is implemented.

[0047] The beneficial effects of the present invention are:

[0048] (1) By introducing a correction value output by an image correction model, the present invention enables the image correction model to learn and identify an acquisition mode for counteracting environmental influences (such as sudden braking) in the acquisition system, and perform real-time correction on sensor measurement errors and system drift, so that the acquisition system can better resist the accumulation of errors during long-term use, thereby improving the accuracy of the collected inertial navigation data;

[0049] (2) The present invention uses an image sensor to obtain visual information of the surrounding environment of an object, which can provide additional reference for the acquisition system. By processing the image and extracting features related to the target position and movement, the combination of the image sensor information and the neural network correction term effectively reduces the sensitivity of the acquisition system to temporary interference, thereby improving the robustness of the acquired inertial navigation data.

[0050] (3) By introducing the error term of the navigation data, the present invention comprehensively considers the uncertainty of the sensor measurements of the traditional inertial navigation system, which helps to more accurately simulate the possible errors during actual acquisition, and improves the accuracy and robustness of the acquired inertial navigation data. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a scene diagram of a method for acquiring inertial navigation data provided by the present invention;

[0052] Figure 2 It is a flowchart of a method for acquiring inertial navigation data provided by the present invention;

[0053] Figure 3 It is a schematic structural diagram of an apparatus for acquiring inertial navigation data provided by the present invention;

[0054] Figure 4 It is a schematic hardware structure diagram of a possible electronic device provided by the present invention;

[0055] Figure 5 It is a schematic hardware structure diagram of a possible computer-readable storage medium provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0056] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts fall within the scope of protection of the present application.

[0057] In the description of the present application, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of the described features. In the description of the present application, "a plurality" means two or more, unless otherwise specifically defined.

[0058] In the description of the present application, the term "for example" is used to mean "serving as an example, illustration, or explanation". Any embodiment described as "for example" in the present application is not necessarily construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to implement and use the present invention. In the following description, details are set forth for purposes of explanation. It should be understood that those of ordinary skill in the art can recognize that the present invention can be implemented without the use of these specific details. In other instances, well-known structures and processes are not elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but rather to be in line with the broadest scope consistent with the principles and features disclosed in the present application.

[0059] Please refer to Figure 1 , Figure 1 which is a scenario diagram of a method for collecting inertial navigation data provided by the present invention. As Figure 1 shown, the terminal and the server are connected through a network, for example, through a wired or wireless network connection, etc. Among them, the terminal may include, but is not limited to, portable terminals such as mobile phones and tablets installed with various network platform applications, as well as fixed terminals such as computers, query machines, and advertising machines. Among them, the server provides various business services for users, including service push servers, user recommendation servers, etc.

[0060] It should be noted that Figure 1 the application scenario schematic diagram of the drone integrated detection system based on multi-source intelligence fusion shown is only an example. The terminal, server, and application scenario described in the embodiments of the present application are for more clearly explaining the technical solutions of the embodiments of the present application, and do not generate limitations on the technical solutions provided by the embodiments of the present application. Those of ordinary skill in the art can know that with the evolution of the system and the emergence of new business scenarios, the technical solutions provided by the embodiments of the present application are equally applicable to similar technical problems.

[0061] Among them, the terminal can be used for:

[0062] Obtaining the initial navigation data of the object to be collected at the first moment;

[0063] Obtaining the real-time navigation data of the object to be collected at the second moment and the error of the real-time navigation data, and the influence value of the change in the motion state of the object to be collected caused by the real-time navigation data and the error of the real-time navigation data;

[0064] Obtaining the correction value output after processing the first image through an image correction model; the first image is an image of the environment where the object to be collected is located at the second moment;

[0065] Based on the influence value and the correction value, obtain updated navigation data for representing the updated motion state of the object to be collected.

[0066] According to the initial navigation data and the updated navigation data, determine and collect inertial navigation data for representing the motion state of the object to be collected at the target moment.

[0067] It should be noted that the steps of the above method for the terminal to collect inertial navigation data can also be executed by the server.

[0068] Please refer to Figure 2 , which provides a flowchart of a method for collecting inertial navigation data according to the present invention, including the following steps:

[0069] Step 201, obtain the initial navigation data of the object to be collected at the first moment.

[0070] Wherein, the object to be collected refers to the object for which inertial navigation data is to be collected. For example, the object to be collected can be a vehicle, an aircraft, etc. In some embodiments, the initial navigation data can be the speed, acceleration, and angular velocity of the object to be collected.

[0071] In some embodiments, an inertial navigation sensor (such as an accelerometer, a gyroscope, etc.) can be used to measure the acceleration A(t) and angular velocity ω(t) of the object to be collected at the first moment. The above sensors can provide information about the linear and angular motion of the target. At the same time, the images around the target can be captured by an image sensor (such as a camera), and through steps such as preprocessing and feature extraction, image information I(t) related to the target motion can be generated.

[0072] In some embodiments, the acceleration and angular velocity information measured by the inertial navigation sensor can be fused with the image information provided by the image sensor to obtain an initial estimate of the navigation data. For example, some sensor fusion algorithms, such as Kalman filtering or a fusion neural network model, can be used to obtain more accurate and robust initial navigation data.

[0073] In some embodiments, the fused navigation data can be subjected to coordinate transformation to obtain the speed V(t), angle θ(t), and angular velocity ω(t) at the initial moment.

[0074] In summary, the acquisition of the initial navigation data is achieved by fusing the information of multiple sensors, including inertial navigation sensors and image sensors, so as to improve the accuracy and robustness of the target motion state. The above initial navigation data will be used as the starting point for the subsequent navigation data acquisition process.

[0075] Step 202: Obtain the real-time navigation data of the object to be collected at the second moment, the error of the real-time navigation data, and the influence value of the change in the motion state of the object to be collected caused by the real-time navigation data and the error of the real-time navigation data.

[0076] Where t is the first moment and τ is the second moment. In some embodiments, the inertial navigation data may include velocity update data V(t + Δt), angle update data θ(t + Δt), and angular velocity update data ω(t + Δt).

[0077] In some embodiments, the initial navigation data V(t), θ(t), ω(t) at the first moment can be used to perform update calculations of the navigation data to obtain the real-time navigation data at the second moment τ, such as velocity update V(τ), angle update θ(τ), and angular velocity update ω(τ).

[0078] In some embodiments, the real-time navigation data can be compared with the real motion state (if available) or previous measurement values to calculate the error term E, which represents the difference between the actual motion state and the estimated value of the navigation system. At the same time, considering the sensor measurement error or the inaccuracy of the system model, the error term ∈ can be calculated.

[0079] In some embodiments, according to the error terms E and ∈, as well as the corresponding weights and correction terms, their influence values on the change in the motion state of the object to be collected can be calculated. For example, the error terms can be introduced into the formula for updating the navigation data to correct the real-time navigation data.

[0080] In some embodiments, the calculated influence value can be applied to the real-time navigation data to obtain the finally corrected navigation data, such as velocity update V(τ), angle update θ(τ), and angular velocity update ω(τ).

[0081] Through the above method, by comparing the difference between the actual motion state and the estimated value of the navigation system, calculating the error and considering the correction term, the navigation data can be finally updated to improve the accurate estimation of the motion state of the object to be collected. Such a real-time navigation data correction mechanism helps to reduce the system error and improve the robustness of the navigation system.

[0082] Further, step 202 further includes:

[0083] For each type of the inertial navigation data, obtain the first data and the second data in the real-time navigation data;

[0084] Obtain the first error term corresponding to the first data and the second error term corresponding to the second data;

[0085] Fusion processing is performed on the first data, the second data, the first error term, and the second error term to obtain a fusion feature, and then integral processing is performed on the fusion feature to obtain the influence value.

[0086] In some embodiments, the influence value corresponding to the speed update data V(t + Δt) is The influence value corresponding to the angle update data θ(t + Δt) is The influence value corresponding to the angular velocity update data ω(t + Δt) is Taking as an example, the first data is A(τ), the second data is The first error term is E a (τ), and the second error term is ∈2(τ).

[0087] In some embodiments, for each type of inertial navigation data, such as the speed update data V(t + Δt), the angle update data θ(t + Δt), and the angular velocity update data ω(t + Δt), first obtain the first data and the second data in the real-time navigation data, that is, A(τ) and For the speed update data, ω(τ) for the angle update data, and For the angular velocity update data.

[0088] In some embodiments, for the first data, generate a corresponding first error term, such as E a (τ) for the speed update data, Eω(τ) for the angle update data, and Eα(τ) for the angular velocity update data. For the second data, generate a corresponding second error term, such as ∈2(τ) for the speed update data, ∈3(τ) for the angle update data, and ∈4(τ) for the angular velocity update data.

[0089] In some embodiments, fusion processing can be performed on the first data and the first error term, and the second data and the second error term to obtain a fusion feature. For example, through mathematical operations, weighted sums, etc., information from different sources can be fused according to specific situations to improve the accuracy and robustness of the navigation data.

[0090] Furthermore, the step "Fusion processing is performed on the first data, the second data, the first error term, and the second error term to obtain a fusion feature, and then integral processing is performed on the fusion feature to obtain the influence value" further includes:

[0091] Perform square root processing on the second data to obtain second square root data;

[0092] Fuse the second square root data and the second error term to obtain a first influence value that characterizes the change in the motion state of the object to be collected caused by the acquisition error of the second data;

[0093] After fusing the first data, the first error term, and the first influence value, perform an integration process to obtain the influence value.

[0094] In some embodiments, still taking as an example, A(τ) is the acceleration of the target at time τ, E a (τ) is the error term related to the acceleration, V(τ) is the square root of the velocity, sin(θ(τ)) is the sine value of the angle, ∈2(τ) is the error term related to the velocity update, and cos(θ(τ)) is the cosine value of the angle.

[0095] To facilitate understanding of the process of calculating the influence value, numerical examples are given below. Assume that: A(τ) = 2, indicating that the acceleration at time τ is a constant 2, E α (τ) = 0.5, indicating the error term related to the acceleration, V(τ) = 16, indicating that the square root of the velocity is a constant 4, θ(τ) = π / 4, indicating that the angle is 45 degrees, ∈2(τ) = 0.2, indicating the error term related to the velocity update, and the integration interval Then substituting the above data, we get:

[0096]

[0097] Step 203: Obtain the correction value output after processing the first image through the image correction model.

[0098] Among them, the first image is an image of the environment where the object to be collected is located at the second moment, including the velocity image I V (τ), the angle image I θ (τ), and the angular velocity image I ω (τ). The correction values corresponding to the three first images are respectively and

[0099] It can be understood that these correction values process the input image information through the neural network model, use the learning ability of the neural network to correct the image, and then generate correction values for the motion state, enabling the entire process to achieve intelligent processing of the image information and making the inertial navigation data more accurate and reliable.

[0100] Furthermore, step 203 further includes:

[0101] For each type of the inertial navigation data, obtain the image information corresponding to the inertial navigation data;

[0102] Integrate the image information to obtain the cumulative value corresponding to the image information;

[0103] Call the image correction model to correct the cumulative value to obtain the correction result corresponding to the image information; the correction result is the weight or offset term for correcting the image information;

[0104] Apply the correction result to the image information to obtain the corrected image information as the correction value.

[0105] In some embodiments, each type of image information can be integrated to obtain the corresponding cumulative value. For example, for the velocity image, the integration means the cumulative value of the velocity image in the time interval [t, t+Δt]. In some embodiments, the obtained cumulative value can be input into the image correction model, and the image correction model processes the cumulative value to generate the corresponding correction result, and the correction result can be the weight or offset term for correcting the image information.

[0106] In some embodiments, the correction result can be applied to the original image information to obtain the corrected image information as the correction value. In some embodiments, the correction value can represent the image information after being processed by the image correction model, and these correction information reflect the correction weight or offset of the inertial navigation data.

[0107] In summary, through integration, calling the correction model, and applying the correction result, the whole process realizes the accumulation and correction of the image information, provides more accurate image information for the subsequent inertial navigation data acquisition, and thus improves the accuracy and reliability of the acquisition.

[0108] Further, the step of "for each type of the inertial navigation data, obtain the image information corresponding to the inertial navigation data" further includes:

[0109] Call an image sensor to obtain an initial image that meets the preset acquisition conditions; the initial image includes video frames and static images;

[0110] Preprocess the initial image to obtain a preprocessed image;

[0111] Extract the motion features related to the object to be acquired and the inertial navigation data from the preprocessed image;

[0112] Perform a fusion process on the motion features to synthesize the image information corresponding to each type of the inertial navigation data.

[0113] In some embodiments, the preset acquisition condition may be a condition that affects the corresponding type of inertial navigation data. For example, the preset acquisition condition that affects the acquisition of the velocity image I V (τ) may be the appearance of an obstacle ahead, thus affecting the acquisition of the inertial navigation data of speed. In a specific implementation, assume that we want to acquire image information related to the vehicle speed, the velocity image I V (τ) may represent the environmental image around the vehicle at time τ.

[0114] Merely as an example, when the vehicle is driving normally, the velocity image I V (τ) may include images of the road ahead, road signs, other vehicles, and the surrounding environment. These images can be used to identify the current location of the vehicle, detect obstacles ahead, etc. When the vehicle makes an emergency brake, the velocity image I V (τ) may display an obstacle, pedestrian, or other traffic participants that suddenly appear in front of the vehicle. This image information is crucial for timely judgment and taking emergency braking operations. When the vehicle is changing lanes, the velocity image I V (τ) can capture the lane lines around the vehicle, adjacent vehicles, and the traffic conditions during lane change. These images help to judge the safety of lane change and select an appropriate timing for lane change. When the vehicle is driving at high speed, the velocity image I V (τ) may include the road in the distance ahead of the vehicle, other vehicles, and distant landmarks, etc. These images help to provide the overall operating state of the vehicle and are very important for grasping the situation of the vehicle driving at high speed.

[0115] In some embodiments, an image sensor mounted on the vehicle, such as a camera, may be used to obtain an initial image that meets the preset acquisition condition, which may be a real-time video frame or a series of static images, depending on the design and requirements of the acquisition system.

[0116] In some embodiments, the obtained initial image may be preprocessed to improve the effect of subsequent motion feature extraction. The preprocessing may include operations such as image denoising, contrast adjustment, and grayscale conversion to ensure image quality and feature clarity.

[0117] In some embodiments, motion features related to the object to be acquired (such as a vehicle) and inertial navigation data may be extracted from the preprocessed image. For example, it may include information such as the position, speed, direction of the vehicle, and changes in the surrounding environment. The methods for extracting motion features may involve techniques of computer vision and image processing, such as object detection, motion tracking, etc.

[0118] In some embodiments, the extracted motion features can be fused to synthesize the image information corresponding to each type of inertial navigation data. For example, it may involve adding the weights of different features or performing other mathematical operations to obtain the image information related to the inertial navigation data.

[0119] The goal of the above process is to use the image information to enhance the accuracy and real-time performance of the inertial navigation data. By fusing the image features with the inertial navigation data, a more comprehensive understanding of the target's motion state can be achieved, improving the performance and robustness of the navigation system.

[0120] Take as an example. The actual calculation process is as follows: Obtain a series of image frames of I V (τ), each image frame containing features related to velocity. Integrate the velocity image information, that is, integrate I V (τ) within the time range [t, t + Δt] to obtain Input the above integration result into the neural network model NN V . The image correction model will correct the velocity information according to the weights and biases obtained from its training and output the correction value.

[0121] Step 204: Based on the influence value and the correction value, obtain the updated navigation data for representing the updated motion state of the object to be collected.

[0122] In some embodiments, the updated navigation data can be expressed as

[0123] Step 205: According to the initial navigation data and the updated navigation data, determine and collect the inertial navigation data for representing the motion state of the object to be collected at the target moment.

[0124] Among them, the target moment is t + △t. In some embodiments, the inertial navigation data of the object to be collected at the target moment can be expressed as

[0125] It can be understood that after calculating the updated navigation data, the corresponding inertial navigation data can be calculated by calculating the sum of the updated navigation data and the initial navigation data.

[0126] In some embodiments, the inertial navigation data can be applied to the navigation system to update the motion state of the target. These data can be used in navigation algorithms to provide more accurate position, direction and other information. By integrating information from multiple aspects such as motion state, sensor error, and image correction, the present application improves the estimation accuracy of the target's motion state, making the collected inertial navigation data more accurate and reliable.

[0127] In some embodiments, the inertial navigation data may further include position update data P(t+Δt).

[0128] Further, the method of the present invention further includes:

[0129] Obtaining the horizontal motion state of the object to be collected at a second moment; the second moment is any moment within the range from the first moment to the target moment;

[0130] Based on the influence value, the correction value, and the horizontal motion state, obtaining first updated navigation data for representing the updated motion state of the object to be collected;

[0131] According to the initial navigation data and the first updated navigation data, determining and collecting the position update data of the object to be collected at the target moment.

[0132] In some embodiments, the position update data of the object to be collected at the target moment can be expressed as:

[0133]

[0134] where P(t+Δt) is the position update data, P(t) is the position of the object to be collected at the first moment, V(τ)·cos(θ(τ) is the horizontal motion state, is the influence value, is the correction value.

[0135] Further, P(t+Δt) represents the position update data of the object to be collected at the target moment t+Δt. V(τ)·cos(θ(τ) represents the horizontal motion state at time τ, where V(τ) is the velocity and θ(τ) is the angle, which is part of the current horizontal motion state. represents the influence on position update, including the influence of acceleration A(τ), correction term E_a(τ), velocity V(τ), and angle θ(τ). ∈1(τ) is a correction term related to image correction. is the correction term for correcting image information through a neural network, NN P represents the correction function of the neural network, which receives the integral of the image information I P (τ) in the time range [t,t+Δt] as input and generates a correction value.

[0136] For the convenience of understanding the process of calculating and collecting P(t+Δt), an example is given below. Assume: P(t) is 100 meters, V(τ)=20m / s, the influence value is 5, and the correction value is 2, △t = 0.1, substituting the above values into the formula:

[0137]

[0138] In some embodiments, P(t + △t) being 102.7 means that within the time interval [t, t + 0.1], due to the influence of the horizontal motion state, the position of the object to be collected rises from 100 to 102.7. For example, the position of the initial point changes from 100 meters to 102.7 meters. Therefore, the ultimate meaning of P(t + △t) is that at the target time t + Δt, the position of the object to be collected is updated to 102.7 after being affected by the influencing factors.

[0139] Please refer to Figure 3 , Figure 3 which is a schematic structural diagram of an inertial navigation data acquisition device provided by the present invention.

[0140] As Figure 3 shown, the inertial navigation data acquisition device proposed in the embodiment of the present invention includes:

[0141] A first acquisition module, configured to acquire the initial navigation data of the object to be collected at a first moment;

[0142] A second acquisition module, configured to acquire the real-time navigation data of the object to be collected at the second moment and the error of the real-time navigation data, as well as the influence value of the change in the motion state of the object to be collected caused by the real-time navigation data and the error of the real-time navigation data;

[0143] A third acquisition module, configured to acquire the correction value output after processing the first image through an image correction model; the first image is an image of the environment where the object to be collected is located at the second moment;

[0144] An updated navigation data module, configured to obtain updated navigation data representing the updated motion state of the object to be collected based on the influence value and the correction value;

[0145] A data determination module, configured to determine and acquire inertial navigation data representing the motion state of the object to be collected at the target moment according to the initial navigation data and the updated navigation data.

[0146] Please refer to Figure 4 , Figure 4 which is a schematic diagram of an embodiment of an electronic device provided by an embodiment of the present invention. As Figure 4 shown, an embodiment of the present invention provides an electronic device 400, including a memory 410, a processor 420, and a computer program 411 stored in the memory 410 and executable on the processor 420. When the processor 420 executes the computer program 411, the following steps are implemented:

[0147] Obtain the initial navigation data of the object to be collected at the first moment;

[0148] Obtain the real-time navigation data of the object to be collected at the second moment, the error of the real-time navigation data, and the influence value of the change of the real-time navigation data and the error of the real-time navigation data on the motion state of the object to be collected;

[0149] Obtain the correction value output after processing the first image through the image correction model; the first image is the image of the environment where the object to be collected is located at the second moment;

[0150] Based on the influence value and the correction value, obtain the updated navigation data for representing the updated motion state of the object to be collected;

[0151] According to the initial navigation data and the updated navigation data, determine and collect the inertial navigation data for representing the motion state of the object to be collected at the target moment.

[0152] Please refer to Figure 5 , Figure 5 which is a schematic diagram of an embodiment of a computer-readable storage medium provided by an embodiment of the present invention. As Figure 5 shown, this embodiment provides a computer-readable storage medium 500, on which a computer program 511 is stored. When the computer program 511 is executed by a processor, the following steps are implemented:

[0153] Obtain the initial navigation data of the object to be collected at the first moment;

[0154] Obtain the real-time navigation data of the object to be collected at the second moment, the error of the real-time navigation data, and the influence value of the change of the real-time navigation data and the error of the real-time navigation data on the motion state of the object to be collected;

[0155] Obtain the correction value output after processing the first image through the image correction model; the first image is the image of the environment where the object to be collected is located at the second moment;

[0156] Based on the influence value and the correction value, obtain the updated navigation data for representing the updated motion state of the object to be collected;

[0157] According to the initial navigation data and the updated navigation data, determine and collect the inertial navigation data for representing the motion state of the object to be collected at the target moment.

[0158] It should be noted that in the above embodiments, the descriptions of the respective embodiments have their own focuses. For parts not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0159] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0160] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0161] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that implement the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0162] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0163] Although the preferred embodiments of the present invention have been described, additional changes and modifications can be made to these embodiments by those skilled in the art once they learn of the basic inventive concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications that fall within the scope of the present invention.

[0164] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. A method for collecting inertial navigation data, characterized in that: The method comprises: Acquire initial navigation data of the object to be collected at the first moment; Acquire the real-time navigation data of the object to be collected at a second moment and an error of the real-time navigation data, and an influence value of the real-time navigation data and the error of the real-time navigation data on the change of the motion state of the object to be collected; Obtaining a correction value output after the first image is processed by the image correction model; the first image is an image of the environment where the object to be collected is located at the second moment; Based on the impact value and the correction value, obtaining updated navigation data for representing an updated motion state of the object to be captured; Determining and collecting inertial navigation data representing the motion state of the object to be collected at a target time according to the initial navigation data and the updated navigation data; The acquisition of the real-time navigation data and the influence value of the error of the real-time navigation data on the change of the motion state of the object to be collected include: For each type of the inertial navigation data, acquiring first data and second data in the real-time navigation data; Obtaining a first error term corresponding to the first data and a second error term corresponding to the second data; fusing the first data, the second data, the first error term and the second error term to obtain a fused feature, and then integrating the fused feature to obtain the influence value; The obtaining of the correction value output after the first image is processed by the image correction model comprises: For each type of inertial navigation data, acquiring image information corresponding to the inertial navigation data; Performing integration processing on the image information to obtain a cumulative value corresponding to the image information; Calling the image correction model to correct the accumulated value to obtain a correction result corresponding to the image information; the correction result is a weight or offset item for correcting the image information; The correction result is applied to the image information to obtain corrected image information as the correction value.

2. The method for collecting inertial navigation data according to claim 1, characterized in that: The inertial navigation data includes speed update data, angle update data and angular velocity update data.

3. The method for collecting inertial navigation data according to claim 2, characterized in that: The fusing the first data, the second data, the first error term and the second error term to obtain a fused feature, and then integrating the fused feature to obtain the influence value, comprises: Performing square root processing on the second data to obtain second square root data; The second square root data and the second error term are fused to obtain a first impact value for representing a change in the motion state of the object to be collected caused by the collection error of the second data; The first data, the first error term and the first influence value are fused and then integrated to obtain the influence value.

4. The method for collecting inertial navigation data according to claim 1, characterized in that: For each type of inertial navigation data, acquiring image information corresponding to the inertial navigation data includes: Calling an image sensor to obtain an initial image that meets a preset acquisition condition; the initial image includes a video frame and a static image; Preprocessing the initial image to obtain a preprocessed image; Extracting motion features related to the object to be collected and the inertial navigation data from the preprocessed image; The motion features are fused to synthesize the image information corresponding to each type of inertial navigation data.

5. The method for collecting inertial navigation data according to any one of claims 1 to 4, characterized in that: The inertial navigation data includes position update data, and the method further includes: Acquire the horizontal motion state of the object to be collected at a second moment; the second moment is any moment in the range from the first moment to the target moment; Based on the impact value, the correction value and the horizontal motion state, obtaining first updated navigation data for representing the updated motion state of the object to be captured; The position update data of the object to be collected at the target time is determined and collected according to the initial navigation data and the first updated navigation data.

6. The method for collecting inertial navigation data according to claim 5, characterized in that: The position update data of the object to be collected at the target time is expressed as: ； Among them, t is the first moment, For the second moment, is the position update data, P(t) is the position of the object to be collected at the first moment, is the horizontal motion state, θ ( τ ) is the angle update data, is the impact value, is the acceleration, is the correction term, is the second data, is a correction term related to image correction, is the correction value, represents the correction function of the neural network, For image information.

7. An inertial navigation data collection device, characterized in that: The device comprises: A first acquisition module, used to acquire initial navigation data of the object to be collected at a first moment; A second acquisition module is used to acquire the real-time navigation data of the object to be collected at a second moment and an error of the real-time navigation data, and an influence value of the real-time navigation data and the error of the real-time navigation data on the change of the motion state of the object to be collected; A third acquisition module is used to obtain a correction value output after the first image is processed by the image correction model; the first image is an image of the environment where the object to be collected is located at the second moment; An updating navigation data module, used for obtaining updated navigation data representing an updated motion state of the object to be collected based on the influence value and the correction value; A data determination module, used to determine and collect inertial navigation data representing the motion state of the object to be collected at a target time according to the initial navigation data and the updated navigation data; The acquisition of the real-time navigation data and the influence value of the error of the real-time navigation data on the change of the motion state of the object to be collected include: For each type of the inertial navigation data, acquiring first data and second data in the real-time navigation data; Obtaining a first error term corresponding to the first data and a second error term corresponding to the second data; fusing the first data, the second data, the first error term and the second error term to obtain a fused feature, and then integrating the fused feature to obtain the influence value; The obtaining of the correction value output after the first image is processed by the image correction model comprises: For each type of inertial navigation data, acquiring image information corresponding to the inertial navigation data; Performing integration processing on the image information to obtain a cumulative value corresponding to the image information; Calling the image correction model to correct the accumulated value to obtain a correction result corresponding to the image information; the correction result is a weight or offset item for correcting the image information; The correction result is applied to the image information to obtain corrected image information as the correction value.

8. An electronic device, characterized in that: include: Memory for storing computer software programs; A processor is used to read and execute the computer software program, thereby implementing the inertial navigation data collection method described in any one of claims 1-6.

9. A non-transitory computer-readable storage medium, characterized in that: The storage medium stores a computer software program, and when the computer software program is executed by the processor, the method for collecting inertial navigation data according to any one of claims 1 to 6 is implemented.

Images