AI-based adaptive method and device for inertial sensor fusion strategy

Through the AI-based inertial sensor fusion strategy adaptive method, the use scenarios are analyzed and adaptive strategies are selected or generated, and the applicability problem of MEMS inertial sensors in different scenarios is solved, the accuracy and scope of application are improved, and the sensor intelligence is realized and maximize economic benefits.

CN112880672BActive Publication Date: 2024-11-26WUHAN YESENSE TECH CO LTD
View PDF 10 Cites 0 Cited by

Patent Information

Application Number
CN202110049549.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-01-14
Publication Date
2024-11-26
Estimated Expiration
2041-01-14

AI Technical Summary

Technical Problem

Due to the need to write different software in different usage scenarios, existing MEMS inertial sensors have many software versions, the scope of application of products in a single scenario is limited, and there is a lack of intelligent fusion strategies to improve accuracy.

Method used

Adaptive method of AI-based inertial sensor fusion strategy is adopted to obtain sensor data, extract feature information using pre-trained AI models, analyze usage scenarios, and select or generate inertial sensor fusion strategies based on feature information, including Bias estimation and Bias update strategies , static judgment and fusion weights to achieve adaptive fusion strategy configuration.

Benefits of technology

It improves the accuracy and application scope of inertial sensors in different usage scenarios, realizes the intelligence of traditional MEMS inertial sensors, enhances the accuracy of the fusion output results, reduces the cost of use, and expands the application scenarios of the same sensor.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN112880672B_ABST
    Figure CN112880672B_ABST
Patent Text Reader

Abstract

The present invention provides an AI-based adaptive method and device for inertial sensor fusion strategy, the method comprising: first, obtaining sensor data to be fused at a given moment, and using a pre-trained AI model to perform feature extraction on the sensor data to be fused to obtain feature data information. Then, based on the feature data information, the matching usage scenario is analyzed; finally, according to the feature data information and its matching usage scenario, the inertial sensor fusion strategy is obtained. The present invention combines AI technology, and the user analyzes the user's current usage scenario under different usage scenario conditions through a large amount of data collected in real time by the sensor to be fused, and selects an inertial sensor fusion strategy for the usage scenario. When necessary, the configuration parameters of the inertial sensor fusion strategy are modified to achieve the best usage effect, thereby making the traditional MEMS inertial sensor fusion strategy more intelligent and the fusion output result more accurate.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of inertial sensors, and in particular to an AI-based inertial sensor fusion strategy adaptive method and device. Background Art

[0002] MEMS inertial sensors have a series of advantages that traditional inertial sensors do not have, such as small size, light weight, low power consumption, mass production, low production cost, and high reliability. Therefore, they are increasingly widely used in various industries, such as aviation, aerospace, navigation, automotive industry, industrial monitoring, robotics and consumer electronics.

[0003] Low-precision MEMS inertial sensors are mainly used in consumer electronics products such as mobile phones, game consoles, smart watches and bracelets, etc.

[0004] Intermediate MEMS inertial sensors are mainly used in industrial and automotive products, such as automotive electronic stability devices, automotive airbags, various robots (AGV, commercial services, sweeping robots), etc.

[0005] High-precision MEMS inertial sensors are mainly used as military-grade and aerospace-grade products, such as missile guidance heads, aircraft / missile flight control, battlefield robots, etc.

[0006] MEMS inertial sensors have a variety of usage scenarios that have been discovered or are yet to be discovered. Different usage scenarios will inevitably require different processing methods for the sensors. Therefore, different software needs to be written for different usage scenarios, resulting in a large number of software versions and limited applicability of products for a single scenario. Summary of the invention

[0007] In order to solve the above problems, an embodiment of the present invention provides an AI-based inertial sensor fusion strategy adaptive method and device that overcomes the above problems or at least partially solves the above problems.

[0008] In a first aspect, an embodiment of the present invention provides an AI-based inertial sensor fusion strategy adaptive method, including:

[0009] S1, obtain the sensor data to be fused at a given moment;

[0010] S2, using a pre-trained AI model to extract features of the sensor data to be fused to obtain feature data information;

[0011] S3, analyzing the matching usage scenarios based on the feature data information;

[0012] S4, obtaining an inertial sensor fusion strategy according to the feature data information and its matching usage scenario.

[0013] Preferably, the sensor to be fused in step S1 includes at least a MEMS accelerometer and a MEMS gyroscope.

[0014] Preferably, the sensor to be fused further includes one or more of a magnetometer, a barometer, an odometer and a GNSS.

[0015] Preferably, the sensor data to be fused at a given moment obtained in step S1 includes Pi=F(Ωi, Ai, Mi, ..., Ti, BWi); wherein i represents moment i, Ωi represents the angular velocity value at moment i, Ai represents the acceleration value at moment i, Mi represents the magnetometer data at moment i, Ti represents the temperature value at moment i, and BWi represents the sensor bandwidth at moment i.

[0016] Preferably, the characteristic data information in step S2 includes but is not limited to RMS noise value, Norm value, temperature value and bandwidth.

[0017] Preferably, in steps S3 to S4, if a matched usage scenario is obtained according to the characteristic data information analysis in step S3, a corresponding inertial sensor fusion strategy is selected according to the matched usage scenario in step S4.

[0018] If new usage scenario data is obtained according to the analysis of the feature data information in step S3, the feature data information and its corresponding new usage scenario data are saved, and then the inertial sensor fusion strategy under the new usage scenario is obtained using the neural network model.

[0019] Preferably, the inertial sensor fusion strategy includes but is not limited to Bias estimation, Bias update strategy, static judgment and fusion weight.

[0020] In a second aspect, an embodiment of the present invention further provides an AI-based inertial sensor fusion strategy adaptive device, comprising:

[0021] An acquisition module is used to obtain the sensor data to be fused at a given moment;

[0022] A feature extraction module, used to extract features of the sensor data to be fused using a pre-trained AI model to obtain feature data information;

[0023] A usage scenario matching module is used to analyze the matching usage scenarios based on the feature data information;

[0024] The fusion strategy acquisition module is used to obtain the inertial sensor fusion strategy according to the characteristic data information and the matching usage scenario.

[0025] In a third aspect, an embodiment of the present invention provides an electronic device, comprising a processor, a memory, a communication interface and a bus; wherein the processor, the memory, and the communication interface communicate with each other through the bus; the memory stores program instructions executable by the processor, and the processor calls the program instructions to execute the AI-based inertial sensor fusion strategy adaptation method provided in the embodiment of the first aspect.

[0026] In a fourth aspect, an embodiment of the present invention provides a non-transitory computer-readable storage medium, which stores a computer program, which, when executed by a processor, is implemented to execute the AI-based inertial sensor fusion strategy adaptation method provided in the embodiment of the first aspect.

[0027] The AI-based inertial sensor fusion strategy adaptive method and device provided in the embodiments of the present invention, combined with AI technology, enables the user to analyze the user's current usage scenario under different usage scenarios through a large amount of data collected in real time by the sensor to be fused, and select an inertial sensor fusion strategy for the usage scenario. If necessary, the configuration parameters of the inertial sensor fusion strategy are modified to achieve the best usage effect, thereby making the traditional MEMS inertial sensor fusion strategy more intelligent and the fusion output result more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0029] Figure 1 A schematic diagram of a flow chart of an AI-based inertial sensor fusion strategy adaptive method provided in an embodiment of the present invention;

[0030] Figure 2 The simulation results of the gyroscope and its integral angle under different bandwidths provided by the embodiment of the present invention;

[0031] Figure 3 Accelerometer and angle simulation results under different bandwidths provided by the embodiments of the present invention;

[0032] Figure 4 Simulation results of different Bias update strategies provided by the embodiments of the present invention;

[0033] Figure 5 A schematic diagram of the structure of an AI-based inertial sensor fusion strategy adaptive device provided in an embodiment of the present invention;

[0034] Figure 6 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0035] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0036] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0037] At present, MEMS inertial sensors have a variety of usage scenarios that have been discovered or are yet to be discovered. Different usage scenarios will inevitably require different processing methods for the sensors. Therefore, different software needs to be written for different usage scenarios, resulting in a large number of software versions and limited applicability of products for a single scenario.

[0038] AI (artificial intelligence) is a process of continuous learning that enables intelligent behaviors such as reasoning, thinking, and planning.

[0039] In view of the above problems in the prior art, the embodiments of the present invention use AI technology to provide an AI-based inertial sensor fusion strategy adaptive method to adapt to more usage scenarios and strive to achieve optimal performance for different usage scenarios. The following will be described and introduced through multiple embodiments in conjunction with the accompanying drawings.

[0040] Figure 1 A schematic diagram of a flow chart of an AI-based inertial sensor fusion strategy adaptive method provided in an embodiment of the present invention, such as Figure 1 As shown, the AI-based inertial sensor fusion strategy adaptive method provided by the embodiment of the present invention includes but is not limited to the following steps:

[0041] S1, obtain the sensor data to be fused at a given moment.

[0042] The MEMS inertial sensor device fusion strategy adaptive method provided in the embodiment of the present invention can be applied to a variety of MEMS inertial sensors, thereby increasing the trial scope of the method. The sensor to be fused includes at least a MEMS accelerometer and a MEMS gyroscope. Furthermore, the sensor to be fused may also include one or more of a magnetometer, a barometer, an odometer, and a GNSS (Global Navigation Satellite System).

[0043] In this embodiment, the sensor data to be fused at a given moment includes Pi=F(Ωi, Ai, Mi, ..., Ti, BWi), where i represents moment i, Ωi represents the angular velocity value at moment i, Ai represents the acceleration value at moment i, Mi represents the magnetometer data at moment i, Ti represents the temperature value at moment i, and BWi represents the sensor bandwidth at moment i.

[0044] S2, using a pre-trained AI (Artificial Intelligence) model to extract features of the sensor data to be fused to obtain feature data information.

[0045] The characteristic data information includes but is not limited to RMS noise value, Norm value, temperature value and bandwidth.

[0046] S3, analyzing the matching usage scenarios based on the feature data information.

[0047] S4, obtaining an inertial sensor fusion strategy according to the feature data information and its matching usage scenario.

[0048] Specifically, if a matched usage scenario is obtained according to the characteristic data information analysis in step S3, then in step S4, a corresponding inertial sensor fusion strategy can be selected based on experience according to the matched usage scenario.

[0049] If new usage scenario data is obtained according to the characteristic data information analysis in step S3, the characteristic data information and its corresponding new usage scenario data are saved, and then the inertial sensor fusion strategy and its configuration parameters under the new usage scenario are obtained using the neural network model.

[0050] Figure 2 The simulation results of the gyroscope and its integral angle under different bandwidths provided by the embodiment of the present invention; Figure 3 The accelerometer and angle simulation results under different bandwidths provided by the embodiment of the present invention are shown in FIG. Figure 2 and Figure 3 As shown, Figure 2 and Figure 3 The relationship between the inertial sensor fusion output angle and bandwidth is shown.

[0051] In this embodiment, the inertial sensor fusion strategy adopted in step S4 includes but is not limited to bias estimation, bias update strategy, static judgment and fusion weight. Among them, bias estimation is zero bias estimation, including zero bias estimation of acceleration and gyroscope. Bias update strategy is a calculation method for updating zero bias in real time. Figure 5 The simulation results of different Bias update strategies provided by the embodiments of the present invention. Static judgment refers to judging whether the device is in a static state or a moving state through IMU data. The fusion weight can be understood as the confidence of each sensor data in the fusion algorithm.

[0052] The AI-based inertial sensor fusion strategy adaptive method provided by the embodiment of the present invention combines AI technology. Under different usage scenarios, the user can analyze the user's current usage scenario through a large amount of data collected in real time by the sensor to be fused, and select the inertial sensor fusion strategy for the usage scenario. If necessary, the configuration parameters of the inertial sensor fusion strategy are modified to achieve the best usage effect, thereby making the traditional MEMS inertial sensor fusion strategy more intelligent and the fusion output result more accurate. In addition, this method can increase the usage scenarios applicable to the same sensor at the lowest usage cost, thereby achieving the purpose of maximizing economic benefits.

[0053] In one embodiment, Figure 5 The schematic diagram of the structure of the AI-based inertial sensor fusion strategy adaptive device provided in the embodiment of the present invention is used to execute the AI-based inertial sensor fusion strategy adaptive method in the above method embodiment. Figure 5 As shown, the device comprises:

[0054] An acquisition module 501 is used to acquire sensor data to be fused at a given moment;

[0055] A feature extraction module 502 is used to extract features of the sensor data to be fused using a pre-trained AI model to obtain feature data information;

[0056] A usage scenario matching module 503, used to analyze the matching usage scenarios based on the feature data information;

[0057] The fusion strategy acquisition module 504 is used to obtain an inertial sensor fusion strategy according to the feature data information and its matching usage scenario.

[0058] For details on how to use the acquisition module 501, the feature extraction module 502, the usage scene matching module 503 and the fusion strategy acquisition module 504 to execute the AI-based inertial sensor fusion strategy adaptation method provided in the above embodiment, please refer to the aforementioned method embodiment, and the embodiments of the present invention will not be repeated here.

[0059] In one embodiment, the present invention provides an electronic device, such as Figure 6 As shown, the electronic device may include: a processor (processor) 601, a communication interface (CommunicationsInterface) 602, a memory (memory) 603 and a communication bus 604, wherein the processor 601, the communication interface 602, and the memory 603 communicate with each other through the communication bus 604. The processor 601 can call the logic instructions in the memory 603 to execute the steps of the AI-based inertial sensor fusion strategy adaptive method provided in the above embodiments, for example, including: S1, obtaining the sensor data to be fused at a given moment; S2, using a pre-trained AI model to extract features of the sensor data to be fused to obtain feature data information; S3, analyzing the matching usage scenarios based on the feature data information; S4, obtaining the inertial sensor fusion strategy according to the feature data information and its matching usage scenarios.

[0060] In one embodiment, an embodiment of the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the AI-based inertial sensor fusion strategy adaptation method provided in the above embodiments are implemented, for example, including: S1, obtaining sensor data to be fused at a given moment; S2, using a pre-trained AI model to perform feature extraction on the sensor data to be fused to obtain feature data information; S3, analyzing the matching usage scenarios based on the feature data information; S4, obtaining the inertial sensor fusion strategy based on the feature data information and its matching usage scenarios.

[0061] In summary, the embodiments of the present invention provide an AI-based inertial sensor fusion strategy adaptive method and device. Combined with AI technology, the user can analyze the user's current usage scenario under different usage scenarios through a large amount of data collected in real time by the sensor to be fused, and select the inertial sensor fusion strategy for the usage scenario. If necessary, the configuration parameters of the inertial sensor fusion strategy are modified to achieve the best usage effect, thereby making the traditional MEMS inertial sensor fusion strategy more intelligent and the fusion output result more accurate.

[0062] The various embodiments of the present invention can be combined arbitrarily to achieve different technical effects.

[0063] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0064] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An AI-based inertial sensor fusion strategy adaptive method, characterized in that: include: S1, obtain the sensor data to be fused at a given moment; S2, using a pre-trained AI model to extract features of the sensor data to be fused to obtain feature data information; S3, analyzing the matching usage scenarios based on the feature data information; S4, obtaining an inertial sensor fusion strategy according to the characteristic data information and its matching usage scenario; In steps S3-S4, if a matched usage scenario is obtained according to the characteristic data information analysis in step S3, then in step S4, a corresponding inertial sensor fusion strategy is selected according to the matched usage scenario; If new usage scenario data is obtained according to the characteristic data information analysis in step S3, the characteristic data information and the corresponding new usage scenario data are saved, and then the inertial sensor fusion strategy under the new usage scenario is obtained by using the neural network model; The inertial sensor fusion strategy includes bias estimation, bias update strategy, static judgment and fusion weight.

2. The AI-based inertial sensor fusion strategy adaptive method according to claim 1, characterized in that: The sensors to be fused in step S1 at least include a MEMS accelerometer and a MEMS gyroscope.

3. The AI-based inertial sensor fusion strategy adaptive method according to claim 2, characterized in that: The sensors to be fused also include one or more of a magnetometer, a barometer, an odometer and a GNSS.

4. The AI-based inertial sensor fusion strategy adaptive method according to claim 3, characterized in that: The sensor data to be fused at a given time obtained in step S1 includes Pi = F(Ωi, Ai, Mi, ..., Ti, BWi); wherein i represents time i, Ωi represents the angular velocity value at time i, Ai represents the acceleration value at time i, Mi represents the magnetometer data at time i, Ti represents the temperature value at time i, and BWi represents the sensor bandwidth at time i.

5. The AI-based inertial sensor fusion strategy adaptive method according to claim 4, characterized in that: The characteristic data information in step S2 includes RMS noise value, Norm value, temperature value and bandwidth.

6. An AI-based inertial sensor fusion strategy adaptive device, characterized in that: include: An acquisition module is used to obtain the sensor data to be fused at a given moment; A feature extraction module, used to extract features of the sensor data to be fused using a pre-trained AI model to obtain feature data information; A usage scenario matching module is used to analyze the matching usage scenarios based on the feature data information; A fusion strategy acquisition module, used to obtain an inertial sensor fusion strategy based on the feature data information and its matching usage scenario; In the usage scenario matching module and the fusion strategy acquisition module, if the usage scenario matching module obtains a matched usage scenario according to the characteristic data information analysis, then in the fusion strategy acquisition module, a corresponding inertial sensor fusion strategy is selected according to the matched usage scenario; If the usage scenario matching module obtains new usage scenario data according to the feature data information analysis, the feature data information and its corresponding new usage scenario data are saved, and then the inertial sensor fusion strategy under the new usage scenario is obtained by using the neural network model; The inertial sensor fusion strategy includes bias estimation, bias update strategy, static judgment and fusion weight.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the AI-based inertial sensor fusion strategy adaptation method as described in any one of claims 1 to 5 are implemented.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the AI-based inertial sensor fusion strategy adaptation method as described in any one of claims 1 to 5 are implemented.

Citation Information

Patent Citations

  • Method and device for identifying scene integrated by multi-feature vision codebook

    CN103366181A

  • Integrated navigation data fusion method suitable for weak signals of GNSS

    CN105180935A

  • Multi-sensor-based vehicle-mounted environment recognition system and omni-directional vision module

    CN106878687A

  • Image information and radar information fusion method and system for traffic scene

    CN108983219A

  • Semi-supervised scene classification identification method, system and device based on multimode fusion

    CN109697453A