Motion estimation method under differential control, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a motion estimation method under differential control, electronic equipment and a storage medium, and belongs to the technical field of carrier control. The method comprises the following steps: acquiring real-time wheel speed data of each driving wheel, inertial measurement unit data of a variable wheelbase chassis, a current chassis wheelbase, environment sensing data and longitude and latitude mileage data based on preset metering time; based on the real-time wheel speed data and the current chassis wheelbase, motion mileage analysis data of the carrier within the metering time are output; obtaining motion integral data within metering time based on the data of the inertial measurement unit; performing motion mileage analysis based on the environment perception data to obtain image mileage data within the metering time; and performing motion data fusion based on the motion mileage analysis data, the image mileage data, the motion integral data and the longitude and latitude mileage data to obtain target motion mileage data. According to the embodiment of the invention, the motion mileage estimation algorithm which can adapt to the change of the chassis structure and can compensate for complex slippage can be provided.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle control, in particular to a motion estimation method under differential control, an electronic device and a storage medium. BACKGROUND

[0002] The four-wheel differential chassis driving mode is a chassis driving and steering mode for mobile robots or special vehicles, and the principle is to realize slip steering by controlling the speed difference of the left and right wheels or four wheels.

[0003] However, in the related art, the vehicle under the four-wheel differential chassis driving mode often calibrates the slip coefficient of the chassis in advance, so as to analyze the motion, calculate how far the vehicle has traveled and how much the vehicle has turned, and thus obtain a continuous pose estimation. However, this method cannot be applied to vehicles with changing chassis structures. When the variable wheelbase chassis changes the wheelbase, the kinematic model of the wheels will change significantly. At this time, if the traditional motion analysis model is used, errors will be introduced. Such errors are particularly serious in unknown or complex ground environments, which can lead to inaccurate analysis of the motion of the robot or vehicle.

[0004] Therefore, there is an urgent need for a motion estimation method that can adapt to changes in chassis structure to solve the problem of inaccurate motion analysis caused by changes in wheelbase and road conditions of the variable wheelbase chassis. SUMMARY

[0005] The main purpose of the embodiments of the present application is to provide a motion estimation method under differential control, an electronic device and a storage medium, which can adapt to changes in chassis structure and compensate for complex slip.

[0006] To achieve the above-mentioned purpose, a first aspect of the embodiments of the present application provides a motion estimation method under differential control, applied to a vehicle configured with a variable wheelbase chassis, the variable wheelbase chassis being configured with a plurality of drive wheels, and the method comprising: Based on a preset metering time, real-time wheel speed data of each drive wheel is obtained, inertial measurement unit data and a current chassis wheelbase of the variable wheelbase chassis are obtained, environmental perception data and latitude and longitude mileage data are obtained; Based on the real-time wheel speed data and the current chassis wheelbase, motion analysis of the vehicle is performed, and motion mileage analysis data of the vehicle within the metering time is output; Based on the inertial measurement unit data, motion integral analysis is performed, and motion integral data within the metering time is obtained; Based on the environmental perception data, motion mileage analysis is performed, and image mileage data within the metering time is obtained; Perform motion data fusion based on the motion mileage analysis data, the image mileage data, the motion integral data, and the latitude and longitude mileage data to obtain target motion mileage data.

[0007] In some embodiments, the vehicle motion analysis based on the real-time wheel speed data and the current chassis wheelbase outputs motion mileage analysis data of the vehicle within the metering time, including: Perform environmental terrain analysis based on the collected environmental perception data to obtain a road surface friction coefficient; Perform slip coefficient analysis based on the current chassis wheelbase, the road surface friction coefficient, and the real-time wheel speed data input into a pre-trained slip rate estimation model to output slip coefficient estimates of each of the drive wheels; Perform vehicle motion analysis based on the slip coefficient estimates of each of the drive wheels and the real-time wheel speed data to output motion mileage analysis data of the vehicle within the metering time.

[0008] In some embodiments, the slip coefficient estimates include forward slip coefficient estimates and lateral slip coefficient estimates, The vehicle motion analysis based on the real-time wheel speed data and the current chassis wheelbase outputs motion mileage analysis data of the vehicle within the metering time, including: Determine the travel line speed of the vehicle based on the real-time wheel speed data of each of the drive wheels and the forward slip coefficient estimates; Determine the travel angular speed of the vehicle based on the real-time wheel speed data of each of the drive wheels and the lateral slip coefficient estimates; Determine the motion mileage analysis data of the vehicle within the metering time based on the travel line speed and the travel angular speed.

[0009] In some embodiments, the motion mileage analysis based on the environmental perception data outputs image mileage data within the metering time, including: Obtain a first environmental image and a second environmental image of the environmental perception data based on a preset collection time interval; Perform image feature extraction based on the first environmental image to obtain first image features; Perform image feature extraction based on the second environmental image to obtain second image features; Perform feature matching based on the first image features and the second image features to determine relative displacement data; Update the first environmental image and the second environmental image based on the metering time and the collection time interval to obtain multiple relative displacement data and determine the image mileage data.

[0010] In some embodiments, the motion data fusion based on the motion mileage analysis data, the image mileage data, the motion integral data and the latitude and longitude mileage data obtains target motion mileage data, comprising: The motion mileage analysis data, the image mileage data, the motion integral data and the latitude and longitude mileage data are subjected to confidence evaluation to obtain confidence evaluation data; Based on the confidence evaluation data, the target motion mileage data is determined from the motion mileage analysis data, the image mileage data, the motion integral data and the latitude and longitude mileage data.

[0011] In some embodiments, the motion mileage analysis data, the image mileage data, the motion integral data and the latitude and longitude mileage data are input into a confidence evaluation model for confidence evaluation to obtain confidence evaluation data, comprising: Obtain the signal strength information associated with the latitude and longitude mileage data within the measurement time; Obtain the matching confidence of a plurality of relative displacement data in the image mileage data; Based on the real-time wheel speed data of each driving wheel, the current chassis wheelbase and the inertial measurement unit data within the measurement time, the motion state information of the vehicle is determined; Based on the motion state information, vehicle slip analysis is performed to obtain the number of vehicle slips; Based on the number of vehicle slips, the first confidence of the motion mileage analysis data and the second confidence of the motion integral data are determined; Based on the signal strength information, the matching confidence, the first confidence and the second confidence, the confidence evaluation data is obtained.

[0012] In some embodiments, the target motion mileage data is determined from the motion mileage analysis data, the image mileage data, the motion integral data and the latitude and longitude mileage data based on the confidence evaluation data, comprising: Based on the first confidence and the second confidence, the first weight value of the motion mileage analysis data and the second weight value of the motion integral data are determined; Based on the first weight value, the motion mileage analysis data, the second weight value and the motion integral data, the target motion mileage data is determined.

[0013] In some embodiments, the method further comprises: In response to the signal strength information exceeding a preset first intensity value, the latitude and longitude mileage data is determined as a target reinforcement value; determine the image mileage data as a target reinforcement value in response to the confidence exceeding a preset second confidence value; determine a first confidence of the motion mileage analysis data and a second confidence of the motion integral data based on the target reinforcement value.

[0014] To achieve the above object, a second aspect of the embodiment of the present application provides a motion estimation device under differential control, which is used to implement the motion estimation method under differential control as described in the first aspect.

[0015] To achieve the above object, a third aspect of the embodiment of the present application provides an electronic device, which comprises a memory and a processor, the memory stores a computer program, and the processor implements the motion estimation method under differential control as described in the first aspect when executing the computer program.

[0016] To achieve the above object, a fourth aspect of the embodiment of the present application provides a computer readable storage medium, which stores a computer program, and the computer program implements the motion estimation method under differential control as described in the first aspect when executed by a processor.

[0017] The motion estimation method under differential control, the electronic device and the storage medium provided by the present application obtain real-time wheel speed data of each drive wheel based on a preset measurement time, obtain inertia measurement unit data and a current chassis wheelbase of a variable wheelbase chassis, obtain environmental perception data and latitude and longitude mileage data, perform vehicle motion analysis based on the real-time wheel speed data and the current chassis wheelbase, output motion mileage analysis data of the vehicle within the measurement time, perform motion integral analysis based on the inertia measurement unit data to obtain motion integral data within the measurement time, perform motion mileage analysis based on the environmental perception data to obtain image mileage data within the measurement time, and perform motion data fusion based on the motion mileage analysis data, the image mileage data, the motion integral data and the latitude and longitude mileage data to obtain target motion mileage data. Therefore, by obtaining the current chassis wheelbase in real time and bringing it into the motion analysis process to obtain the motion mileage analysis data, and by performing motion data fusion on the motion mileage analysis data, the image mileage data, the motion integral data and the latitude and longitude mileage data which are different in source and characteristics, the motion analysis deviation caused by the change of the chassis structure and the slip error under complex road conditions are corrected, so that high-precision motion mileage estimation results are provided under the conditions of the change of the chassis structure and complex slip. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 is a flowchart of the motion estimation method under differential control provided by the embodiment of the present application; Figure 2 is Figure 1A flowchart illustrating step 102 in the diagram; Figure 3 yes Figure 2 A flowchart illustrating step 203 in the process; Figure 4 yes Figure 1 A flowchart illustrating step 104 in the diagram; Figure 5 yes Figure 1 A flowchart illustrating step 105 in the diagram; Figure 6 yes Figure 5 A flowchart illustrating step 501 in the diagram; Figure 7 yes Figure 5 A flowchart illustrating step 502 in the diagram; Figure 8 This is another flowchart illustrating the motion estimation method under differential control provided in the embodiments of this application. Figure 9 This is a schematic diagram of the variable wheelbase chassis structure provided in the embodiments of this application; Figure 10 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0020] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0022] This application provides a motion estimation method, electronic device, and storage medium under differential control, aiming to provide a motion mileage estimation algorithm that can adapt to changes in chassis structure and compensate for complex slippage.

[0023] The motion estimation method under differential control provided by the embodiment of the present application is described in the following embodiment. First, the motion estimation method under differential control in the embodiment of the present application is described.

[0024] The embodiment of the present application can acquire and process related data based on artificial intelligence technology. The artificial intelligence (AI) is a theory, method, technology and application system for simulating, extending and expanding human intelligence by using a digital computer or a machine controlled by a digital computer, perceiving environment, acquiring knowledge and using knowledge to obtain optimal results.

[0025] The motion estimation method under differential control provided by the embodiment of the present application relates to the technical field of vehicle control. The motion estimation method under differential control provided by the embodiment of the present application can be applied to a terminal, can be applied to a server end, and can also be software running in a terminal or a server end. In some embodiments, the terminal can be a smart phone, a tablet computer, a notebook computer, a desktop computer, etc.; the server end can be configured as a stand-alone physical server, can be configured as a server cluster or a distributed system composed of multiple physical servers, can also be configured as a cloud server providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and basic cloud computing services such as big data and artificial intelligence platforms; and the software can be an application that implements the motion estimation method under differential control, but is not limited to the above forms.

[0026] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld devices or portable devices, tablet devices, multi-processor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in a distributed computing environment in which tasks are performed by remote processing devices connected by a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media, including storage devices.

[0027] Figure 1 is an optional flowchart of the motion estimation method under differential control provided by the embodiment of the present application, applied to a vehicle configured with a variable wheelbase chassis, the variable wheelbase chassis being configured with a plurality of drive wheels, Figure 1The method in the method can include but is not limited to steps 101-105.

[0028] Step 101, based on the preset measurement time, acquiring real-time wheel speed data of each drive wheel, acquiring inertia measurement unit data and current chassis wheelbase of the variable wheelbase chassis, acquiring environmental perception data and latitude and longitude mileage data; Step 102, based on real-time wheel speed data and current chassis wheelbase, performing vehicle motion analysis, and outputting motion mileage analysis data of the vehicle within the measurement time; Step 103, based on the inertia measurement unit data, performing motion integral analysis to obtain motion integral data within the measurement time; Step 104, based on the environmental perception data, performing motion mileage analysis to obtain image mileage data within the measurement time; Step 105, based on motion mileage analysis data, image mileage data, motion integral data and latitude and longitude mileage data, performing motion data fusion to obtain target motion mileage data.

[0029] The steps 101-105 shown in the embodiments of the present application, by acquiring real-time wheel speed data of each drive wheel based on the preset measurement time, acquiring inertia measurement unit data and current chassis wheelbase of the variable wheelbase chassis, acquiring environmental perception data and latitude and longitude mileage data, based on real-time wheel speed data and current chassis wheelbase, performing vehicle motion analysis, and outputting motion mileage analysis data of the vehicle within the measurement time, based on the inertia measurement unit data, performing motion integral analysis to obtain motion integral data within the measurement time, based on the environmental perception data, performing motion mileage analysis to obtain image mileage data within the measurement time, based on motion mileage analysis data, image mileage data, motion integral data and latitude and longitude mileage data, performing motion data fusion to obtain target motion mileage data. Therefore, by acquiring real-time current chassis wheelbase and bringing it into the motion analysis process, the motion mileage analysis data is obtained, and the motion mileage analysis data, image mileage data, motion integral data and latitude and longitude mileage data with different sources and different characteristics are fused to correct the motion analysis deviation caused by the change of the chassis structure and the slip error under complex road conditions, thereby providing high-precision motion mileage estimation results under the conditions of chassis structure change and complex slip.

[0030] In step 101 of some embodiments, a metering time is first set, and real-time collection of multi-source data is started within this period. Specifically, the data collection module reads real-time wheel speed data from the encoders installed on each drive wheel, which reflects the rotation speed of each wheel. At the same time, considering the variable structure of the chassis, the current chassis track of the variable track chassis is obtained in real time through the joint encoder or displacement sensor. In addition, three-axis acceleration and angular velocity data output by the inertial measurement unit (IMU) are also read synchronously, as well as environmental perception data collected by the laser radar or visual camera, and latitude and longitude mileage data provided by the global positioning system (GPS).

[0031] In step 102 of some embodiments, in-depth vehicle motion analysis is performed on the collected real-time wheel speed data and current chassis track. Since the chassis track is not a fixed value, the traditional kinematic model cannot be directly applied. Therefore, the real-time changing chassis track is substituted into the slip ratio estimation model as a key variable, and the real-time wheel speed data of each drive wheel is combined for calculation. In this way, the linear and angular velocities of the vehicle within the metering time are calculated, and the pose change of the vehicle relative to the previous metering time is derived, and finally the motion mileage analysis data reflecting the change of the chassis structure is output, effectively solving the problem of inaccurate motion mileage estimation caused by chassis structure deformation.

[0032] Referring to Figure 2 In some embodiments, step 102 can include but is not limited to steps 201 to 203: Step 201, based on the collected environmental perception data, the road surface friction coefficient is obtained by analyzing the environmental terrain; Step 202, based on the current chassis track, road surface friction coefficient and real-time wheel speed data, the slip coefficient estimation value of each drive wheel is output by inputting into the pre-trained slip ratio estimation model for slip coefficient analysis; Step 203, based on the slip coefficient estimation value of each drive wheel and the real-time wheel speed data, the motion mileage analysis data of the vehicle within the metering time is output by performing vehicle motion analysis.

[0033] In step 201 of some embodiments, environmental terrain analysis is performed on the current terrain of the vehicle using the collected environmental perception data, for example, by analyzing image data acquired through a forward-facing camera. Computer vision technology or deep learning feature extraction technology is used to identify the texture, material, and smoothness of the road surface. Through quantitative analysis of the road surface features, parameters characterizing the current ground adhesion ability can be estimated, i.e., the road surface friction coefficient is obtained. The road surface friction coefficient reflects the grip or anti-skid ability of the wheels on the road surface, characterizing the probability of the wheels slipping on the current terrain.

[0034] In step 202 of some embodiments, the acquired current chassis wheelbase, the calculated road friction coefficient, and real-time wheel speed data are combined into a multi-dimensional feature vector, which is then input into a pre-trained slip ratio estimation model. The slip ratio estimation model has typically been trained under various typical terrains and different wheelbase settings, and can be constructed based on a Long Short-Term Memory (LSTM) network or other neural networks with temporal processing capabilities. After receiving the multi-dimensional feature vector, the slip ratio estimation model performs online inference, comprehensively considering the influence of chassis structural deformation on motion and the dynamic characteristics under different friction coefficients, analyzing and outputting estimated slip coefficient values ​​for each drive wheel under the current state.

[0035] A four-wheel differential chassis relies on wheel slippage to achieve steering. Therefore, there is a discrepancy between the theoretical wheel speed and the actual vehicle speed. The slip coefficient estimate is a parameter used to quantify this degree of slippage; it describes the degree of slippage that occurs when the wheels roll on the ground. The slip coefficient estimate is used to correct for motion errors caused by slippage.

[0036] In step 203 of some embodiments, the original real-time wheel speed data is corrected and solved using the calculated slip coefficient estimates of each drive wheel. Specifically, a generalized dynamic kinematic model incorporating slip parameters is constructed. The slip coefficient estimates are substituted into this generalized dynamic kinematic model to convert the simple wheel speed pulses into the actual linear and angular velocities of the vehicle body after slip compensation. Subsequently, integration is performed based on the actual linear and angular velocities over a measurement time period to output high-precision vehicle mileage analysis data within the measurement time period.

[0037] Through steps 201 to 203, changes in road surface friction can be sensed in real time, the road surface friction coefficient can be obtained, and vehicle motion analysis can be performed in conjunction with the current chassis wheelbase to predict the nonlinear slip behavior of the wheels. This enables the vehicle to correct motion analysis deviations caused by changes in chassis structure and slip errors under complex road conditions, thereby providing high-precision motion mileage estimation results under chassis structure changes and complex slip conditions.

[0038] Please see Figure 3 In some embodiments, the slip coefficient estimate includes an advance slip coefficient estimate and a lateral slip coefficient estimate, and step 203 may include, but is not limited to, steps 301 to 303: Step 301: Determine the vehicle's linear velocity based on the real-time wheel speed data of each drive wheel and the estimated forward slip coefficient. Step 302: Determine the vehicle's angular velocity based on the real-time wheel speed data of each drive wheel and the estimated lateral slip coefficient. Step 303: Based on the linear velocity and angular velocity of the vehicle, determine the motion mileage analysis data of the vehicle within the measurement time.

[0039] In step 301 of some embodiments, the current linear velocity of the vehicle is determined by combining the real-time wheel speed data of each drive wheel with the estimated forward slip coefficient. Specifically, a generalized differential kinematic model is used, with the estimated forward slip coefficient as a correction factor for the longitudinal velocity. Motion analysis is performed by combining the wheel speeds of all drive wheels to correct motion errors caused by slippage and spinning in the driving forward direction, thereby obtaining the vehicle's true linear velocity in the longitudinal direction.

[0040] The linear velocity can be calculated using the following formula (1): (1), in, Indicates linear velocity. This represents the estimated forward slip coefficient. This indicates the real-time wheel speed of the left front wheel. This indicates the real-time wheel speed of the left rear wheel. This indicates the real-time wheel speed of the right front wheel. This indicates the real-time wheel speed of the right rear wheel.

[0041] In step 302 of some embodiments, the current angular velocity of the vehicle is determined based on the real-time wheel speed data of each drive wheel and the estimated lateral slip coefficient. For vehicles using a differential chassis drive system, the steering motion mainly originates from the speed difference between the left and right wheels. Therefore, a generalized differential kinematic model is used to introduce the estimated lateral slip coefficient for correction, in order to calculate the speed difference between the left and right wheel sets. This can compensate for the theoretically calculated angular velocity, thereby accurately calculating the actual angular velocity of the vehicle in the lateral direction.

[0042] The angular velocity can be calculated using the following formula (2): (2), in, Indicates the angular velocity of travel. This represents the estimated value of the lateral slip coefficient. This indicates the wheelbase, which is the lateral distance between the left and right wheels.

[0043] In step 303 of some embodiments, the vehicle's motion mileage analysis data within a measurement time is calculated based on the determined linear velocity and angular velocity. The linear velocity and angular velocity after slip correction are analyzed over a preset measurement time to deduce the specific pose change of the vehicle relative to the previous moment within that measurement time period, thereby determining the vehicle's motion mileage analysis data.

[0044] Through steps 301 to 303, by refining the slip coefficient estimate into two independent dimensions—forward and lateral—and by providing targeted compensation for linear velocity and angular velocity respectively, a decoupled dynamic kinematic model is constructed. For variable wheelbase chassis, the longitudinal and lateral dynamic characteristics vary significantly with changes in chassis structure. By applying the forward slip coefficient estimate and the lateral slip coefficient estimate separately, the true motion state of the vehicle can be reproduced regardless of the wheelbase configuration or the complex terrain, thus providing high-precision motion mileage estimation results under chassis structure changes and complex slip conditions.

[0045] In step 103 of some embodiments, motion integration analysis is performed using inertial measurement unit (IMU) data. The IMU can sense changes in the vehicle's acceleration and angular velocity at high frequency, unaffected by the wheel-ground contact state. Motion integration processing is performed on this high-frequency IMU data to calculate the vehicle's pose change and displacement increment over a very short time, thus obtaining the motion integration data for that measurement time. The motion integration data can quickly respond to the vehicle's dynamic changes, providing a reference trajectory with high short-term accuracy. Especially when wheels slip or become airborne, the motion integration data can provide reliable motion estimation.

[0046] In step 104 of some embodiments, independent motion mileage analysis is performed based on the acquired environmental perception data. Computer vision algorithms or point cloud matching algorithms can be used to extract and match features from images captured by cameras or point clouds scanned by LiDAR. By calculating the relative positional relationship between the features of the current frame and features of historical frames or a local map, the actual displacement of the vehicle in the environment is calculated, thereby obtaining image mileage data within the measurement time. Since the image mileage data is calculated based on external environmental features and does not depend on the chassis mechanical structure or wheel speed, it can effectively correct the cumulative drift error caused by long-term operation.

[0047] Please see Figure 4 In some embodiments, step 104 may include, but is not limited to, steps 401 to 405: Step 401: Based on the preset acquisition time interval, obtain the first environmental image and the second environmental image of the environmental perception data; Step 402: Extract image features based on the first environmental image to obtain the first image features; Step 403: Extract image features based on the second environment image to obtain the second image features; Step 404: Perform feature matching based on the first image features and the second image features to determine the relative displacement data; Step 405: Based on the measurement time and the acquisition time interval, update the first environmental image and the second environmental image to obtain multiple relative displacement data and determine the image mileage data.

[0048] In step 401 of some embodiments, environmental perception data is continuously acquired according to a pre-set acquisition time interval. Specifically, the surrounding environment is continuously scanned or photographed at a fixed high frequency using, for example, a forward-looking camera or a LiDAR sensor. Two frames of data with temporal correlation are extracted from the environmental perception data and defined as a first environmental image and a second environmental image, respectively. The first environmental image and the second environmental image represent the environmental state of the vehicle at the start time and the end time after a very short time interval, respectively, providing material for subsequent calculation of its own motion by comparing environmental changes.

[0049] In step 402 of some embodiments, an image feature extraction operation is performed on the acquired first environmental image. Using computer vision algorithms (such as corner detection or edge detection feature extraction algorithms), key information points with significant geometric or textural characteristics, such as wall corners, ground textures, and plane edges, are identified from the first environmental image. These key information points are then converted into computer-recognizable digital descriptions to obtain the first image features.

[0050] In step 403 of some embodiments, the second environmental image is processed using the same feature extraction logic as in the previous step. The second environmental image is scanned and analyzed to extract key information points that potentially correspond to those in the first environmental image, thereby generating second image features.

[0051] In step 404 of some embodiments, feature matching operations are performed on the first image features and the second image features to determine the relative motion of the vehicle. For example, by using the Iterative Closest Point (ICP) algorithm or the Feature Descriptor Matching algorithm, the correspondence between the first image features and the second image features is found, and by minimizing the geometric distance or reprojection error, the changes in position and attitude of the vehicle in space from the start time to the end time are calculated and determined as relative displacement data, thereby accurately quantifying the movement of the vehicle relative to the surrounding static environment within a very short acquisition time interval.

[0052] In step 405 of some embodiments, during the measurement time, the first and second environmental images are cyclically updated and iteratively calculated through multiple acquisition time intervals. Since a relatively long measurement time typically contains multiple shorter acquisition time intervals, the relative displacement data between this series of consecutive frames can be calculated. Subsequently, these minute relative displacement data are stitched together to form the complete motion trajectory of the vehicle throughout the entire measurement time, thereby ultimately determining the image mileage data for that measurement time.

[0053] Steps 401 to 405 employ an inter-frame feature matching-based approach to determine image mileage data. Since static features in the environment are used as reference anchors, the calculation process is completely independent of the chassis's mechanical transmission system. Therefore, even if the vehicle chassis experiences severe slippage, wheel spin, or if wheel track adjustments cause the kinematic model to temporarily fail, the image mileage data based on environmental features can still maintain a certain level of accuracy. This provides subsequent motion data fusion with objective, reliable motion mileage data that is unaffected by ground friction conditions.

[0054] In step 105 of some embodiments, the motion mileage analysis data, image mileage data, motion integral data and latitude and longitude mileage data obtained in the aforementioned steps are aggregated together to perform motion data fusion in order to obtain target motion mileage data that integrates multiple sources of data.

[0055] The Extended Kalman Filter (EKF) or graph optimization algorithm is typically used as the fusion framework. During the fusion process, the confidence levels of various data sources are evaluated; for example, when wheel slippage is detected, the weight of the motion mileage analysis data is reduced, while the weight of image mileage data or latitude / longitude mileage data is increased. After weighted fusion processing, the final output is a target motion mileage dataset that accurately describes the vehicle's motion process.

[0056] Please see Figure 5 In some embodiments, step 105 may include, but is not limited to, steps 501 to 502: Step 501: Calculate the confidence level of the motion mileage analysis data, image mileage data, motion integral data, and latitude and longitude mileage data to obtain confidence level assessment data. Step 502: Based on the confidence assessment data, determine the target mileage data from the mileage analysis data, image mileage data, mileage integral data, and latitude and longitude mileage data.

[0057] In step 501 of some embodiments, real-time confidence assessments are performed on the input motion mileage analysis data, image mileage data, motion integral data, and latitude and longitude mileage data to obtain confidence assessment data. By analyzing multi-dimensional state information and monitoring the signal quality of the sensors themselves, such as the number of GPS satellites and laser matching scores, and considering the current motion state and chassis shape of the vehicle, such as whether it is turning at high speed and changes in the chassis wheelbase, it is determined whether each data source deviates from the true physical laws at the current moment. For example, if a violent fluctuation in wheel speed is detected while the inertial data is stable, it is determined that the wheels are slipping, and the reliability of each data source at the current moment is quantified to generate corresponding confidence assessment data.

[0058] In step 502 of some embodiments, the motion mileage analysis data, image mileage data, motion integral data, and latitude and longitude mileage data are weighted and fused based on the generated confidence assessment data to determine the final target motion mileage data. Specifically, the weight allocation in the fusion filter is dynamically adjusted according to the confidence assessment data, automatically assigning greater weight to data sources with high confidence, while suppressing or isolating the influence of low-confidence data sources on the final result.

[0059] For example, when confidence assessment data indicates that GPS signals are unreliable, the fusion algorithm reduces their weight and relies more on image mileage data or motion integral data. Through this adaptive weighted calculation, an optimal pose estimate that integrates the advantages of multiple sources and eliminates interference errors is output as the target motion mileage data.

[0060] By using confidence assessment data as the basis for weight allocation in steps 501 and 502, the fusion weight ratio of each sensor data is dynamically adjusted in real time, effectively avoiding motion positioning divergence caused by single sensor failure or interference under specific working conditions. By intelligently filtering and utilizing the most reliable data source at the current moment, it ensures that the final output target motion mileage data always maintains high accuracy and high robustness under various complex and changing working environments.

[0061] Please see Figure 6 In some embodiments, step 501 may include, but is not limited to, steps 601 to 606: Step 601: Obtain signal strength information associated with latitude and longitude mileage data within the measurement time period; Step 602: Obtain the matching confidence of multiple relative displacement data in the image mileage data; Step 603: Determine the motion state information of the vehicle based on the real-time wheel speed data of each drive wheel, the current chassis wheelbase, and the inertial measurement unit data within the measurement time. Step 604: Analyze vehicle slippage based on motion state information to obtain the number of times the vehicle slips. Step 605: Based on the number of vehicle skidding incidents, determine the first confidence level of the motion mileage analysis data and the second confidence level of the motion integral data; Step 606: Based on signal strength information, matching confidence, first confidence and second confidence, obtain confidence assessment data.

[0062] In step 601 of some embodiments, signal strength information associated with latitude and longitude mileage data is read in real time during the measurement period. Indicators such as the number of satellites that the Global Positioning System (GPS) receiver can capture, the Horizontal Dilution of Precision (HDOP), or the carrier-to-noise ratio are obtained. Signal strength information directly reflects the reliability of satellite positioning signals. For example, signal strength is usually high in open areas, while it drops significantly indoors or in environments with many tall buildings. Signal strength information provides an objective basis for assessing the reliability of latitude and longitude mileage data.

[0063] In step 602 of some embodiments, matching confidence scores for multiple relative displacement data are obtained during the image odometry data calculation process. In lidar-based or vision-based odometry methods, a confidence score is generated for each inter-frame match to measure the richness of current environmental features and the overlap of matching results. This matching confidence score can effectively identify the risk of visual positioning failure caused by missing environmental features or excessive dynamic changes in the environment, providing an objective basis for assessing the reliability of image odometry data.

[0064] In step 603 of some embodiments, the motion state information of the vehicle is determined by combining the real-time wheel speed data of each drive wheel within the measurement time, the current chassis wheelbase, and the inertial measurement unit data. By analyzing the dynamic characteristics of these data, it is identified whether the vehicle is currently in a constant-speed straight-line driving, rapid acceleration, high-speed steering, or an unstable state during wheelbase adjustment. The current chassis wheelbase is taken into consideration because the same wheel speed difference corresponds to drastically different dynamic characteristics under different wheelbases.

[0065] In step 604 of some embodiments, vehicle slippage analysis is performed based on the determined motion state information, and the number of vehicle slippages is counted and obtained. The theoretical motion calculated from wheel speed is compared with the actual inertia sensed by the inertial measurement unit. When the deviation between the two exceeds a preset threshold, or when a large steering is detected under a large wheelbase, it is determined that a slippage event has occurred. The statistics of the number of slippages intuitively quantify the physical stability of the current road surface and chassis interaction, and are the core indicator for judging the reliability of motion mileage analysis data.

[0066] In step 605 of some embodiments, a first confidence level for the motion mileage analysis data and a second confidence level for the motion integral data are determined based on the statistically obtained number of vehicle slippages. Logically, a higher number of slippages means a greater error in the wheel speed-based motion mileage analysis data, thus assigning it a lower first confidence level. In contrast, inertial measurement unit (IMU) data is not directly affected by wheel slippage and is relatively more reliable in the short term; therefore, a higher number of slippages will assign a relatively higher second confidence level to the motion integral data.

[0067] In some embodiments, the first confidence level of the motion odometer data and the second confidence level of the motion integral data are also determined based on the magnitude of the slip coefficient estimate. A larger slip coefficient estimate indicates more severe vehicle slippage, which means a larger error is included in the wheel speed-based motion odometer data, thus assigning it a lower first confidence level. In contrast, the inertial measurement unit data is not directly affected by wheel slippage, so a higher number of slippages will result in a relatively higher second confidence level for the motion integral data.

[0068] Inertial measurement units (IMUs) also have errors, mainly manifested as drift and bias that accumulate over time. Therefore, motion integral data obtained based on IMU data also contains errors.

[0069] In step 606 of some embodiments, final confidence assessment data is generated based on the acquired signal strength information, matching confidence, first confidence, and second confidence.

[0070] By incorporating the quality of external environmental signals and the internal physical motion state into the same evaluation system through steps 601 to 606, the generated confidence evaluation data can accurately guide subsequent motion data fusion.

[0071] In some embodiments, the method further includes using a neural network model to perform confidence assessment on motion mileage analysis data, image mileage data, motion integral data, and latitude and longitude mileage data to obtain confidence assessment data. Please see Figure 7 In some embodiments, step 502 may include, but is not limited to, steps 701 to 702: Step 701: Determine the first weight value of the exercise mileage analysis data and the second weight value of the exercise integral data based on the first confidence level and the second confidence level; Step 702: Determine the target mileage data based on the first weight value, mileage analysis data, second weight value, and mileage integral data.

[0072] In step 701 of some embodiments, weight parameters for subsequent fusion calculations are quantified based on the first and second confidence levels calculated in previous steps. For the first confidence level, which reflects the reliability of wheel speed estimation, a higher value indicates less vehicle slippage, thus assigning it a larger first weight value. Conversely, if the first confidence level decreases due to detected frequent slippage, the first weight value decreases accordingly. Similarly, the corresponding second weight value is determined based on the level of the second confidence level. Through numerical methods, the qualitative assessment of slippage degree and motion state is transformed into weight coefficients that can be directly used by the mathematical model. In the application scenario of extended Kalman filtering, this process can be represented by dynamically adjusting the process noise matrix Q and the observation noise matrix R.

[0073] In step 702 of some embodiments, a final fusion calculation is performed using determined weight values. This step applies a first weight value to the motion mileage analysis data, applies a second weight value to the motion integral data, and calculates the optimal target motion mileage data using a state update equation based on a weighted average or Kalman filter.

[0074] Through this weighted dynamic synthesis, the position and attitude information that is closest to the real physical motion trajectory can be calculated in real time, ensuring that the final output target motion mileage data is the optimal solution after slip compensation and dynamic weighting.

[0075] Please see Figure 8 In some embodiments, the embodiments of this application may also include, but are not limited to, steps 801 to 803: Step 801: In response to the signal strength information exceeding a preset first strength value, the latitude and longitude mileage data is determined as the target enhancement value; Step 802: In response to the matching confidence exceeding a preset second confidence value, the image mileage data is determined as the target enhancement value; Step 803: Based on the target reinforcement value, determine the first confidence level of the exercise mileage analysis data and the second confidence level of the exercise integral data.

[0076] In step 801 of some embodiments, the signal quality of the satellite positioning system is continuously monitored. When the signal strength information is detected to exceed a preset first strength value, it indicates that the vehicle is currently in an open, unobstructed area, and the satellite positioning data has extremely high accuracy and reliability. At this time, the latitude and longitude mileage data is directly determined as the target enhancement value, that is, it is regarded as the true value used for the target motion mileage data at the current moment; In step 802 of some embodiments, the matching quality of the visual or laser sensor is evaluated. When the matching confidence score output by the environment perception algorithm exceeds a preset second confidence value, it indicates that the current environment has rich features, clear texture, and the algorithm has converged well. At this time, the image odometry data is determined as the target enhancement value and used as the ground truth from another reliable source.

[0077] In step 803 of some embodiments, the confidence level of the internal model is back-calibrated and updated using the determined target reinforcement value. The deviation between the motion distance analysis data and the motion integral data and the target reinforcement value is calculated, and this deviation is used as a feedback signal. Based on this deviation, the error level of the current kinematic model and integral algorithm under specific terrain can be accurately assessed, thereby redetermining the first confidence level of the motion distance analysis data and the second confidence level of the motion integral data. Furthermore, an online learning mechanism can be combined to construct a loss function using the calculated error, fine-tuning the weights of the neural network model so that the neural network model can dynamically adapt to the current chassis morphology and road surface characteristics.

[0078] Through steps 801 to 803, when the high-precision sensor is functioning well, these high-quality data can be automatically captured as true values, and confidence correction can be performed on motion mileage analysis data and motion integral data that are prone to cumulative errors. This enables continuous adaptation to unknown complex terrain and errors caused by chassis mechanical aging through online learning, thereby significantly improving the robustness and generalization ability in the process of motion data fusion to obtain target motion mileage data.

[0079] Please see Figure 9This application embodiment also provides a schematic structure of a variable wheelbase chassis, which can realize the motion estimation method under the above differential control. The variable wheelbase chassis includes a supporting front leg 50 and a supporting rear leg 60, and a supporting joint constructed by a joint drive device 20 and a joint shaft 21. The joint drive device 20 can be a high torque density servo motor or a drive module with deceleration function. Its output end is connected to the joint shaft 21, and it can accurately output rotational torque and angle.

[0080] The supporting front leg 50 and supporting rear leg 60 are hinged or fixedly connected around the joint axis 21, and the supporting joint can adjust the position of the supporting front leg 50 and supporting rear leg 60. Driven by the joint drive device 20, the supporting joint can flexibly adjust the angle or relative position between the supporting front leg 50 and supporting rear leg 60. For example, the joint drive device 20 can control the two legs to retract inward to shorten the wheelbase, or to extend outward to lower the center of gravity, or even control the pitch angle of the two legs relative to the horizontal plane, thereby realizing the robot's folding and unfolding actions, providing the mechanical execution basis for the aforementioned center of gravity planning and obstacle crossing actions.

[0081] Front wheels 30 are mounted at the ends of the supporting front legs 50, and rear wheels 40 are mounted at the ends of the supporting rear legs 60. These two wheels typically serve as active drive wheels, integrating hub motors or connecting to drive motors via transmission mechanisms, responsible for providing the traction required for the robot to move forward, backward, and perform differential steering. During obstacle crossing, the front wheels 30 and rear wheels 40 not only undertake the task of gripping the ground and walking, but also, in coordination with the lifting and lowering movements of the supporting legs, alternately serve as the robot's fulcrum.

[0082] In some embodiments, to further improve the chassis's passability in complex terrain, the wheeled robot also has driven wheels 70 at the support joints. These driven wheels 70 are typically designed as passively rotating driven wheels, and their installation height or radius may differ from the front and rear wheels. The main function of the driven wheels 70 is to prevent the chassis from touching the ground. When the robot crosses sharp, protruding obstacles or moves in a large-angle folding posture, the middle driven wheel 70 can contact the obstacle or the ground to provide additional support points, preventing the delicate joint drive mechanism 20 from directly colliding rigidly with the environment, thereby protecting the core joints and assisting the robot in smooth transitions.

[0083] The specific implementation of the motion estimation device under differential control is basically the same as the specific implementation of the motion estimation method under differential control described above, and will not be repeated here.

[0084] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the motion estimation method under differential control described above. This electronic device can be any smart terminal, including a tablet computer, an in-vehicle computer, or similar device.

[0085] Please see Figure 10 , Figure 10 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes: The processor 1001 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 1002 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 1002 can store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1002 and is called and executed by the processor 1001 using the motion estimation method under differential control according to the embodiments of this application. Input / output interface 1003 is used to implement information input and output; The communication interface 1004 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 1005 transmits information between various components of the device (e.g., processor 1001, memory 1002, input / output interface 1003, and communication interface 1004); The processor 1001, memory 1002, input / output interface 1003 and communication interface 1004 are connected to each other within the device via bus 1005.

[0086] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the motion estimation method under differential control described above.

[0087] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0088] The motion estimation method, electronic device, and storage medium under differential control provided in this application embodiment acquire real-time wheel speed data of each drive wheel, acquire inertial measurement unit data and current chassis wheelbase of the variable wheelbase chassis, acquire environmental perception data and latitude and longitude mileage data based on a preset measurement time, perform vehicle motion analysis based on real-time wheel speed data and current chassis wheelbase, output motion mileage analysis data of the vehicle within the measurement time, perform motion integral analysis based on inertial measurement unit data to obtain motion integral data within the measurement time, perform motion mileage analysis based on environmental perception data to obtain image mileage data within the measurement time, and perform motion data fusion based on motion mileage analysis data, image mileage data, motion integral data, and latitude and longitude mileage data to obtain target motion mileage data. Therefore, this application obtains the real-time current chassis wheelbase and incorporates it into the motion analysis process to obtain motion mileage analysis data. It then fuses motion mileage analysis data, image mileage data, motion integral data, and latitude and longitude mileage data from different sources and with different characteristics to correct motion analysis deviations caused by chassis structure changes and slippage errors under complex road conditions. This provides high-precision motion mileage estimation results under chassis structure changes and complex slippage conditions.

[0089] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0090] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0091] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0092] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0093] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0094] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0095] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. The coupling or direct coupling or communication connection between the shown or discussed units may be through some interfaces, or indirect coupling or communication connection between the apparatus or units, and may be electrical, mechanical, or other forms.

[0096] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0097] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0098] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0099] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A motion estimation method under differential control, characterized in that, Applied to a vehicle equipped with a variable wheelbase chassis, the variable wheelbase chassis being configured with a plurality of drive wheels, the method includes: Based on a preset measurement time, real-time wheel speed data of each drive wheel is acquired, inertial measurement unit data and current chassis wheelbase of the variable wheelbase chassis are acquired, and environmental perception data and latitude and longitude mileage data are acquired. Based on the real-time wheel speed data and the current chassis wheelbase, vehicle motion analysis is performed, and the vehicle's motion mileage analysis data within the measurement time is output. Motion integral analysis is performed based on the data from the inertial measurement unit to obtain the motion integral data within the measurement time. Based on the environmental perception data, motion mileage analysis is performed to obtain image mileage data within the measurement time. The target motion mileage data is obtained by fusing motion data based on the motion mileage analysis data, the image mileage data, the motion integral data, and the latitude and longitude mileage data.

2. The method according to claim 1, characterized in that, The process of analyzing vehicle motion based on the real-time wheel speed data and the current chassis wheelbase, and outputting vehicle mileage analysis data within the measured time period, includes: Environmental terrain analysis is performed based on the collected environmental perception data to obtain the road surface friction coefficient; Based on the current chassis wheelbase, the road surface friction coefficient, and the real-time wheel speed data, the slip coefficient is analyzed by inputting the data into a pre-trained slip ratio estimation model, and the slip coefficient estimate of each drive wheel is output. Based on the estimated slip coefficient of each drive wheel and the real-time wheel speed data, vehicle motion analysis is performed, and the vehicle's motion mileage analysis data within the measurement time is output.

3. The method according to claim 2, characterized in that, The slip coefficient estimates include forward slip coefficient estimates and lateral slip coefficient estimates. The process of analyzing vehicle motion based on the real-time wheel speed data and the current chassis wheelbase, and outputting vehicle mileage analysis data within the measured time period, includes: The vehicle's linear velocity is determined based on the real-time wheel speed data of each of the drive wheels and the estimated forward slip coefficient. The vehicle's angular velocity is determined based on the real-time wheel speed data of each of the drive wheels and the estimated lateral slip coefficient. Based on the linear velocity and angular velocity, the motion mileage analysis data of the vehicle within the measurement time is determined.

4. The method according to claim 1, characterized in that, The step of performing motion mileage analysis based on the environmental perception data to obtain image mileage data within the measurement time includes: Based on a preset acquisition time interval, a first environmental image and a second environmental image of the environmental perception data are obtained; Based on the first environmental image, image features are extracted to obtain the first image features; Image features are extracted based on the second environmental image to obtain the second image features; Based on the first image features and the second image features, feature matching is performed to determine the relative displacement data; Based on the measurement time and the acquisition time interval, the first environmental image and the second environmental image are updated to obtain multiple relative displacement data and determine the image mileage data.

5. The method according to claim 1, characterized in that, The process of fusing motion data based on the motion mileage analysis data, the image mileage data, the motion integral data, and the latitude and longitude mileage data to obtain target motion mileage data includes: The confidence level of the motion mileage analysis data, the image mileage data, the motion integral data, and the latitude and longitude mileage data is evaluated to obtain confidence evaluation data; Based on the confidence assessment data, the target mileage data is determined from the mileage analysis data, the image mileage data, the mileage integral data, and the latitude and longitude mileage data.

6. The method according to claim 5, characterized in that, The step of inputting the motion mileage analysis data, the image mileage data, the motion integral data, and the latitude and longitude mileage data into the confidence assessment model for confidence assessment to obtain confidence assessment data includes: Obtain signal strength information associated with the latitude and longitude mileage data within the measurement time period; Obtain the matching confidence of multiple relative displacement data in the image mileage data; Based on the real-time wheel speed data of each drive wheel, the current chassis wheelbase, and the inertial measurement unit data within the measurement time, the motion state information of the vehicle is determined. Based on the motion state information, vehicle slippage analysis is performed to obtain the number of vehicle slippages; Based on the number of vehicle skidding incidents, a first confidence level for the motion mileage analysis data and a second confidence level for the motion integral data are determined. The confidence assessment data is obtained based on the signal strength information, the matching confidence level, the first confidence level, and the second confidence level.

7. The method according to claim 6, characterized in that, The step of determining the target mileage data from the mileage analysis data, the image mileage data, the mileage integral data, and the latitude and longitude mileage data based on the confidence assessment data includes: The first weight value of the exercise mileage analysis data and the second weight value of the exercise points data are determined based on the first confidence level and the second confidence level. The target mileage data is determined based on the first weight value, the mileage analysis data, the second weight value, and the mileage integral data.

8. The method according to claim 6, characterized in that, The method further includes: In response to the signal strength information exceeding a preset first strength value, the latitude and longitude mileage data is determined as the target enhancement value; In response to the matching confidence exceeding a preset second confidence value, the image mileage data is determined as the target enhancement value; Based on the target enhancement value, a first confidence level for the exercise mileage analysis data and a second confidence level for the exercise integral data are determined.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the motion estimation method under differential control as described in any one of claims 1 to 8.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the motion estimation method under differential control as described in any one of claims 1 to 8.

Images