A Transformer Oil Sampling Robot Localization Method and System Based on Multi-Source Data Fusion

Abstract

This invention discloses a method and system for locating a transformer oil sampling robot based on multi-source data fusion. The method includes: acquiring IMU, optical flow sensor, and GPS data; preprocessing the data using a second-order Butterworth filter; and automatically selecting a fusion algorithm based on the scene: indoors, complementary filtering is used to dynamically fuse optical flow and IMU data, and tilt interference is compensated through weight factor adjustment and error feedback closed-loop; outdoors, extended state Kalman filtering is used to fuse GPS and IMU data, and electromagnetic interference is suppressed by introducing accelerometer zero-bias compensation and dynamic noise covariance adaptation. This invention effectively solves the problems of insufficient robot positioning accuracy, weak anti-interference capability, and scene switching discontinuity in the complex environment of converter stations, achieving high-precision continuous positioning with high positioning error indoors, and meeting the operational requirements for precise docking of transformer oil sampling ports.

Description

Technical Field

[0001] This invention belongs to the field of robot positioning technology, and in particular relates to a positioning method and system for a transformer oil sampling robot based on multi-source data fusion. Background Technology

[0002] With the continuous upgrading and intelligent transformation of power systems, the requirements for condition monitoring of key equipment are increasing. Oil sampling of converter station transformers is a crucial step in ensuring their safe and stable operation. Traditional manual oil sampling methods are not only inefficient but also pose safety risks to personnel in the high-voltage, strong electromagnetic field environment of converter stations. Therefore, intelligent robots capable of automating oil sampling operations have become an important development direction. One of the core supporting technologies for achieving autonomous and precise robot operation is its high-precision positioning system.

[0003] Currently, the positioning technology for oil extraction robots has evolved from relying on a single sensor to employing a multi-sensor data fusion approach. Inertial Measurement Units (IMUs) offer good dynamic response performance, but their integration operations can lead to error accumulation and divergence. Optical flow sensors achieve high accuracy in well-lit, textured indoor environments, but their velocity measurements are susceptible to interference from changes in the carrier's tilt angle. Global Positioning System (GPS) is suitable for open outdoor environments, but the presence of high-voltage equipment and metal structures within converter stations generates strong electromagnetic interference and multipath effects, significantly degrading GPS signal quality and causing substantial fluctuations in its observations. Summary of the Invention

[0004] This invention provides a transformer oil sampling robot positioning method and system based on multi-source data fusion, which is used to solve the technical problem that the strong electromagnetic interference and multipath effect generated by the high-voltage equipment and metal structure distributed in the converter station will significantly degrade the GPS signal quality, resulting in large fluctuations in its observation values.

[0005] In a first aspect, the present invention provides a method for locating a transformer oil sampling robot based on multi-source data fusion, comprising:

[0006] The IMU data, optical flow sensor data, and GPS data of the transformer oil sampling robot are acquired, and the IMU data, optical flow sensor data, and GPS data are preprocessed to obtain target IMU data, target optical flow sensor data, and target GPS data, respectively.

[0007] Based on the target GPS data and the target optical flow sensor data, the current environmental scene of the transformer oil sampling robot is determined, and the environmental scene includes indoor scene and outdoor scene;

[0008] If the scene is indoors, the target optical flow sensor data and the target IMU data are fused based on a preset complementary filtering strategy to obtain an indoor location estimate.

[0009] If the scene is outdoors, the target IMU data and the target GPS data are fused based on the Kalman filter algorithm to obtain an outdoor location estimate.

[0010] Secondly, the present invention provides a transformer oil sampling robot positioning system based on multi-source data fusion, comprising:

[0011] The acquisition module is configured to acquire IMU data, optical flow sensor data, and GPS data of the transformer oil sampling robot, and preprocess the IMU data, optical flow sensor data, and GPS data to obtain target IMU data, target optical flow sensor data, and target GPS data, respectively.

[0012] The judgment module is configured to determine the current environmental scene of the transformer oil sampling robot based on the target GPS data and the target optical flow sensor data, wherein the environmental scene includes indoor scene and outdoor scene;

[0013] The first positioning module is configured to, if in an indoor scene, fuse the target optical flow sensor data and the target IMU data based on a preset complementary filtering strategy to obtain an indoor position estimate.

[0014] The second positioning module is configured to, if in an outdoor scene, fuse the target IMU data and the target GPS data based on the Kalman filter algorithm to obtain an outdoor location estimate.

[0015] Thirdly, an electronic device is provided, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the steps of the transformer oil sampling robot positioning method based on multi-source data fusion according to any embodiment of the present invention.

[0016] Fourthly, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor performs the steps of the transformer oil sampling robot positioning method based on multi-source data fusion according to any embodiment of the present invention.

[0017] This application presents a transformer oil sampling robot localization method and system based on multi-source data fusion. First, in indoor scenarios, a complementary filtering strategy with dynamically adjusted weight factors is used to fuse optical flow and IMU data. A three-level correction mechanism based on closed-loop feedback of position and velocity errors is introduced to effectively compensate for the velocity measurement deviation of the optical flow sensor caused by the robot climbing inclined cabinets. Second, in outdoor scenarios, by employing an extended state vector (introducing an accelerometer zero-bias compensation term) and a Kalman filter algorithm adapted to dynamic noise covariance, the system's ability to resist GPS observation noise under strong electromagnetic interference is significantly enhanced, ensuring the safety and accuracy of path planning for long-distance outdoor movement. Finally, through a dual-mode fusion architecture of "indoor complementary filtering + outdoor Kalman filtering" and automatic scene switching logic based on signal strength, the discontinuity problem during indoor and outdoor positioning scene switching is successfully solved, achieving continuous, stable, and high-precision positioning of the robot throughout the entire operation process. This comprehensively ensures the reliability, accuracy, and continuity of the transformer oil sampling robot's autonomous operation. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A flowchart illustrating a transformer oil sampling robot positioning method based on multi-source data fusion, provided as an embodiment of the present invention;

[0020] Figure 2 A schematic diagram of the positioning trajectory fused from optical flow sensor data and IMU data is provided as an embodiment of the present invention.

[0021] Figure 3 An indoor fusion positioning error comparison chart is provided for a specific embodiment of the present invention;

[0022] Figure 4 A schematic diagram of GPS data and IMU data fusion positioning trajectory is provided as an embodiment of the present invention;

[0023] Figure 5 An outdoor fusion positioning error comparison chart is provided for a specific embodiment of the present invention;

[0024] Figure 6 This is a structural block diagram of a transformer oil sampling robot positioning system based on multi-source data fusion, provided in an embodiment of the present invention.

[0025] Figure 7This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] Please see Figure 1 The diagram shows a flowchart of a transformer oil sampling robot positioning method based on multi-source data fusion, as described in this application.

[0028] like Figure 1 As shown, the transformer oil sampling robot positioning method specifically includes the following steps:

[0029] Step S101: Acquire the IMU data, optical flow sensor data, and GPS data of the transformer oil sampling robot, and preprocess the IMU data, optical flow sensor data, and GPS data to obtain target IMU data, target optical flow sensor data, and target GPS data, respectively.

[0030] In this step, dynamic zero-bias compensation and nonlinear error calibration are performed on the IMU data, and the optical flow sensor data and the GPS data are preprocessed using a second-order Butterworth low-pass filter to obtain the target IMU data, target optical flow sensor data and target GPS data, respectively.

[0031] Step S102: Based on the target GPS data and the target optical flow sensor data, determine the current environmental scene of the transformer oil sampling robot. The environmental scene includes indoor scene and outdoor scene.

[0032] Step S103: If the scene is indoors, the target optical flow sensor data and the target IMU data are fused based on a preset complementary filtering strategy to obtain an indoor location estimate.

[0033] In this step, during the horizontal positioning process of the transformer oil sampling robot, its motion state induces changes in the feature images captured by the optical flow sensor, thereby causing deviations in the horizontal position calculation results. To effectively correct this error, gyroscope data needs to be introduced, and a complementary filtering strategy is implemented to achieve the fusion correction of the optical flow sensor position estimation and gyroscope information.

[0034] The optical flow velocity and the gyroscope angular velocity are fused to obtain the fused optical flow velocity. Based on this fused optical flow velocity and the height between the optical flow sensor and the ground, the planar optical flow velocity is calculated. The expression for calculating the fused optical flow velocity is as follows:

[0035] ,

[0036] In the formula, To merge optical flow velocities, These are the weighting coefficients. Optical flow velocity, This refers to the angular velocity of the gyroscope.

[0037] The expression for calculating the plane velocity of the optical flow is:

[0038] ,

[0039] ,

[0040] In the formula, The optical flow plane velocity is in the X-axis direction. The fused optical flow velocity is located in the X-axis direction. The optical flow plane velocity is in the Y-axis direction. The fused optical flow velocity is located in the Y-axis direction. The height between the optical flow sensor and the ground;

[0041] Integrating the plane velocity of the optical flow using the forward Euler method yields the plane optical flow position estimate, expressed as:

[0042] ,

[0043] ,

[0044] In the formula, Let X be the position of the planar optical flow along the X-axis after integration. This represents the position of the planar optical flow along the Y-axis after integration. The position of the planar optical flow in the X-axis direction. The position of the planar optical flow in the Y-axis direction. It is a quantity that changes over time;

[0045] The position error between the optical flow position estimate and the IMU inertial navigation solution is calculated using the following expression:

[0046] ,

[0047] ,

[0048] ,

[0049] ,

[0050] In the formula, The position error is calculated by integrating the optical flow plane position along the X-axis with the IMU accelerometer. The position error is calculated by integrating the optical flow plane position along the Y-axis with the IMU accelerometer. The velocity error is calculated by integrating the optical flow plane velocity along the X-axis with the IMU accelerometer readings. The velocity error is calculated by integrating the optical flow plane velocity along the Y-axis with the IMU accelerometer readings. The position of the planar optical flow in the X-axis direction. The position of the planar optical flow in the Y-axis direction. The position is calculated by integrating the IMU accelerometer readings along the X-axis. The position is calculated by integrating the IMU accelerometer readings along the Y-axis. The velocity generated by integrating the IMU accelerometer readings in the X-axis direction. The velocity generated is calculated by integrating the IMU accelerometer readings in the Y-axis direction;

[0051] Based on the position error, an acceleration correction, a velocity correction, and a position correction are generated by a proportional-integral controller. The expression for calculating the acceleration correction is as follows:

[0052] ,

[0053] ,

[0054] In the formula, X-axis direction Constant acceleration correction amount Y-axis direction Constant acceleration correction amount This represents the acceleration correction amount in the X-axis direction at time t. This is the acceleration correction amount in the Y-axis direction t. This is the acceleration correction coefficient;

[0055] The expression for calculating the speed correction amount is:

[0056] ,

[0057] ,

[0058] In the formula, This is the planar optical flow velocity correction amount in the X-axis direction. This is the planar optical flow velocity correction amount in the Y-axis direction. For speed correction coefficient;

[0059] The expression for calculating the position correction amount is:

[0060] ,

[0061] ,

[0062] In the formula, This is the planar optical flow position correction amount in the X-axis direction. This is the planar optical flow position correction amount in the Y-axis direction. This is the position correction coefficient;

[0063] Based on the acceleration correction, velocity correction, and position correction, a closed-loop feedback correction is performed on the velocity state and position state to obtain an indoor position estimate. The expression for the corrected velocity state is:

[0064] ,

[0065] ,

[0066] ,

[0067] ,

[0068] ,

[0069] ,

[0070] In the formula, The fusion speed is in the X-axis direction. The original inertial navigation system velocity is located in the X-axis direction. This represents the change in velocity along the X-axis. The fusion speed is in the Y-axis direction. The original inertial navigation system velocity is in the Y-axis direction. The change in velocity along the Y-axis. The fusion acceleration is in the X-axis direction. The fusion acceleration is in the Y-axis direction. The acceleration of the original inertial navigation system in the X-axis direction. The original inertial navigation system acceleration is in the Y-axis direction;

[0071] The expression for correcting the position state is:

[0072] ,

[0073] ,

[0074] In the formula, This represents the fusion position along the X-axis. This represents the fusion position along the Y-axis. This represents the original position of the inertial navigation system along the X-axis. The original position of the inertial navigation system in the Y-axis direction;

[0075] In one specific embodiment, a simulation experiment of indoor fusion positioning is conducted based on a complementary filtering strategy. A particle is placed in the simulation environment and moves along a closed curve. During the particle's movement, an indeterminate tilt angle perturbation within the range of ±0.1 radians is introduced, and the horizontal position perturbation is calculated based on the height relative to the ground. This perturbation is then incorporated into the particle's trajectory to simulate the trajectory of optical flow-only positioning. Figure 2 As shown, the angular velocity calculated from the tilt disturbance is measured using a simulated gyroscope. The experimental results are as follows:

[0076] from Figure 3 The experimental results show that the positioning effect after complementary filtering by the optical flow sensor and IMU is significantly better than that of optical flow positioning alone. The positioning error of optical flow positioning alone reached a maximum of 0.14 meters during the experiment. The introduction of IMU gyroscope data effectively compensates for the deficiency of optical flow sensor in positioning under tilt interference, and can maintain high positioning accuracy, especially under large turns. After fusion positioning, the positioning error does not exceed 0.07 meters.

[0077] Step S104: If the scene is outdoors, the target IMU data and the target GPS data are fused based on the Kalman filter algorithm to obtain an outdoor location estimate.

[0078] In this step, based on the fusion of target GPS data and target IMU data using an extended Kalman filter, the state vector of the Kalman filter is defined as follows:

[0079] ,

[0080] In the formula, For state vectors, The horizontal position. The velocity is in the horizontal direction;

[0081] During the prediction phase, based on IMU acceleration data, state prediction is performed using a kinematic model, and the predicted state covariance is calculated. The expression for state prediction is:

[0082] ,

[0083] In the formula, This is the predicted position value at time k. This is the predicted position value at time k-1. For the time change, For the measured acceleration, To achieve zero bias in the accelerometer, The velocity prediction value at time k. The velocity prediction value at time k-1;

[0084] The expression for calculating the predicted state covariance is:

[0085] ,

[0086] ,

[0087] ,

[0088] In the formula, This is the predicted value of the state covariance. Here is the state transition matrix. The initial state value, For process noise covariance, For location noise covariance, For velocity noise covariance;

[0089] During the update phase, the Kalman gain is calculated using GPS-observed location data, the state vector and covariance matrix are updated, and the accelerometer bias is dynamically corrected through residual feedback to obtain the final outdoor position estimate.

[0090] The expression for calculating Kalman gain is:

[0091] ,

[0092] In the formula, For Kalman gain, To estimate the error covariance matrix a priori, For the observation matrix, To observe the noise covariance, , and These are the position observation noise covariance and the velocity observation noise covariance, respectively.

[0093] The expression for state correction is:

[0094] ,

[0095] In the formula, For posterior state estimation, For prior state estimation, These are GPS observations;

[0096] The expression for covariance update is:

[0097] ,

[0098] In the formula, It is the identity matrix. To estimate the covariance matrix for the posterior time;

[0099] Accelerometer zero bias is corrected through residual feedback. The expression is:

[0100] ,

[0101] In the formula, For the Kalman gain component for accelerometer zero bias correction;

[0102] In one specific embodiment, a point mass is placed in the simulation environment with an initial velocity of 0, and it undergoes non-uniform acceleration along a tangential direction of a path 50 meters long in the east and 30 meters long in the north. Gaussian noise is added to the feedback position of this point mass to simulate the GPS observation position in a real environment.

[0103] In summary, the method of this application can achieve the following technical effects:

[0104] Significantly improves indoor positioning accuracy and solves tilt interference problems: By fusing optical flow sensor and IMU data through complementary filtering, and introducing gyroscope angular velocity to compensate for the tilt interference defects of the optical flow sensor, experiments show that the maximum positioning error of optical flow alone is 0.14 meters, while the maximum positioning error after fusion is no more than 0.07 meters, an improvement in accuracy of 50%. Especially in scenarios where the robot moves tilted along the transformer cabinet (simulating ±0.1 radian tilt angle disturbance) or turns, the weighting factor α can dynamically adjust the dependence ratio of gyroscope and optical flow, effectively suppressing optical flow jitter and ensuring the high-precision positioning required for oil port alignment.

[0105] Optimizing outdoor positioning anti-interference capabilities and reducing the impact of GPS noise: A GPS and IMU fusion scheme based on Kalman filtering compensates for long-term IMU drift by incorporating the accelerometer zero bias (ba) into the extended state vector, and dynamically optimizes the process noise covariance Q and observation noise covariance R. Experiments show that the average error of a single GPS observation is 0.6 meters, with a maximum of 1 meter. After fusion, the positioning error is reduced to about 0.3 meters, with a maximum of no more than 0.5 meters. This can cope with the electromagnetic interference of outdoor high-voltage equipment in converter stations and meet the path planning accuracy requirements for long-distance outdoor robot movement.

[0106] Achieving continuous positioning across all scenarios and adapting to oil extraction operation requirements: A dual-scheme architecture combining indoor optical flow-IMU complementary filtering and outdoor GPS-IMU Kalman filtering, along with error feedback closed-loop control (three-level correction for acceleration, velocity, and position), covers the entire operational scenario of the transformer oil extraction robot, from the indoor equipment area to the outdoor passageway. Closed-loop control quickly converges accumulated errors, avoiding sudden positional changes or velocity oscillations, ensuring the stability and continuity of the robot's autonomous navigation and precise docking with the oil extraction port.

[0107] from Figure 4 Experimental results and Figure 5 The error comparison images clearly show that the average observation error of GPS is around 0.6 meters, with a maximum of 1 meter. After data fusion using a Kalman filter and IMU, the error between the estimated position and the actual trajectory is reduced to around 0.3 meters, with a maximum of no more than 0.5 meters. Experimental results indicate that the error between the estimated position and the actual trajectory after GPS / IMU fusion using a Kalman filter is significantly smaller than the observation error of GPS.

[0108] Please see Figure 6 The diagram shows a structural block diagram of a transformer oil sampling robot positioning system based on multi-source data fusion according to this application.

[0109] like Figure 6 As shown, the transformer oil sampling robot positioning system 200 includes an acquisition module 210, a judgment module 220, a first positioning module 230, and a second positioning module 240.

[0110] The acquisition module 210 is configured to acquire IMU data, optical flow sensor data, and GPS data of the transformer oil sampling robot, and preprocess the IMU data, optical flow sensor data, and GPS data to obtain target IMU data, target optical flow sensor data, and target GPS data, respectively. The judgment module 220 is configured to determine the current environmental scene of the transformer oil sampling robot based on the target GPS data and the target optical flow sensor data, wherein the environmental scene includes an indoor scene and an outdoor scene. The first positioning module 230 is configured to fuse the target optical flow sensor data and the target IMU data based on a preset complementary filtering strategy to obtain an indoor position estimate if the robot is in an indoor scene. The second positioning module 240 is configured to fuse the target IMU data and the target GPS data based on a Kalman filter algorithm to obtain an outdoor position estimate if the robot is in an outdoor scene.

[0111] It should be understood that Figure 6 The modules and references described in the document Figure 1 The steps described in the text correspond to those in the method described above. Therefore, the operations, features, and corresponding technical effects described above also apply to the method described in the text. Figure 6 The various modules in the document will not be described in detail here.

[0112] In other embodiments, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor performs the transformer oil sampling robot positioning method based on multi-source data fusion in any of the above method embodiments.

[0113] In one embodiment, the computer-readable storage medium of the present invention stores computer-executable instructions, which are configured as follows:

[0114] The IMU data, optical flow sensor data, and GPS data of the transformer oil sampling robot are acquired, and the IMU data, optical flow sensor data, and GPS data are preprocessed to obtain target IMU data, target optical flow sensor data, and target GPS data, respectively.

[0115] Based on the target GPS data and the target optical flow sensor data, the current environmental scene of the transformer oil sampling robot is determined, and the environmental scene includes indoor scene and outdoor scene;

[0116] If the scene is indoors, the target optical flow sensor data and the target IMU data are fused based on a preset complementary filtering strategy to obtain an indoor location estimate.

[0117] If the scene is outdoors, the target IMU data and the target GPS data are fused based on the Kalman filter algorithm to obtain an outdoor location estimate.

[0118] Computer-readable storage media may include a stored program area and a stored data area, wherein the stored program area may store an operating system and an application program required for at least one function; the stored data area may store data created based on the use of the transformer oil sampling robot positioning system based on multi-source data fusion, etc. Furthermore, the computer-readable storage medium may include high-speed random access memory, and may also include memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, the computer-readable storage medium may optionally include memory remotely disposed relative to a processor, which can be connected to the transformer oil sampling robot positioning system based on multi-source data fusion via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0119] Figure 7 This is a schematic diagram of the structure of the electronic device provided in the embodiment of the present invention, such as... Figure 7As shown, the device includes a processor 310 and a memory 320. The electronic device may also include an input device 330 and an output device 340. The processor 310, memory 320, input device 330, and output device 340 can be connected via a bus or other means. Figure 7 Taking a bus connection as an example, the memory 320 is the computer-readable storage medium described above. The processor 310 executes various server functions and data processing by running non-volatile software programs, instructions, and modules stored in the memory 320, thereby implementing the transformer oil sampling robot positioning method based on multi-source data fusion as described in the above embodiment. The input device 330 can receive input digital or character information and generate key signal inputs related to user settings and function control of the transformer oil sampling robot positioning system based on multi-source data fusion. The output device 340 may include a display screen or other display device.

[0120] The aforementioned electronic device can execute the method provided in the embodiments of the present invention, and has the corresponding functional modules and beneficial effects for executing the method. Technical details not described in detail in this embodiment can be found in the method provided in the embodiments of the present invention.

[0121] In one implementation, the aforementioned electronic device is applied to a transformer oil sampling robot positioning system based on multi-source data fusion, serving as a client, and includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to:

[0122] The IMU data, optical flow sensor data, and GPS data of the transformer oil sampling robot are acquired, and the IMU data, optical flow sensor data, and GPS data are preprocessed to obtain target IMU data, target optical flow sensor data, and target GPS data, respectively.

[0123] Based on the target GPS data and the target optical flow sensor data, the current environmental scene of the transformer oil sampling robot is determined, and the environmental scene includes indoor scene and outdoor scene;

[0124] If the scene is indoors, the target optical flow sensor data and the target IMU data are fused based on a preset complementary filtering strategy to obtain an indoor location estimate.

[0125] If the scene is outdoors, the target IMU data and the target GPS data are fused based on the Kalman filter algorithm to obtain an outdoor location estimate.

[0126] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of embodiments.

[0127] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A transformer oil taking robot positioning method based on multi-source data fusion, characterized in that, The application relates to a transformer oil extraction robot and a position estimation method thereof. IMU data, optical flow sensor data and GPS data of the transformer oil extraction robot are acquired, and the IMU data, the optical flow sensor data and the GPS data are preprocessed to obtain target IMU data, target optical flow sensor data and target GPS data respectively. The IMU data is subjected to dynamic zero offset compensation and nonlinear error calibration, and the optical flow sensor data and the GPS data are preprocessed by adopting a second-order Butterworth low-pass filter to obtain target IMU data, target optical flow sensor data and target GPS data respectively. According to the target GPS data and the target optical flow sensor data, the current environment scene of the transformer oil extraction robot is judged, and the environment scene comprises an indoor scene and an outdoor scene. If the indoor scene is adopted, the target optical flow sensor data and the target IMU data are fused based on a preset complementary filtering strategy to obtain indoor position estimation. If the outdoor scene is adopted, the target IMU data and the target GPS data are fused based on a Kalman filtering algorithm to obtain outdoor position estimation. The target GPS data and the target IMU data are fused based on an extended Kalman filter, and a state vector of the Kalman filter is defined as follows: ， wherein is the state vector, is the horizontal position, is the horizontal velocity; In the prediction stage, state prediction is carried out through a kinematic model based on acceleration data of the IMU, and a state covariance prediction value is calculated, and the expression of the state prediction is as follows: ， wherein is a position prediction value at time k, is a position prediction value at time k-1, is a time variation, is a measured acceleration, is an accelerometer zero offset, is a speed prediction value at time k, is a speed prediction value at time k-1; The expression of the state covariance prediction value is as follows: ， ， ， wherein is a state covariance prediction, is a state transition matrix, is a state initial value, is a process noise covariance, is a position noise covariance, is a velocity noise covariance; In the update stage, the Kalman gain is calculated by using the position data of the GPS observation, the state vector and the covariance matrix are updated, and the accelerometer zero offset is dynamically corrected through residual feedback to obtain the final outdoor position estimation.

2. The transformer oil taking robot positioning method based on multi-source data fusion according to claim 1, characterized in that, The target optical flow sensor data and the target IMU data are fused based on the preset complementary filtering strategy to obtain indoor position estimation, and the method comprises the following steps: According to the complementary filtering strategy, the optical flow velocity and the gyro angular velocity are fused to obtain fused optical flow velocity, and the optical flow plane velocity is calculated according to the fused optical flow velocity and the height between the optical flow sensor and the ground, wherein the expression of the fused optical flow velocity is as follows: ， In the formula, is the fused optical flow velocity, is the weight coefficient, is the optical flow velocity, is the gyro angular velocity; The expression of the optical flow plane velocity is as follows: ， ， wherein is the X-axis direction optical flow planar velocity, is the X-axis direction fused optical flow velocity, is the Y-axis direction optical flow planar velocity, is the Y-axis direction fused optical flow velocity, is the height between the optical flow sensor and the ground; The optical flow plane velocity is integrated according to the forward Euler method to obtain a plane optical flow position estimation, and the expression is as follows: ， ， In the formula, is the position of the planar optical flow in the X-axis direction after integration, is the position of the planar optical flow in the Y-axis direction after integration, is the position of the planar optical flow in the X-axis direction, is the position of the planar optical flow in the Y-axis direction, is the amount of change in time; The position error between the optical flow position estimation and the IMU inertial navigation solution position is calculated, and the expression is as follows: ， ， ， ， wherein, is the position error resulting from the optical flow planar position in the X-axis direction and IMU accelerometer integration calculation, is the position error resulting from the optical flow planar position in the Y-axis direction and IMU accelerometer integration calculation, is the velocity error resulting from the optical flow planar velocity in the X-axis direction and IMU accelerometer integration calculation, is the velocity error resulting from the optical flow planar velocity in the Y-axis direction and IMU accelerometer integration calculation, is the planar optical flow position in the X-axis direction, is the planar optical flow position in the Y-axis direction, is the position resulting from the IMU accelerometer integration calculation in the X-axis direction, is the position resulting from the IMU accelerometer integration calculation in the Y-axis direction, is the velocity resulting from the IMU accelerometer integration calculation in the X-axis direction, is the velocity resulting from the IMU accelerometer integration calculation in the Y-axis direction; According to the position error, an acceleration correction amount, a velocity correction amount and a position correction amount are generated through a proportional-integral controller, wherein the expression of the acceleration correction amount is as follows: ， ， wherein is the X-axis direction is the X-axis direction is the Y-axis direction is the Y-axis direction is the X-axis direction is the Y-axis direction is the acceleration correction coefficient The expression of the velocity correction amount is as follows: ， ， In the formula, is a planar optical flow velocity correction amount in the X-axis direction, is a planar optical flow velocity correction amount in the Y-axis direction, is a velocity correction amount coefficient; The expression of the position correction amount is as follows: ， ， In the formula, is a position correction amount of the planar optical flow in the X-axis direction, is a position correction amount of the planar optical flow in the Y-axis direction, is a position correction amount coefficient; The velocity state and the position state are subjected to closed-loop feedback correction according to the acceleration correction amount, the velocity correction amount and the position correction amount to obtain indoor position estimation, wherein the expression of the corrected velocity state is as follows: ， ， ， ， ， ， In the formula, is a fusion speed in the X-axis direction, is an original inertial navigation system speed in the X-axis direction, is a speed change amount in the X-axis direction, is a fusion speed in the Y-axis direction, is an original inertial navigation system speed in the Y-axis direction, is a speed change amount in the Y-axis direction, is a fusion acceleration in the X-axis direction, is a fusion acceleration in the Y-axis direction, is an original inertial navigation system acceleration in the X-axis direction, is an original inertial navigation system acceleration in the Y-axis direction; The expression of the corrected position state is as follows: , , In the formula, is a fusion position in the X-axis direction, is a fusion position in the Y-axis direction, is an original inertial navigation system position in the X-axis direction, is an original inertial navigation system position in the Y-axis direction.

3. A transformer oil taking robot positioning system based on multi-source data fusion, used to implement the transformer oil taking robot positioning method based on multi-source data fusion according to any one of claims 1 to 2, characterized in that, The application relates to a transformer oil extraction robot and a position estimation method thereof. An acquisition module is configured to acquire IMU data, optical flow sensor data and GPS data of the transformer oil extraction robot, and to preprocess the IMU data, the optical flow sensor data and the GPS data to obtain target IMU data, target optical flow sensor data and target GPS data respectively; A judgment module is configured to judge a current environment scenario of the transformer oil extraction robot according to the target GPS data and the target optical flow sensor data, wherein the environment scenario includes an indoor scenario and an outdoor scenario; A first positioning module is configured to fuse the target optical flow sensor data and the target IMU data based on a preset complementary filtering strategy to obtain an indoor position estimation if the indoor scenario is present. A second positioning module is configured to fuse the target IMU data and the target GPS data based on a Kalman filtering algorithm to obtain an outdoor position estimation if the outdoor scenario is present.

4. An electronic device, comprising: comprise: at least one processor, and a memory connected with the at least one processor in communication, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1 to 2.

5. A computer-readable storage medium having stored thereon a computer program, characterized in that, The program is executed by the processor to implement the method of any one of claims 1 to 2.

Images