Automatic orientation technology

Abstract

The invention relates to the technical field of automatic orientation, and discloses an automatic orientation technology which comprises the following steps: integrating a global navigation satellite system receiver, a geomagnetic field sensor and an inertial measurement unit, and calculating three-dimensional coordinates and direction angles to obtain equipment position data; and preprocessing the equipment position data by adopting a Kalman filtering algorithm to obtain preprocessed equipment position data. According to the automatic orientation technology, by integrating a global navigation satellite system receiver, a geomagnetic field sensor and an inertial measurement unit, calculation of high-precision three-dimensional coordinates and direction angles is achieved, the accuracy of equipment position data is ensured, the equipment position data is preprocessed through a Kalman filtering algorithm, and the accuracy of the equipment position data is improved. Noise is effectively removed, data accuracy is improved, follow-up analysis is more reliable, a prediction model is constructed based on a machine learning algorithm, time sequence analysis and an environment perception technology are combined, and the pointing path of the holder can be accurately predicted.

Description

Technical Field

[0001] The present invention relates to the field of automatic orientation technology, in particular to automatic orientation technology. Background Art

[0002] Automatic orientation technology is a system that combines advanced technologies such as multi-sensor data fusion, machine learning, and intelligent beamforming. It aims to achieve intelligent planning and control of the gimbal pointing path by accurately calculating and predicting the position and direction angle of the device. This technology is widely used in communications, navigation, monitoring and other fields. It is particularly suitable for application scenarios that require high-precision positioning and dynamic adjustment. However, traditional methods rely on a single type of sensor and are easily interfered with in complex environments, resulting in reduced positioning accuracy. Traditional models are difficult to adapt to environmental changes quickly, especially in complex electromagnetic environments, which can easily lead to pointing errors or communication interruptions. At the same time, static or simple prediction models cannot accurately capture the movement trend of the device, especially in the case of nonlinear changes. Over time, the model will overfit or lose generalization ability due to insufficient data accumulation. Summary of the invention

[0003] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] Automatic orientation technology, including: integrating global navigation satellite system receivers, geomagnetic field sensors and inertial measurement units to calculate three-dimensional coordinates and direction angles to obtain device location data;

[0006] The Kalman filter algorithm is used to pre-process the device location data to obtain the pre-processed device location data;

[0007] A prediction model is built based on the machine learning algorithm and the preprocessed device location data, and the preprocessed device location data is input into the prediction model, and the prediction value of the gimbal pointing path planning is output;

[0008] Based on the predicted value of the gimbal pointing path planning, the adaptive antenna array technology is introduced, and the gimbal pointing is dynamically adjusted according to the electromagnetic environment;

[0009] Apply smart beamforming algorithms to optimize the weighting coefficients of the antenna array and generate antenna array patterns;

[0010] Continuously collect device location data, and regularly train machine learning models based on the accumulated device location data to iteratively update the device location data.

[0011] As a further solution of the present invention: the integrated global navigation satellite system receiver, the geomagnetic field sensor and the inertial measurement unit calculate the three-dimensional coordinates and direction angles to obtain the device position data, and the specific steps are as follows:

[0012] Obtain the device's geographic location information by integrating multiple global navigation satellite system receivers;

[0013] The geomagnetic field sensor detects changes in the earth's magnetic field to assist and determine the absolute azimuth of the device;

[0014] The inertial measurement unit collects the acceleration and angular velocity motion parameters of the device;

[0015] The device's geographic location information, the device's absolute azimuth, and the device's acceleration and angular velocity motion parameters are integrated and marked as device location data.

[0016] As a further solution of the present invention: the Kalman filter algorithm is used to pre-process the device location data to obtain the pre-processed device location data, and the specific steps are:

[0017] The Kalman filter algorithm is used to denoise the device location data and improve the accuracy of the data. The expression is:

[0018]

[0019] in, represents the preprocessed device position data, X(t-1) is the state estimate at the previous moment, F is the state transfer matrix, B is the control input matrix, U(t) is the control vector, H is the environment perception factor matrix, and E(t) is the real-time environmental condition;

[0020] Get the pre-processed device location data

[0021] As a further solution of the present invention: the prediction model is constructed based on the machine learning algorithm and the preprocessed device position data, and the preprocessed device position data is input into the prediction model, and the prediction value of the gimbal pointing path planning is output, and the specific steps are:

[0022] Build a prediction model based on machine learning algorithms and preprocessed device location data;

[0023] The model adopts a hybrid model that combines time series analysis and environmental perception;

[0024] The pre-processed device location data The real-time environmental conditions E(t) are input into the trained model, and the predicted value of the gimbal pointing path planning is output. The expression is:

[0025]

[0026] Among them, Y(t) represents the predicted value of the gimbal pointing path planning, It is a combination of the current device location data and a nonlinear function of the real-time environmental conditions E(t), is a time lag function, which represents the influence of the historical position data of the past τ time steps, w(τ) is a time weight function, which is used to adjust the importance of different historical moments, and β is a regulation parameter that controls the degree of influence of historical data;

[0027] The output is the predicted value Y(t) of the gimbal pointing path planning.

[0028] As a further solution of the present invention: the predicted value based on the pan-tilt pointing path planning introduces the adaptive antenna array technology, and dynamically adjusts the pan-tilt pointing according to the electromagnetic environment. The specific steps are:

[0029] Monitor the electromagnetic environment around the device and collect received signal strength indication and signal-to-noise ratio data from each antenna unit;

[0030] Establish a three-dimensional model of the current electromagnetic environment of the equipment;

[0031] A dynamic adjustment factor is introduced to adjust the gimbal pointing and the antenna array directivity diagram. The expression is:

[0032] D(t)=Y(t)+γ·A(t);

[0033] Where D(t) represents the final gimbal pointing direction, Y(t) is the predicted value of the gimbal pointing path planning obtained from the prediction model, A(t) is the dynamic adjustment factor, which represents the impact of the electromagnetic environment on the performance of the antenna array, and γ is a regulation parameter that controls the degree of influence of the electromagnetic environment.

[0034] As a further solution of the present invention: the application of the intelligent beamforming algorithm to optimize the weighted coefficients of the antenna array and generate the antenna array pattern comprises the following specific steps:

[0035] Based on the maximum ratio combining principle, the received useful signal power is maximized while minimizing the impact of noise and interference;

[0036] Define the objective function, the expression is:

[0037]

[0038] Where W is the weight vector of the antenna array, H represents the channel matrix, and R is the correlation matrix of noise plus interference. The predicted value Y(t) of the gimbal pointing path planning and the electromagnetic environment data E(t) are used as input, and the initial weight coefficient is estimated through the nonlinear mapping function h(D(t), E(t)).

[0039] Importing historical location data Using the time lag function To adjust the weighting coefficient, weighted summation is performed through the time weight function k(τ);

[0040] Set the adjustment parameter α to control the influence of historical data on the current weighting coefficient. The expression is:

[0041]

[0042] Among them, W(t) is the optimized weighting coefficient, h(D(t), E(t)) is the nonlinear mapping function, k(τ) is the time weight function, is the time lag function;

[0043] According to the optimized weighting coefficient W(t), the phase and amplitude of each antenna element are calculated;

[0044] Use the antenna array simulation tool in MATLAB to adjust the transmission parameters of each antenna unit according to the calculation results and generate the antenna array radiation pattern.

[0045] As a further solution of the present invention: the device location data is continuously collected, and the machine learning model is regularly trained based on the accumulated device location data to iteratively update the device location data. The specific steps are:

[0046] An evaluation function that comprehensively considers prediction error and model complexity is introduced to quantify the performance of the machine learning model at the current time t. The expression is:

[0047]

[0048] Among them, y i represents the actual value of the i-th sample, is the corresponding predicted value, N is the total number of samples, w j is an element in the model parameter vector, M is the number of parameters, λ is a hyperparameter that controls the strength of regularization, and σ 2 is the standard deviation of the prediction error, F(t) is the evaluation function;

[0049] According to the model evaluation function F(t), an adaptive learning rate adjustment rule is designed, which changes over time to accelerate the convergence to the local optimal solution;

[0050] Use the evaluation function to guide the training process of the machine learning model, that is, use the newly collected data set to update the model weights and dynamically adjust the learning rate according to the adaptive learning rate adjustment rule;

[0051] The model performance after each iteration is recorded to form a historical record for subsequent analysis and further optimization.

[0052] As a further solution of the present invention: according to the model evaluation function F(t), an adaptive learning rate adjustment rule is designed, which changes over time to accelerate the convergence of the local optimal solution, and the expression is:

[0053] η(t)=η0·e -γ·F(t) ;

[0054] Among them, η0 is the initial learning rate, γ is the decay coefficient, and η(t) is the adaptive learning rate adjustment rule.

[0055] The present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the automatic orientation technology described in the first aspect of the present invention is implemented.

[0056] The present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the automatic orientation technology described in the first aspect of the present invention is implemented.

[0057] Compared with the prior art, the present invention has the following beneficial effects:

[0058] By integrating global navigation satellite system receivers, geomagnetic field sensors and inertial measurement units, high-precision three-dimensional coordinate and direction angle calculations are achieved, ensuring the accuracy of device location data. The Kalman filter algorithm is used to pre-process the device location data, effectively removing noise and improving data accuracy, making subsequent analysis more reliable. A prediction model is built based on a machine learning algorithm, and combined with time series analysis and environmental perception technology, it can accurately predict the gimbal pointing path, improving the system's response speed and positioning accuracy. The adaptive antenna array technology and intelligent beamforming algorithm are introduced to dynamically adjust the gimbal pointing and optimize the antenna array weighting coefficient according to the electromagnetic environment, significantly improving the quality and stability of the communication link. The device location data is continuously collected, and the machine learning model is regularly trained to iteratively update the device location data. The evaluation function that comprehensively considers the prediction error and model complexity and the adaptive learning rate adjustment rule not only ensures the long-term stability and efficiency of the model, but also achieves continuous optimization of the device behavior pattern. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1Schematic diagram of the method flow of automatic orientation technology. DETAILED DESCRIPTION

[0060] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and understandable, the specific implementation modes of the present invention are described in detail below in conjunction with the accompanying drawings.

[0061] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0062] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0063] Example 1

[0064] See also Figure 1 , which is the first embodiment of the present invention, and which provides an automatic orientation technology, including:

[0065] S1, integrates global navigation satellite system receiver, geomagnetic field sensor and inertial measurement unit to calculate three-dimensional coordinates and direction angles to obtain device location data;

[0066] Specifically, the global navigation satellite system receiver, the geomagnetic field sensor and the inertial measurement unit are integrated to calculate the three-dimensional coordinates and direction angles to obtain the device location data. The specific steps are as follows:

[0067] Obtain the device's geographic location information by integrating multiple global navigation satellite system receivers;

[0068] The geomagnetic field sensor detects changes in the earth's magnetic field to assist and determine the absolute azimuth of the device;

[0069] The inertial measurement unit collects the acceleration and angular velocity motion parameters of the device;

[0070] The device's geographic location information, the device's absolute azimuth, and the device's acceleration and angular velocity motion parameters are integrated and marked as device location data.

[0071] It should be noted that by integrating multiple global navigation satellite system receivers, geomagnetic field sensors, and inertial measurement units, high-precision three-dimensional coordinate and direction angle calculations can be achieved;

[0072] GNSS provides precise geographic location information, the geomagnetic field sensor assists in determining the absolute azimuth, and the IMU is responsible for capturing the dynamic motion parameters of the device. The fusion of data not only improves the accuracy of positioning, but also enhances the robustness and adaptability of the system.

[0073] Data from all sensors are collected at a high frequency and processed in real time, ensuring the timeliness and reliability of device location information.

[0074] S2. Preprocess the device location data using a Kalman filter algorithm to obtain preprocessed device location data;

[0075] Specifically, the Kalman filter algorithm is used to preprocess the device location data to obtain the preprocessed device location data. The specific steps are: the Kalman filter algorithm is used to denoise the device location data and improve the accuracy of the data. The expression is:

[0076]

[0077] in, represents the preprocessed device location data, X(t-1) is the state estimate at the previous moment, F is the state transfer matrix, B is the control input matrix, U(t) is the control vector, H is the environment perception factor matrix, and E(t) is the real-time environmental condition; the preprocessed device location data is obtained

[0078] It should be noted that the Kalman filter effectively filters out noise interference through a recursive estimation method, making the output position data more stable and smooth;

[0079] The algorithm combines the state transition matrix, control input matrix, and environmental perception factor matrix to achieve accurate prediction and real-time update of the device position change trend;

[0080] Based on real-time environmental conditions, the system can dynamically adjust filtering parameters to ensure optimal performance in different environments.

[0081] S3, construct a prediction model based on the machine learning algorithm and the preprocessed device location data, input the preprocessed device location data into the prediction model, and output the prediction value of the gimbal pointing path planning; specifically, construct a prediction model based on the machine learning algorithm and the preprocessed device location data, input the preprocessed device location data into the prediction model, and output the prediction value of the gimbal pointing path planning, the specific steps are: construct a prediction model based on the machine learning algorithm and the preprocessed device location data; the model adopts a hybrid model combining time series analysis and environmental perception; the preprocessed device location data The real-time environmental conditions E(t) are input into the trained model, and the predicted value of the gimbal pointing path planning is output. The expression is:

[0082]

[0083] Among them, Y(t) represents the predicted value of the gimbal pointing path planning, It is a combination of the current device location data and a nonlinear function of the real-time environmental conditions E(t), is a time lag function, which represents the influence of the historical position data of the past τ time steps. w(τ) is a time weight function, which is used to adjust the importance of different historical moments. β is an adjustment parameter, which controls the degree of influence of historical data. The output is the predicted value Y(t) of the gimbal pointing path planning.

[0084] It should be noted that the model not only considers the impact of historical data in time series analysis, but also takes into account real-time environmental conditions, thereby improving the accuracy and adaptability of the prediction;

[0085] function The introduction of allows the model to better capture nonlinear relationships in complex environments and enhances predictive capabilities;

[0086] By continuously inputting new data into the trained model, the system can continuously optimize itself and ensure long-term stable high-precision performance.

[0087] S4, based on the predicted value of the gimbal pointing path planning, introduce the adaptive antenna array technology and dynamically adjust the gimbal pointing according to the electromagnetic environment;

[0088] Specifically, based on the predicted value of the gimbal pointing path planning, the adaptive antenna array technology is introduced, and the gimbal pointing is dynamically adjusted according to the electromagnetic environment. The specific steps are: monitor the electromagnetic environment around the device and collect the received signal strength indication and signal-to-noise ratio data from each antenna unit; establish a three-dimensional model of the current electromagnetic environment of the device; introduce a dynamic adjustment factor to adjust the gimbal pointing and the antenna array directivity diagram, and the expression is:

[0089] D(t)=Y(t)+γ·A(t);

[0090] Where D(t) represents the final gimbal pointing direction, Y(t) is the predicted value of the gimbal pointing path planning obtained from the prediction model, A(t) is the dynamic adjustment factor, which represents the impact of the electromagnetic environment on the performance of the antenna array, and γ is a regulation parameter that controls the degree of influence of the electromagnetic environment.

[0091] It should be noted that the combination of adaptive antenna array technology and electromagnetic environment monitoring enables intelligent adjustment of the PTZ pointing;

[0092] Through real-time monitoring of the surrounding electromagnetic environment, a three-dimensional model is established, providing a scientific basis for adjusting the antenna array's directional pattern;

[0093] The dynamic adjustment factor A(t) in the formula D(t)=Y(t)+γ·A(t) reflects the influence of the electromagnetic environment, while the adjustment parameter γ is used to flexibly control the degree of this influence, ensuring the flexibility and adaptability of the system.

[0094] S5. Applying the intelligent beamforming algorithm to optimize the weighting coefficients of the antenna array and generate the antenna array pattern;

[0095] Specifically, the intelligent beamforming algorithm is applied to optimize the weighting coefficients of the antenna array and generate the antenna array pattern. The specific steps are: based on the maximum ratio combining principle, the received useful signal power is maximized while minimizing the noise and interference effects; the objective function is defined, and the expression is:

[0096]

[0097] Where W is the weighted vector of the antenna array, H represents the channel matrix, and R is the correlation matrix of noise plus interference; the predicted value Y(t) of the gimbal pointing path planning and the electromagnetic environment data E(t) are used as input, and the initial weighting coefficient is estimated through the nonlinear mapping function h(D(t), E(t)); historical position data is introduced Using the time lag function To adjust the weighting coefficient, weighted sum is performed through the time weight function k(τ); set the adjustment parameter α to control the influence of historical data on the current weighting coefficient, the expression is:

[0098]

[0099] Among them, W(t) is the optimized weighting coefficient, h(D(t), E(t)) is the nonlinear mapping function, k(τ) is the time weight function, is a time lag function; according to the optimized weighting coefficient W(t), the phase and amplitude of each antenna unit are calculated; using the antenna array simulation tool in MATLAB, the transmission parameters of each antenna unit are adjusted according to the calculation results, and the antenna array radiation pattern is generated.

[0100] It should be noted that the application of smart beamforming algorithms significantly improves the quality of the communication link;

[0101] The objective function based on the maximum ratio combining principle ensures the maximization of the received useful signal and the minimization of noise and interference;

[0102] Through the nonlinear mapping function h(D(t),E(t)) and the time lag function The system can make full use of historical data while maintaining high sensitivity to current environmental changes;

[0103] The resulting antenna array pattern is carefully designed to ensure that the main beam points in the optimal direction, while reducing the sidelobe level and improving communication quality.

[0104] S6. Continuously collect device location data, and regularly train the machine learning model based on the accumulated device location data to iteratively update the device location data;

[0105] Specifically, the device location data is continuously collected, and the machine learning model is regularly trained based on the accumulated device location data to iteratively update the device location data. The specific steps are: introducing an evaluation function that comprehensively considers the prediction error and model complexity to quantify the performance of the machine learning model at the current time t. The expression is:

[0106]

[0107] Among them, y i represents the actual value of the i-th sample, is the corresponding predicted value, N is the total number of samples, w j is an element in the model parameter vector, M is the number of parameters, λ is a hyperparameter that controls the strength of regularization, and σ 2 is the standard deviation of the prediction error, F(t) is the evaluation function; according to the model evaluation function F(t), an adaptive learning rate adjustment rule is designed, which changes over time to accelerate the convergence to the local optimal solution; the evaluation function is used to guide the training process of the machine learning model, that is, the model weights are updated using the newly collected data set, and the learning rate is dynamically adjusted according to the adaptive learning rate adjustment rule; the model performance after each iteration is recorded to form a historical record for subsequent analysis and further optimization; according to the model evaluation function F(t), an adaptive learning rate adjustment rule is designed, which changes over time to accelerate the convergence to the local optimal solution, and the expression is:

[0108] η(t)=η0·e -γ·F(t) ;

[0109] Among them, η0 is the initial learning rate, γ is the decay coefficient, and η(t) is the adaptive learning rate adjustment rule.

[0110] It should be noted that effective management and optimization of machine learning models are achieved through the evaluation function that comprehensively considers prediction error and model complexity;

[0111] Formula η(t)=η0·e -γ·F(t) Dynamically adjust the learning rate to accelerate the convergence of the model and improve the ability to escape from the local optimal solution;

[0112] By regularly collecting new data and updating model weights, the system forms a closed-loop feedback mechanism, ensuring long-term stability and efficiency, while also providing a solid foundation for subsequent technical improvements.

[0113] This embodiment also provides a computer device, which is suitable for the case of automatic orientation technology, including: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions to implement the automatic orientation technology proposed in the above embodiment.

[0114] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0115] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the automatic orientation technology proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable ReadOnly Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0116] In summary, by integrating the global navigation satellite system receiver, the geomagnetic field sensor and the inertial measurement unit, the calculation of high-precision three-dimensional coordinates and direction angles is realized, ensuring the accuracy of the device location data. The Kalman filter algorithm is used to pre-process the device location data, which effectively removes noise and improves data accuracy, making subsequent analysis more reliable. A prediction model is built based on the machine learning algorithm, and combined with time series analysis and environmental perception technology, it can accurately predict the gimbal pointing path, improve the response speed and positioning accuracy of the system, introduce adaptive antenna array technology and intelligent beamforming algorithm, dynamically adjust the gimbal pointing and optimize the antenna array weighting coefficient according to the electromagnetic environment, significantly improve the quality and stability of the communication link, continuously collect device location data, and regularly train the machine learning model to iteratively update the device location data. By comprehensively considering the evaluation function of the prediction error and model complexity and the adaptive learning rate adjustment rule, not only the long-term stability and efficiency of the model are guaranteed, but also the continuous optimization of the device behavior pattern is achieved.

[0117] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. Automatic orientation technology, characterized by: include: Integrate global navigation satellite system receiver, geomagnetic field sensor and inertial measurement unit to calculate three-dimensional coordinates and direction angles to obtain device location data; The Kalman filter algorithm is used to pre-process the device location data to obtain the pre-processed device location data; A prediction model is built based on the machine learning algorithm and the preprocessed device location data, and the preprocessed device location data is input into the prediction model, and the prediction value of the gimbal pointing path planning is output; Based on the predicted value of the gimbal pointing path planning, the adaptive antenna array technology is introduced, and the gimbal pointing is dynamically adjusted according to the electromagnetic environment; Apply smart beamforming algorithms to optimize the weighting coefficients of the antenna array and generate antenna array patterns; Continuously collect device location data, and regularly train machine learning models based on the accumulated device location data to iteratively update the device location data.

2. The automatic orientation technology according to claim 1, characterized in that: The integrated global navigation satellite system receiver, geomagnetic field sensor and inertial measurement unit calculate the three-dimensional coordinates and direction angles to obtain the device location data, and the specific steps are as follows: Obtain the device's geographic location information by integrating multiple global navigation satellite system receivers; The geomagnetic field sensor detects changes in the earth's magnetic field to assist and determine the absolute azimuth of the device; The inertial measurement unit collects the acceleration and angular velocity motion parameters of the device; The device's geographic location information, the device's absolute azimuth, and the device's acceleration and angular velocity motion parameters are integrated and marked as device location data.

3. The automatic orientation technology according to claim 1, characterized in that: The Kalman filter algorithm is used to preprocess the device location data to obtain the preprocessed device location data. The specific steps are: The Kalman filter algorithm is used to denoise the device location data and improve the accuracy of the data. The expression is: ; in, Represents the pre-processed device location data. is the estimated value of the state at the previous moment, is the state transition matrix, is the control input matrix, is the control vector, is the environment perception factor matrix, is the real-time environmental condition; Get the pre-processed device location data .

4. The automatic orientation technology according to claim 1, characterized in that: The prediction model is constructed based on the machine learning algorithm and the preprocessed device position data, and the preprocessed device position data is input into the prediction model, and the prediction value of the gimbal pointing path planning is output. The specific steps are: Build a prediction model based on machine learning algorithms and preprocessed device location data; The model adopts a hybrid model that combines time series analysis and environmental perception; The pre-processed device location data and real-time environmental conditions Input into the trained model and output the predicted value of the gimbal pointing path planning, the expression is: ; in, Indicates the predicted value of the gimbal pointing path planning, It is a combination of the current device location data and real-time environmental conditions A nonlinear function of is a time lag function, representing the past The impact of historical position data of time steps, is a time weight function, which is used to adjust the importance of different historical moments. It is a tuning parameter that controls the degree of influence of historical data; Output the predicted value of the gimbal pointing path planning .

5. The automatic orientation technology according to claim 1, characterized in that: The predicted value based on the gimbal pointing path planning introduces the adaptive antenna array technology and dynamically adjusts the gimbal pointing according to the electromagnetic environment. The specific steps are as follows: Monitor the electromagnetic environment around the device and collect received signal strength indication and signal-to-noise ratio data from each antenna unit; Establish a three-dimensional model of the current electromagnetic environment of the equipment; A dynamic adjustment factor is introduced to adjust the gimbal pointing and the antenna array directivity diagram. The expression is: ； in, Indicates the final gimbal pointing direction. is the predicted value of the gimbal pointing path planning obtained from the prediction model, is a dynamic adjustment factor, which indicates the impact of the electromagnetic environment on the performance of the antenna array. It is an adjustment parameter that controls the degree of influence of the electromagnetic environment.

6. The automatic orientation technology according to claim 1, characterized in that: The application of the intelligent beamforming algorithm to optimize the weighted coefficients of the antenna array and generate the antenna array pattern comprises the following specific steps: Based on the maximum ratio combining principle, the received useful signal power is maximized while minimizing the impact of noise and interference; Define the objective function, the expression is: ; in is the weight vector of the antenna array, represents the channel matrix, is the correlation matrix of noise plus interference; Point the gimbal to the predicted path plan value and electromagnetic environment data As input, through a nonlinear mapping function Estimate initial weighting coefficients; Importing historical location data , using the time lag function To adjust the weighting coefficient, through the time weight function Perform weighted summation; Setting the tuning parameters , controls the influence of historical data on the current weighting coefficient, the expression is: ; in, is the optimized weighting coefficient, is a nonlinear mapping function, is the time weight function, is the time lag function; According to the optimized weighting coefficient , calculate the phase and amplitude of each antenna element; Use the antenna array simulation tool in MATLAB to adjust the transmission parameters of each antenna unit according to the calculation results and generate the antenna array radiation pattern.

7. The automatic orientation technology according to claim 1, characterized in that: The device location data is continuously collected, and the machine learning model is regularly trained based on the accumulated device location data to iteratively update the device location data. The specific steps are as follows: Introduce an evaluation function that comprehensively considers prediction error and model complexity to quantify the current moment The performance of the machine learning model is expressed as: ; in, Indicates The actual value of the samples, is the corresponding predicted value, is the total number of samples, are the elements in the model parameter vector, is the number of parameters, is a hyperparameter that controls the strength of regularization, is the standard deviation of the prediction error, Evaluation function; According to the model evaluation function , design an adaptive learning rate adjustment rule that changes over time to accelerate convergence to the local optimal solution; Use the evaluation function to guide the training process of the machine learning model, that is, use the newly collected data set to update the model weights and dynamically adjust the learning rate according to the adaptive learning rate adjustment rule; The model performance after each iteration is recorded to form a historical record for subsequent analysis and further optimization.

8. The automatic orientation technology according to claim 1, characterized in that: According to the model evaluation function , design an adaptive learning rate adjustment rule that changes over time to accelerate convergence to the local optimal solution, expressed as: in, is the initial learning rate, is the attenuation coefficient, Adaptive learning rate adjustment rule.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the automatic orientation technology described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the automatic orientation technology described in any one of claims 1 to 7 are implemented.

Images