IMU-PID-RTK fused crawler conveyor control method and system

Abstract

The invention discloses an IMU-PID-RTK fused crawler conveyor control method and system. The method comprises the following steps: remotely inputting a target trajectory shape and analyzing to obtain a curvature and a step length; decomposing the target trajectory into equidistant scatter points according to the curvature and the step length, and calculating a target coordinate and a target course angle; acquiring corresponding current state information in real time through the IMU and the RTK; calculating position deviation and course angle deviation according to the input target track shape and the current state information; according to the position deviation and the correction amount of the calculated course angle, generating a course angle total deviation, and finally outputting the motor speed correction amount; and controlling the crawler chassis according to the motor speed correction to enable the crawler chassis to walk according to the target track. The system comprises a remote input unit, a data acquisition unit, a control unit and an execution unit. According to the invention, accurate control of the arc running state of the crawler conveyor in a complex environment is realized. The method can be widely applied to the field of positioning control.

Description

Technical Field

[0001] The present invention relates to the field of positioning control, and in particular to an IMU-PID-RTK fused crawler conveyor control method and system. Background Art

[0002] With the rapid development of industrial automation and robotics technology, the demand for precise control in complex environments is increasing. As an important automation equipment, crawler conveyors are widely used in industry, agriculture, military and other fields. Although the traditional crawler conveyor control system can meet the basic motion control needs, its limitations gradually emerge when facing the changing terrain, high-precision positioning requirements and dynamic environmental interference, especially when performing arc driving or path tracking tasks. How to improve control accuracy and stability has become an important issue in technological development.

[0003] In the prior art, most of the control methods of crawler conveyors rely on PLC programmable controllers or PID controllers as core control units, combined with GPS or RTK systems for position tracking, and use IMU to obtain attitude data to achieve basic path tracking. However, the existing technical solutions still have significant deficiencies, especially in control accuracy and real-time response capabilities in complex environments. For example, the positioning accuracy of GPS or RTK systems in signal-restricted areas such as cities and forests may drop significantly, and although the IMU system can provide vehicle body attitude information, it is easily affected by external disturbances, resulting in error accumulation and affecting the stability of control. In addition, the traditional PID control algorithm cannot dynamically adapt to the changes in the operating state of the conveyor, especially in the case of load changes, terrain fluctuations, etc., the control accuracy and response speed are poor. In order to solve these problems, there is an urgent need for a new control method that deeply integrates IMU, PID control and RTK positioning systems to improve the accuracy and stability of arc driving of crawler conveyors in complex environments and meet the high requirements of modern applications for precise control. Summary of the invention

[0004] In view of this, in order to solve the problem that the existing crawler conveyor control method cannot realize the precise control of the arc driving state of the crawler conveyor in a complex environment, the present invention proposes an IMU-PID-RTK fused crawler conveyor control method, the method comprising the following steps:

[0005] Remotely input the target trajectory shape and analyze it to obtain the curvature and step length;

[0006] Decompose the target trajectory into equidistant scattered points according to the curvature and step size, and calculate the target coordinates and target heading angle;

[0007] During the driving process of the crawler conveyor, the corresponding current status information is obtained in real time through IMU and RTK, including position coordinates and heading angle;

[0008] Calculate the position deviation and heading angle deviation based on the input target trajectory shape and current state information;

[0009] According to the position deviation and the calculated heading angle correction, the total heading angle deviation is generated, and finally the motor speed correction is output;

[0010] The crawler chassis is controlled according to the motor speed correction amount so that it moves along the target trajectory.

[0011] In some embodiments, it also includes:

[0012] Unscented Kalman filter (UKF) is used to optimize IMU data and output high-precision attitude angle information.

[0013] In some embodiments, it also includes:

[0014] The dynamic weight adjustment mechanism is used to coordinate the relative weights of the two control loops, combined with adaptive PID parameter optimization to improve the robustness and dynamic response capability of system control.

[0015] In some embodiments, it also includes:

[0016] Build a prediction model based on LSTM;

[0017] When the RTK signal is limited, the model uses IMU sensor data and historical heading angle deviation information to predict the total heading angle deviation and dynamically optimize PID parameters, thereby achieving high-precision trajectory tracking control.

[0018] Based on the above scheme, the present invention provides a crawler conveyor control method and system with IMU-PID-RTK fusion, which uses the attitude angle data obtained by IMU as the core input of PID control, replacing the direct processing of the acceleration and angular velocity of the conveyor in the traditional method, so as to obtain the heading angle of the conveyor more accurately and in real time. By combining the IMU attitude angle data with the cascaded PID controller and the dynamic PID parameter adjustment strategy, the controller can adaptively optimize the control parameters according to the real-time state of the crawler conveyor to adapt to different operating conditions and achieve high-precision arc driving trajectory control; in addition, the present invention introduces the RTK high-precision positioning system, which works with the IMU data and uses the real-time three-dimensional position information and heading angle deviation provided by RTK to correct the attitude data of the IMU. By fusing the data of the two, the system can dynamically adjust the trajectory correction amount and control parameters during the actual driving process, further improving the trajectory tracking accuracy. On this basis, the present invention combines machine learning technology, optimizes the trajectory control model by training the historical trajectory data and control error of the crawler conveyor, and maintains a high trajectory control accuracy through IMU data and the model parameters obtained by learning even when the RTK signal is limited. Ultimately, the crawler conveyor can travel smoothly along the preset target arc trajectory, achieving high precision and high reliability of trajectory control. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a flow chart of steps of a crawler conveyor control method of IMU-PID-RTK fusion of the present invention;

[0020] Figure 2 It is a data flow diagram of the PID cascade control of the present invention;

[0021] Figure 3 It is a schematic diagram of the hardware architecture of the present invention. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0023] It should be noted that, for the convenience of description, only the parts related to the relevant invention are shown in the drawings. In the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other.

[0024] It should be understood that the "system", "device", "unit" and / or "module" used in this application is a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the word can be replaced by other expressions.

[0025] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "a kind" and / or "the" do not refer to the singular, but also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of clearly identified steps and elements, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements. The elements defined by the sentence "includes a..." do not exclude the existence of other identical elements in the process, method, commodity or device that includes the elements.

[0026] In addition, flow charts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, the various steps may be processed in reverse order or simultaneously. At the same time, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0027] Reference Figure 1 , which is a flow chart of an optional example of the crawler conveyor control method of IMU-PID-RTK fusion proposed in the present invention. The method can be applied to computer equipment. The control method proposed in this embodiment may include but is not limited to the following steps:

[0028] Step S1, generating a target trajectory in response to an input operation;

[0029] Step S2, obtaining current status information through IMU and RTK;

[0030] Step S3, calculating the position deviation and heading angle deviation according to the target trajectory and current state information, and generating control instructions through cascade PID;

[0031] Step S4: Execute actions according to the control instructions.

[0032] By remotely inputting the target trajectory shape, Jetson Nano, as the core control module, receives data from the data acquisition module, processes and calculates the position deviation and heading angle deviation in real time, and inputs it into the cascade PID controller to generate speed difference adjustment instructions for the motors on both sides of the track to ensure that the conveyor moves accurately along the set trajectory.

[0033] The present invention integrates the IMU-PID-RTK crawler conveyor precise arc control method, and proposes a control scheme that can correct path deviation in real time and achieve high-precision arc driving by deeply integrating the IMU attitude angle data, PID controller and RTK high-precision positioning data. The system makes full use of the attitude angle information provided by the IMU to achieve accurate measurement of the heading angle, and dynamically corrects the trajectory of the crawler conveyor in combination with the real-time three-dimensional position information provided by RTK. Through multi-sensor fusion technology, the system significantly improves the real-time and robustness of the control, and can ensure that the crawler conveyor runs smoothly along the predetermined arc trajectory even in complex terrain and dynamic environments.

[0034] In some feasible embodiments, step S1 specifically includes:

[0035] In order to realize the arbitrary arc trajectory of the crawler conveyor, the system uses parameterized expression to input trajectory information. The user inputs the geometric shape of the target arc (such as circular arc, elliptical arc, etc.) through remote communication, and Jetson Nano parses the corresponding curvature k and step length Δs according to the geometric characteristics of the input shape. The initial coordinates (x0, y0) and the initial heading angle θ0 are obtained in real time by the IMU and RTK system to determine the starting point position and initial direction of the trajectory.

[0036] According to the generated curvature k and step length Δs, the system decomposes the target arc into a series of equidistant discrete points and calculates the spatial coordinates and heading angle of each point. The specific steps are as follows:

[0037] Discrete point decomposition: Assume that the target trajectory is decomposed into N discrete points, and the coordinates of each point are (x i ,y i ), the heading angle is θ i , for the case of constant curvature, the recursive formula for discrete points is as follows:

[0038] x i+1 =x i +Δs·cosθ i

[0039] y i+1 =y i +Δs·sinθ i

[0040] θ i+1 =θ i +k·Δs

[0041] Among them, (x0, y0) and θ0 are the initial states of the starting point of the trajectory, k is the target curvature, and Δs is the step size.

[0042] For the case where the curvature changes with the trajectory, the coordinate changes of the discrete points need to reflect the latest heading angle update results, so the discrete point generation formula is adjusted to:

[0043] θ i+1 =θ i +k(s i )·Δs

[0044] x i+1 =x i +Δs·Cosθ i+1

[0045] y i+1 =y i +Δs·sinθ i+1

[0046] Among them, k(s i ) is the curvature function at point s i To improve the calculation accuracy, the curvature derivative is further introduced

[0047]

[0048] Dynamic trajectory adjustment: To support real-time tracking and correction, the system calculates the deviation between the current position of the transport aircraft and the target point at each discrete point, including position deviation and heading angle deviation:

[0049] Δx=x i -x t , Δy=y i -y t , Δθ=θ i -θ t

[0050] Among them, (x t ,y t ) and θ t are the coordinates and heading angle of the real-time target point. By calculating the above deviations, the system can dynamically adjust the generation and switching of target trajectory points during driving. The system saves the generated discrete points and related attributes as a parallel data structure (such as a pointer array). Each discrete point records the following information to support real-time query, dynamic adjustment, and target switching:

[0051] P i ={x i ,y i ,θ i , Δx, Δy, Δθ}

[0052] By calling the above data in real time, the system can accurately track the current trajectory points and dynamically predict the subsequent trajectory points.

[0053] When no stop signal is received, the crawler conveyor continues to move along the current track. The specific logic is as follows:

[0054] Dynamic update of discrete points: When the transport aircraft is near the end of the current trajectory, the system generates subsequent trajectory segments according to the set trajectory shape and parameters; the starting point of each trajectory segment is the end point of the previous segment, and the direction is the heading angle at the end point. For trajectories with continuously changing curvature, the system updates the value of k(s) in real time and calculates the subsequent trajectory segments through the discrete point generation formula.

[0055] Circular trajectory calculation: If the arc trajectory is a closed curve, the system will automatically detect the loop point and calculate the subsequent trajectory to achieve continuous driving; if a new trajectory shape is entered, it will switch to the new trajectory.

[0056] Trajectory interruption and resumption: When receiving a stop command, the system immediately stops calculating the subsequent trajectory, and the conveyor slows down and maintains the current state. If it is necessary to resume driving, the system regenerates the trajectory from the current position and continues driving.

[0057] In some feasible embodiments, step S2 specifically includes:

[0058] The main data collected by IMU are as follows:

[0059] Gyroscope dataω x ,ω y ,ω z : Measure the angular velocity of the transport aircraft around each axis; accelerometer data a x , a y , a z :Measure the acceleration of the conveyor in each axis direction; magnetometer data m x , m y , m z : Measure the magnetic field strength of the transport aircraft in each axis direction; output attitude angles (φ, ψ, θ): pitch angle, roll angle and heading angle respectively.

[0060] The RTK module provides the current position coordinates (x r ,y r ).

[0061] In some feasible embodiments, step S2 further includes:

[0062] IMU is the main data source for the real-time heading angle of transport aircraft, but it is easily affected by noise and drift. Therefore, the unscented Kalman filter (UKF) is used to optimize the IMU data and output high-precision attitude angle information.

[0063] Through UKF state estimation and optimization, the IMU raw data is processed into the state vector x in the following form, where T represents the matrix transposition:

[0064] x=[φ,ψ,θ,ω x ,ω y ,ω z ] T

[0065] Input variable u k :

[0066] u k =[ω x ,ω y ,ω z ] T

[0067] The state transfer equation is established through the relationship between Euler angle and angular velocity:

[0068]

[0069] in, represents the derivative form of φ, ψ, and θ. The equation is discretized by numerical integration method to obtain the nonlinear state transfer function f(x k ,u k ), the state transfer function describes how the state of the system evolves from the current time k to the next time k+1 through the laws of system dynamics.

[0070] x k+1 =x k +Δt·f(x k ,u k )+w k

[0071]

[0072] Among them, w k is the process noise, assumed to be Gaussian distributed Q is the process noise covariance matrix.

[0073] The observation equations of the accelerometer and magnetometer are:

[0074]

[0075] m x =B x cosθ+B y sinθ,m y =-B x sinθ+B y cosθ,m z =B z

[0076] Among them, z k is the actual observation value of IMU, h(xk ) is the observation function, v k is the observation noise, assumed to follow a Gaussian distribution v k ~N(0,R), R is the observation noise covariance matrix, g is the gravitational acceleration constant, B x , B y , B z is the vector component of the Earth's magnetic field and depends on its geographic location and direction.

[0077] Use unscented transformation instead of linearization to propagate the system state and covariance through a set of specific sampling points (Sigma points). and covariance P k Generate a set of 2L+1 Sigma points:

[0078]

[0079] Where L is the state dimension, λ = α 2 (L+κ)-L, λ is the adjustment parameter, α controls the distribution range of Sigma points, and κ is a constant. k It is a Sigma point set, which is used to capture the nonlinear changes of state distribution. According to the state transfer equation, the state mean and covariance at the next moment are predicted:

[0080]

[0081] in, It is the Sigma point predicted at the next moment k+1 based on the current moment k. is the state mean of the next moment k+1 predicted based on the current moment k, P k+1|k It is the covariance predicted at the next moment k+1 based on the current moment k. is the weight used for state mean calculation, is the weight used for covariance calculation, defined as different values ​​to enhance the filter's independent control of the state mean and covariance. According to the observation function, update the state estimate and covariance:

[0082]

[0083] in, is the predicted value of the i-th Sigma point in the observation space, is the predicted observation mean. Furthermore, the observation covariance matrix P zz and the joint covariance P of the state and observation xz for:

[0084]

[0085] Get the Kalman gain K:

[0086]

[0087] The system state and covariance update methods are as follows:

[0088]

[0089] P k+1 =P k+1|k -KP zz K T

[0090] Through the above steps, the optimized attitude angle data output by UKF is:

[0091] x k+1 =[θ k+1 ,φ k+1 , ψ k+1 ,ω x，k+1 ,ω y，k+1 ,ω z，k+1 ] T

[0092] The heading angle θ k+1 That is, θ r , as the core input of the subsequent heading angle PID controller, it provides high-quality attitude data support for the precise arc control of the crawler transporter.

[0093] In some feasible embodiments, steps S3 and S4 specifically include:

[0094] In order to achieve accurate driving of the crawler conveyor along any arc trajectory, this system designs a multi-level dynamic optimization control logic based on cascade control. The control architecture consists of a position PID control loop and a heading angle PID control loop, and accurate trajectory tracking is achieved through the synergy of the outer position control and the inner heading control. The dynamic weight adjustment mechanism is used to coordinate the relative weights of the two control loops, combined with adaptive PID parameter optimization to improve the robustness and dynamic response capability of the system control.

[0095] The outer position PID loop is based on the current position of the crawler conveyor (x r ,y r ) and the target point (x t ,y t ), calculate the position deviation (Δx, Δy), and output the target heading angle θ t The correction value Δθ p The calculation formula is:

[0096]

[0097] Where Δr represents the position deviation, are the proportional, integral and differential gain parameters of the position PID controller respectively. In order to balance the control weight between the position deviation and the heading angle deviation, the system introduces a dynamic weight adjustment mechanism to calculate the total heading angle deviation Δθ based on the position deviation Δr. total :

[0098] Δθ total =Δθ+W p ·Δθ p

[0099] Where Δθ is the target heading angle θ t and the actual heading angle θ r Deviation of W p is the dynamic weight of the position PID output, defined as:

[0100]

[0101] Among them, the threshold δ p Used to distinguish between larger and smaller deviations (for example, set to 0.5m). When the position deviation is large, the position PID loop takes the lead in the correction to quickly correct the path deviation; when the position deviation is small, the heading angle PID loop takes over the lead to enhance the stability of the heading angle control.

[0102] The inner heading angle PID control loop is based on the total heading angle deviation Δθ total As input, the motor speed correction Δv is output, thereby realizing the differential adjustment of the left and right tracks. The control formula is:

[0103]

[0104] in, They are the proportional, integral and differential gain parameters of the heading angle PID. The final motor speed adjustment formula is as follows:

[0105] v left =v base -Δv,v right =v base +Δv

[0106] Among them, v base is the reference speed, and Δv is used to dynamically adjust the speed difference on both sides of the track to achieve precise control of the heading angle.

[0107] In order to meet the requirements of different trajectory curvature k and heading angle deviation Δθ on control accuracy, the system introduces an adaptive parameter optimization mechanism for the inner heading angle PID controller. Heading angle PID parameters and Dynamic adjustment with curvature k:

[0108]

[0109] in, and is the initial heading angle PID parameter, α and β are adjustment coefficients. When the curvature k of the target trajectory is large (i.e., the curve radius is small) or the heading angle deviation Δθ is large, increase Improve the system's response speed and enhance the ability to quickly correct deviations; when the trajectory is relatively straight or the heading angle deviation is small, reduce Increase Reduce system oscillation and overshoot and maintain the stability of the heading angle.

[0110] Among them, the heading angle deviation refers to the angle difference between the current heading angle and the next discrete point of the trajectory calculated; the total heading angle deviation refers to the sum of the heading angle deviation correction output by the coordinate deviation PID and the current heading angle deviation.

[0111] The overall process diagram of the PID control is shown in Figure 1. Figure 2 .

[0112] The present invention integrates IMU-PID-RTK to realize precise tracking of the arc trajectory by introducing an adaptive PID controller to dynamically adjust the speed difference of the motors on both sides of the track. The core technology is based on the heading angle information fed back by the IMU in real time and the high-precision coordinates provided by the RTK positioning system. By monitoring and correcting the coordinate deviation and heading angle deviation, the PID control parameters are dynamically adjusted to ensure that the crawler conveyor strictly follows the target trajectory. In addition, this method combines the advantages of IMU and RTK data, overcomes the trajectory deviation problem caused by a single sensor error in traditional methods, and significantly improves the accuracy and response speed of path tracking through the dynamic adjustment mechanism of PID parameters. It is suitable for high-precision trajectory control requirements in complex road conditions and dynamic environments.

[0113] In some feasible embodiments, it also includes:

[0114] In order to ensure the accurate driving of the crawler conveyor along any arc trajectory when the RTK signal is limited (if the positioning accuracy drift exceeds a certain threshold, which is set here as drift exceeding 5cm and lasting for more than 3 seconds, the RTK signal is judged to be unstable or lost.), this system trains a machine learning model based on the long short-term memory network (LSTM) through historical trajectory data to construct a nonlinear mapping relationship between the heading angle PID parameters, heading angle deviation and RTK data. The model uses IMU sensor data and historical heading angle deviation information to predict the total heading angle deviation Δθ total And dynamically optimized PID parameters to achieve high-precision trajectory tracking control.

[0115] In the multiple arc trajectory experiments of the crawler conveyor, the following key variables are recorded and time series data is formed:

[0116] Target trajectory shape and heading angle: Target position coordinates (x t ,y t ), calculated by JetsonNano based on the parameterized arc trajectory; target heading angle θ t , through the target position and actual position (x r ,y r )calculate:

[0117]

[0118] RTK data: actual position (x r ,y r ), provided by RTK in real time; Position deviation (Δx, Δy):

[0119] Δx=x t -x r , Δy=y t -y t ,

[0120] Heading angle data: actual heading angle θ r , provided by IMU in real time; total heading angle deviation Δθ total :

[0121] Δθ total =Δθ+W p ·Δθ p

[0122] Where Δθ=θ t -θ r is the deviation between the target heading angle and the actual heading angle; Δθ p W is the heading angle correction of the position PID; p is the dynamic weight of the location PID.

[0123] IMU data: three-axis angular velocity (ω x ,ω y ,ω z ), three-axis acceleration (a x , a y , a z );PID parameters: heading angle PID parameters And the adjustment values ​​under different heading angle deviation conditions.

[0124] In the data preprocessing stage, the input and output features of each time step are defined as follows:

[0125] Input features:

[0126] Among them, Δr t is the deviation between the current position and the target position; Δθ total，t is the total heading angle deviation.

[0127] Output features: total heading angle deviation prediction Δθ total，t+1 ;PID parameter prediction,

[0128] In order to model the complex dynamic relationship between the above features, the LSTM structure is used to train the time series data. The network architecture is as follows:

[0129] Input layer: accepts time series input X t ; LSTM layer: stack multiple layers of LSTM units to capture the temporal dynamic characteristics and long-term dependencies of the input features; fully connected layer: map the output features of LSTM to the target output space; output layer: output the total deviation prediction Δθ of the heading angle respectively total，t+1 and PID parameters

[0130] In order to optimize the heading angle deviation prediction and PID parameter adjustment at the same time, a composite loss function is designed:

[0131] L=w1·L Δθ +w2·L PID

[0132] Among them, the mean square error of the heading angle deviation is L Δθ And the mean square error L of PID parameters PID As shown below, the weight coefficients w1 and w2 are adjusted according to the actual task.

[0133]

[0134] The Adam optimizer is used for model training, and the learning rate is set to 10 -3 The dataset is divided into 70% training set and 30% validation set, and training is performed until the loss function converges.

[0135] The model performance was evaluated through the validation set, and the results showed that the mean square error of the total heading deviation prediction and the mean square error of the PID parameter prediction both reached the expected target. The application of the model in the actual arc trajectory task verified its effectiveness in dynamic trajectory control, especially in the scenario where the RTK signal is limited, which significantly improved the trajectory tracking accuracy.

[0136] When the system detects that the RTK signal is limited, the control model is automatically enabled. The model has been trained with historical data and has established a mapping relationship between IMU angle data, RTK positioning data, heading angle deviation, and PID parameters. Therefore, in the absence of RTK signals, the system can adjust the motor speed in real time based on the IMU attitude data and the output of the control model to ensure that the transport aircraft can still travel along the target trajectory.

[0137] This prediction model is mainly used to predict the total heading angle deviation and the optimization direction of the adaptive PID parameters in advance, but the final motor control command is output by the control model instead of directly relying on the data of the prediction model. The prediction data is used as auxiliary input to dynamically adjust the heading angle and PID parameters to enable the transport aircraft to maintain the accuracy and stability of trajectory control.

[0138] Based on the above solution, the present invention also provides a corresponding hardware architecture. Figure 3 For example:

[0139] The system hardware structure is physically connected by solid lines. The IMU module and RTK positioning module are connected to Jetson Nano through the UART interface, and Jetson Nano sends the generated speed correction command to the track motor drive module through the RS232 interface. The main hardware components of the system are as follows:

[0140] Inertial Measurement Unit (IMU): A high-precision inertial measurement unit that integrates a three-axis accelerometer, gyroscope, and magnetometer, outputs the attitude information of the crawler conveyor in real time (including roll angle, pitch angle, and heading angle). Through the filtering algorithm, the noise interference of the IMU data is significantly reduced, thereby providing a more stable attitude angle input for arc trajectory control.

[0141] Real-time Kinematic Differential (RTK) Positioning Module: It consists of a high-precision GNSS receiver and a wireless data transmission unit, and can provide centimeter-level three-dimensional position information (including longitude, latitude and altitude) in real time. The system greatly improves positioning accuracy through differential data correction between the base station and the mobile station, thereby achieving precise navigation in complex environments.

[0142] Jetson Nano module: The Jetson Nano module integrates a high-performance NVIDIA GPU and a quad-core ARM CPU, with powerful computing power and energy efficiency. The module processes sensor data in real time, executes deep learning algorithms and path planning tasks, and achieves environmental perception, autonomous walking, and precise control. Jetson Nano also uses a multi-threaded architecture to separate data acquisition, fusion computing, and instruction generation into independent threads, significantly improving the system's real-time performance and operating efficiency.

[0143] Crawler chassis: A universal explosion-proof robot crawler chassis with standard communication interfaces (such as CAN and RS232), supporting various control methods such as speed loop and torque loop, which can meet the needs of crawler conveyors to perform tasks on different tracks. The crawler chassis also has good impact resistance and terrain adaptability, and is suitable for complex operating environments.

[0144] After receiving the data from IMU and RTK, Jetson Nano uses the unscented Kalman filter algorithm (UKF) to optimize the IMU raw data to reduce noise interference and improve data stability and accuracy. The processed heading angle data is used as the input variable of the heading angle PID controller. The internal algorithm is used to compare the real-time position information of the crawler conveyor with the trajectory information generated after the arc shape of the target is remotely input, and the coordinate deviation and heading angle deviation are calculated, which are used as the input of the position PID and heading angle PID respectively.

[0145] Jetson Nano is the core control unit of the system, responsible for data fusion and deviation calculation. Based on the cascade PID controller, it generates motor speed correction instructions to achieve precise arc control. Through the cascade PID controller, Jetson Nano converts the position deviation into the angle difference between the current trajectory and the desired trajectory, and then makes a comprehensive trade-off between this angle difference and the heading angle deviation value calculated by the system, calculates the real-time speed correction of the crawler motor, and achieves precise arc control of the crawler conveyor.

[0146] Jetson Nano supports WiFi remote control, and acts as a TCP server to receive the target trajectory sent by the control end (such as a PC or mobile app). To ensure the stability of network communication, the system has designed a heartbeat packet mechanism for real-time monitoring and recovery of connection status, supporting remote real-time operation and trajectory setting. To ensure the accuracy of control command transmission, the system uses a cyclic redundancy check mechanism to detect data integrity and designs a command confirmation feedback process. If there is an abnormality in the command transmission, JetsonNano will trigger a retransmission mechanism to ensure that the control command can be accurately executed. Through this double verification mechanism, the system effectively avoids the risk of communication interference or command loss.

[0147] After the data collected by the IMU and RTK modules are uploaded to JetsonNano through the UART interface, the data fusion and cascade PID control algorithm integrate and process the information to generate speed correction instructions. Finally, the correction instructions are sent to the crawler motor drive module through the RS232 interface. The drive module adjusts the speed difference between the left and right crawler motors in real time according to the instructions to ensure that the conveyor can accurately travel along the predetermined arc trajectory. At the same time, the remote control module provides a flexible trajectory setting function, which enables the system to adapt to complex and changing environments and mission requirements, thereby improving the reliability and operating efficiency of crawler conveyors in actual application scenarios.

[0148] An IMU-PID-RTK fused crawler conveyor control system, comprising:

[0149] A remote input unit generates a target trajectory in response to an input operation;

[0150] Data acquisition unit, which obtains current status information through IMU and RTK;

[0151] A control unit calculates a position deviation and a heading angle deviation according to the target trajectory and current state information; and generates a control instruction through a cascade PID according to the position deviation and the heading angle deviation;

[0152] An execution unit executes an action according to the control instruction.

[0153] The contents of the above method embodiments are all applicable to the present system embodiments. The functions specifically implemented by the present system embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0154] A crawler conveyor control device with IMU-PID-RTK fusion:

[0155] at least one processor;

[0156] at least one memory for storing at least one program;

[0157] When the at least one program is executed by the at least one processor, the at least one processor implements the IMU-PID-RTK fusion crawler conveyor control method as described above.

[0158] The contents of the above method embodiments are all applicable to the present device embodiments. The functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0159] A storage medium stores processor-executable instructions, which are used to implement the above-mentioned IMU-PID-RTK fused crawler conveyor control method when executed by the processor.

[0160] The contents of the above method embodiments are all applicable to the present storage medium embodiments. The functions specifically implemented by the present storage medium embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0161] The above is a specific description of the preferred implementation of the present invention, but the invention is not limited to the embodiments. Those skilled in the art may make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of this application.

Claims

1. A crawler conveyor control method fused with IMU-PID-RTK, characterized in that: The following steps are involved: generating a target trajectory in response to an input operation; Get current status information through IMU and RTK; According to the target trajectory and the current state information, the position deviation and the heading angle deviation are calculated, and a control instruction is generated through a cascade PID; An action is executed according to the control instruction.

2. According to claim 1, a crawler conveyor control method of IMU-PID-RTK fusion is characterized in that: The step of generating a target trajectory in response to the input operation specifically includes: Enter the geometry of the target arc; Analytically obtaining a curvature and a step length according to the geometric shape; According to the curvature and the step length, the target arc is decomposed into a series of discrete points, and the spatial coordinates and heading angle of each point are calculated.

3. According to claim 2, a crawler conveyor control method of IMU-PID-RTK fusion, characterized in that: The step of obtaining current status information through IMU and RTK also includes: UKF is used to optimize the IMU data and output optimized attitude angle information.

4. According to claim 2, the IMU-PID-RTK fusion crawler conveyor control method is characterized in that: The step of calculating the position deviation and the heading angle deviation according to the target trajectory and the current state information, and generating the control instruction through cascade PID specifically includes: Calculate the position deviation according to the current coordinates in the current state information and the target point coordinates in the target trajectory; Outputting a correction value of the heading angle according to the position deviation; Generating a total heading angle deviation according to the position deviation and the heading angle correction amount; According to the total heading angle deviation, a motor speed correction amount is output to obtain a control instruction.

5. According to claim 4, a crawler conveyor control method of IMU-PID-RTK fusion is characterized in that: The calculation formula of the correction amount of the heading angle is expressed as follows: Where Δr represents the position deviation, They are the proportional, integral and differential gain parameters of the position PID controller respectively.

6. According to claim 4, the IMU-PID-RTK fusion crawler conveyor control method is characterized in that: The calculation formula of the motor speed correction is as follows: in, are the proportional, integral and differential gain parameters of the heading angle PID, Δθ total Indicates the total heading angle deviation.

7. According to claim 4, the IMU-PID-RTK fusion crawler conveyor control method is characterized in that: Also includes: An adaptive parameter optimization mechanism is introduced to dynamically adjust the parameters of the heading angle PID according to the curvature.

8. According to claim 6, the IMU-PID-RTK fusion crawler conveyor control method is characterized in that: Also includes: Build a prediction model based on LSTM; Constructing a nonlinear mapping relationship between heading angle PID parameters, heading angle deviation and RTK data based on the prediction model; The total heading angle deviation and dynamically optimized PID parameters are predicted based on the prediction model.

9. An IMU-PID-RTK fusion crawler conveyor control system, characterized in that: include: A remote input unit generates a target trajectory in response to an input operation; Data acquisition unit, which obtains current status information through IMU and RTK; A control unit calculates a position deviation and a heading angle deviation according to the target trajectory and current state information; and generates a control instruction through a cascade PID according to the position deviation and the heading angle deviation; An execution unit executes an action according to the control instruction.

Images