An adaptive tracking and positioning method for engineering vehicles

Abstract

The present invention discloses an adaptive tracking and positioning method for construction vehicles, relating to the field of vehicle navigation and control technology. The method comprises collecting position information and operating parameters, transmitting them to an onboard control unit via an onboard sensor network; fusing the position information and operating parameters via the onboard control unit, employing a Kalman filter algorithm to eliminate GPS signal noise and accumulated inertial navigation errors, generating real-time position trajectory data, and combining this with a pre-stored electronic map of the construction area. An adaptive particle filter algorithm is then employed to calculate the deviation between the construction vehicle and a predetermined path, which is then input into a constructed neural network model to predict the next position of the construction vehicle. Based on the error between the predicted and actual positions of the construction vehicle, the neural network model parameters are dynamically adjusted to generate steering angle corrections and speed adjustment commands, which are then executed by a controller to achieve trajectory tracking and positioning of the construction vehicle. The present invention effectively solves the problems of precise positioning and trajectory tracking of construction vehicles in complex environments.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle navigation and control, in particular to an adaptive tracking and positioning method for an engineering vehicle. Background Art

[0002] Currently, construction vehicle positioning systems primarily rely on a combination of global positioning systems (GPS) and inertial navigation systems (INS). However, in complex engineering environments, GPS signals are susceptible to obstruction and multipath effects, resulting in reduced positioning accuracy. Furthermore, INS systems exhibit cumulative errors, which can cause positioning deviations to increase over time. Furthermore, traditional positioning algorithms often employ fixed-parameter filtering methods, which struggle to adapt to the dynamic characteristics of construction vehicles under varying operating conditions, impacting the accuracy and real-time nature of trajectory tracking.

[0003] Existing engineering vehicle trajectory tracking technologies typically employ classic control methods such as PID control or fuzzy control, which have significant limitations when dealing with nonlinear and time-varying system characteristics. Particularly in complex terrain, where factors such as ground friction and slope vary, traditional control methods struggle to achieve accurate trajectory tracking. Furthermore, existing technologies lack the ability to effectively predict the vehicle's motion state, making it impossible to plan and optimize control strategies in advance. This results in delayed tracking responses and is prone to trajectory deviations. Furthermore, existing systems commonly suffer from low sensor data fusion efficiency and poor anti-interference capabilities, making it difficult to ensure sustained and stable operation in harsh environments.

[0004] In response to the above problems, the present invention proposes an adaptive tracking and positioning method for engineering vehicles. This method effectively solves the problems of precise positioning and trajectory tracking of engineering vehicles in complex environments by combining multi-source sensor data fusion, adaptive filtering algorithm and neural network model, thereby improving the environmental adaptability and control accuracy of the system. Summary of the Invention

[0005] In view of the problems that the existing engineering vehicle trajectory tracking technology has in complex working conditions, such as low positioning accuracy, poor anti-interference ability, delayed control response and insufficient environmental adaptability, the present invention is proposed.

[0006] Therefore, the problem to be solved by the present invention is how to improve the positioning accuracy and trajectory tracking control performance of engineering vehicles in complex environments and achieve the adaptability of the system.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0008] In a first aspect, an embodiment of the present invention provides an adaptive tracking and positioning method for an engineering vehicle, which includes collecting position information and operating parameters of the engineering vehicle during driving, and transmitting them to an on-board control unit through on-board sensors; fusing the position information and the operating parameters through the on-board control unit, and using a Kalman filter algorithm to eliminate GPS signal noise and inertial navigation cumulative errors to generate real-time position trajectory data of the engineering vehicle; based on the real-time position trajectory data, in combination with a pre-stored electronic map of the engineering area, using an adaptive particle filter algorithm to calculate the deviation value between the engineering vehicle and the predetermined path; inputting the deviation value and the operating parameters into a constructed neural network model to predict the next moment position of the engineering vehicle; dynamically adjusting the parameters of the neural network model according to the error between the predicted position and the actual position of the engineering vehicle to generate a steering angle correction value and a speed adjustment instruction; executing the steering angle correction value and the speed adjustment instruction through a controller to achieve trajectory tracking and positioning of the engineering vehicle.

[0009] As a preferred solution of the adaptive tracking and positioning method for engineering vehicles described in the present invention, wherein: the position information includes GPS coordinate data and inertial navigation data; the operating parameters include acceleration, angular velocity, steering angle and vehicle posture data; the method for generating the real-time position trajectory data is to synchronize the GPS coordinate data and the inertial navigation data according to timestamps, establish a state equation and a measurement equation for combined navigation of the GPS coordinate data and the inertial navigation data; predict the position state of the engineering vehicle according to the state equation, and calculate the covariance matrix of the state prediction at the same time, wherein the covariance matrix includes the process noise covariance matrix and the measurement noise covariance matrix; according to the The Kalman gain matrix is ​​calculated using the measurement equation, and the GPS coordinate data is used as the observation quantity to correct the predicted state and eliminate the accumulated drift of the inertial navigation data; an adaptive Kalman filter algorithm is used to dynamically adjust the process noise covariance matrix and the measurement noise covariance matrix, wherein the process noise covariance matrix includes the random error of the inertial sensor; the measurement noise covariance matrix includes the multipath effect and atmospheric delay of the GPS signal; based on the filtered position coordinates, the motion trajectory of the engineering vehicle is continuously recorded according to the sampling period, the position coordinates between two adjacent sampling points are smoothly connected by the cubic spline interpolation algorithm to form the real-time position trajectory data of the engineering vehicle, and the deviation value between the engineering vehicle and the predetermined path is calculated.

[0010] As a preferred solution of the adaptive tracking and positioning method for engineering vehicles of the present invention, the specific formula of the state equation is as follows:

[0011]

[0012] Among them, X k is the state vector at time k, Φ k,k-1is the state transfer matrix, B k is the control input matrix at time k, u k is the control input vector at time k, α i is the weight coefficient of the i-th inertial sensor, σ i is the standard deviation of the i-th inertial sensor, w i is the measurement value of the i-th inertial sensor, and n is the number of inertial sensors.

[0013] The specific formula of the measurement equation is as follows:

[0014]

[0015] Among them, Z k is the observation vector at time k, H k is the observation matrix at time k, μ i is the mean value of the i-th inertial sensor, β j is the weight coefficient of the jth GPS receiver channel, and m is the number of GPS receiver channels.

[0016] As a preferred embodiment of the adaptive tracking and positioning method for engineering vehicles according to the present invention, the deviation value is calculated by retrieving an electronic map of the engineering area from a memory and generating N initial particles on the electronic map of the engineering area, wherein the electronic map of the engineering area includes road grid data, road attribute data, and a sequence of predetermined path coordinate points; matching the real-time position trajectory data with the sequence of predetermined path coordinate points, calculating the likelihood of the particles based on the road grid data, and updating the particle weights; using an adaptive resampling threshold, when the particle weight variance is greater than a first threshold, triggering the resampling process, and calculating the optimal estimate of the current position of the engineering vehicle based on the resampled particle swarm; subtracting the optimal estimate from the coordinates of the corresponding point on the predetermined path to obtain a deviation value, wherein the deviation value includes a lateral deviation value and a longitudinal deviation value; inputting the deviation value and the operating parameters into the constructed neural network model, and outputting a prediction result.

[0017] As a preferred solution of the adaptive tracking and positioning method for engineering vehicles described in the present invention, the neural network model includes an input layer, a hidden layer and an output layer; the input layer receives position information, operating parameters and deviation values; the hidden layer uses the tanh activation function to extract the combination relationship between features; and the output layer outputs the predicted position at the next moment through a fully connected layer.

[0018] As a preferred solution of the adaptive tracking and positioning method for engineering vehicles described in the present invention, the method includes: dynamically adjusting the parameters of the neural network model according to the error between the predicted position and the actual position of the engineering vehicle, and generating a steering angle correction value and a speed adjustment instruction, including: calculating the Euclidean distance between the predicted position and the actual position, and using the calculated position error, error change rate and error accumulation value as feedback signals; using the backpropagation algorithm to update the weight parameters of the neural network model, and using the Adam optimizer to adaptively adjust the learning rate, wherein the weight update amount is proportional to the position error value; based on the directional component of the position error and the current heading angle of the engineering vehicle, calculating the steering angle value to be corrected, and mapping it to the effective range of the steering actuator through a sigmoid function to generate a steering angle correction value; setting the speed adjustment coefficient according to the position error value, reducing the driving speed when the position error is greater than or equal to a second threshold; and restoring the normal speed when the position error is less than the second threshold, converting the speed adjustment coefficient into a speed adjustment instruction recognizable by the engine control unit, and visually analyzing the intermediate layer features of the neural network model to extract the feature combination that contributes most to the prediction accuracy, and dynamically adjusting the discard rate and attention weight.

[0019] As a preferred solution of the adaptive tracking and positioning method for engineering vehicles described in the present invention, the controller executes the steering angle correction and the speed adjustment instruction to achieve trajectory tracking and positioning of the engineering vehicle, including: sending the steering angle correction to the steering controller via the CAN bus, and using the PID algorithm to perform closed-loop control on the speed and angle of the steering motor; at the same time, sending the speed adjustment instruction to the engine control unit via the CAN bus, adjusting the throttle opening and fuel injection amount according to the speed adjustment instruction to control the engine output torque; real-time monitoring of the response status of the steering actuator and the power output system, and when it is detected that the execution deviation exceeds a third threshold, issuing an alarm signal and initiating emergency processing operations.

[0020] In a second aspect, an embodiment of 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 instructions are executed by the processor, the steps of the adaptive tracking and positioning method for engineering vehicles as described in the first aspect of the present invention are implemented.

[0021] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, the steps of the adaptive tracking and positioning method for engineering vehicles as described in the first aspect of the present invention are implemented.

[0022] The beneficial effects of the present invention are as follows: real-time collection and transmission of multi-source data are achieved through the on-board sensor network, and the errors of the GPS and inertial navigation systems are eliminated in combination with the Kalman filter algorithm, thereby improving positioning accuracy; an adaptive particle filter algorithm is used to calculate trajectory deviations, thereby enhancing the environmental adaptability of the system under complex terrain conditions; the vehicle position is predicted through a neural network model and the model parameters are dynamically adjusted, thereby achieving forward-looking optimization of the control strategy; and finally, steering angle correction and speed adjustment are executed through a closed-loop feedback control mechanism, thereby ensuring accurate trajectory tracking of the engineering vehicle and improving the efficiency and safety of construction operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort. Among them:

[0024] Figure 1 This is a flow chart of the adaptive tracking and positioning method for an engineering vehicle in Example 1. DETAILED DESCRIPTION

[0025] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0026] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. 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.

[0027] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0028] Example 1

[0029] Reference Figure 1 , which is the first embodiment of the present invention, provides an adaptive tracking and positioning method for an engineering vehicle, comprising:

[0030] S1: Collects the location information and operating parameters of the engineering vehicle while it is driving, and transmits them to the on-board control unit through on-board sensors.

[0031] Specifically, the on-board sensor uses the CAN bus to transmit the collected position information and operating parameters to the on-board control unit, wherein the on-board control unit time-stamps the received data and stores it in the on-board memory.

[0032] It should be noted that the location information includes GPS coordinate data and inertial navigation data; the operating parameters include acceleration, angular velocity, steering angle and vehicle posture data.

[0033] Furthermore, GPS coordinate data is obtained by receiving GPS satellite signals through a GPS antenna; inertial navigation data is obtained by measuring the accelerometer and gyroscope in the inertial measurement unit; acceleration data and angular velocity data are obtained by measuring the six-axis inertial sensor; steering angle data is obtained by measuring the steering angle sensor; and vehicle attitude data is obtained by measuring the attitude sensor.

[0034] S2: The vehicle-mounted control unit fuses the position information and the operating parameters, and uses a Kalman filter algorithm to eliminate GPS signal noise and inertial navigation cumulative error to generate real-time position trajectory data of the engineering vehicle.

[0035] Specifically, the method for generating real-time position trajectory data is to synchronize the GPS coordinate data and the inertial navigation data according to the timestamps, and establish the state equation and measurement equation of the combined navigation of the GPS coordinate data and the inertial navigation data.

[0036] Furthermore, the specific formula of the state equation is as follows:

[0037]

[0038] Among them, X k is the state vector at time k, Φ k,k-1 is the state transfer matrix, B k is the control input matrix at time k, u k is the control input vector at time k, α i is the weight coefficient of the i-th inertial sensor, σ i is the standard deviation of the i-th inertial sensor, w i is the measurement value of the i-th inertial sensor, and n is the number of inertial sensors.

[0039] The specific formula of the measurement equation is as follows:

[0040]

[0041] Among them, Z k is the observation vector at time k, H k is the observation matrix at time k, μ i is the mean value of the i-th inertial sensor, β jis the weight coefficient of the jth GPS receiver channel, and m is the number of GPS receiver channels.

[0042] Furthermore, the position state of the engineering vehicle is predicted based on the state equation, and the covariance matrix of the state prediction is calculated at the same time; the Kalman gain matrix is ​​calculated based on the measurement equation, and the GPS coordinate data is used as the observation quantity to correct the predicted state and eliminate the accumulated drift of the inertial navigation data.

[0043] Specifically, the dimension of the Kalman gain matrix is ​​9×3, where the row vectors correspond to the 9 components of the state vector and the column vectors correspond to the 3 components of the observation vector.

[0044] It should be noted that the covariance matrix includes the process noise covariance matrix and the measurement noise covariance matrix; the process noise covariance matrix includes the random error of the inertial sensor; the measurement noise covariance matrix includes the multipath effect and atmospheric delay of the GPS signal.

[0045] Furthermore, an adaptive Kalman filter algorithm is used to dynamically adjust the process noise covariance matrix and the measurement noise covariance matrix; based on the filtered position coordinates, the motion trajectory of the engineering vehicle is continuously recorded according to the sampling period, and the position coordinates between two adjacent sampling points are smoothly connected through the cubic spline interpolation algorithm to form the real-time position trajectory data of the engineering vehicle, and the deviation value between the engineering vehicle and the planned path is calculated.

[0046] S3: Based on the real-time position trajectory data and in combination with a pre-stored electronic map of the engineering area, an adaptive particle filter algorithm is used to calculate the deviation value between the engineering vehicle and the predetermined path.

[0047] Specifically, the deviation value is calculated by retrieving an electronic map of the project area from a memory and generating N initial particles on the electronic map of the project area; matching the real-time position trajectory data with the sequence of predetermined path coordinate points, calculating the likelihood of the particles based on the road grid data, and updating the particle weights.

[0048] It should be noted that the electronic map of the project area includes road grid data, road attribute data and a sequence of predetermined path coordinate points; each initial particle carries position coordinates and weight information.

[0049] Furthermore, an adaptive resampling threshold is adopted. When the particle weight variance is greater than a first threshold, the resampling process is triggered, and the optimal estimate of the current position of the engineering vehicle is calculated based on the resampled particle swarm. When the particle weight variance is less than or equal to the first threshold, the optimal estimate of the engineering vehicle's position is directly calculated based on the current particle swarm weight. The optimal estimate is subtracted from the coordinates of the corresponding point on the predetermined path to obtain a deviation value, where the deviation value includes a lateral deviation value and a longitudinal deviation value.

[0050] Furthermore, the specific formula for the optimal estimate is as follows:

[0051]

[0052] in, is the optimal estimate at time k, is the prior state estimate at time k, K k is the Kalman gain matrix, which is used to weigh the credibility of the predicted value and the observed value, P k|k-1 is the state prediction error covariance matrix at time k, is the transpose of the observation matrix, R k is the observation noise covariance matrix.

[0053] It should be noted that the first threshold is based on vehicle operation data under different working conditions. It is the optimal value determined by comparing and analyzing the position tracking accuracy and algorithm calculation efficiency under multiple groups of thresholds, and comprehensively weighing the positioning accuracy and real-time requirements.

[0054] S4: Input the deviation value and the operating parameters into the constructed neural network model to predict the next position of the engineering vehicle.

[0055] Specifically, the neural network model includes an input layer, a hidden layer, and an output layer; the input layer receives position information, operating parameters, and bias values; the hidden layer uses the tanh activation function to extract the combination relationship between features; and the output layer outputs the predicted position at the next moment through the fully connected layer.

[0056] It should be noted that the position at the next moment includes longitude coordinates, latitude coordinates and elevation coordinates.

[0057] S5: Based on the error between the predicted position and the actual position of the engineering vehicle, the neural network model parameters are dynamically adjusted to generate steering angle correction and speed adjustment instructions.

[0058] Specifically, the Euclidean distance between the predicted position and the actual position is calculated, and the calculated position error, error change rate, and error accumulation value are used as feedback signals. The backpropagation algorithm is used to update the weight parameters of the neural network model, and the Adam optimizer is used to adaptively adjust the learning rate, where the weight update amount is proportional to the position error value.

[0059] Furthermore, based on the directional component of the position error and the current heading angle of the engineering vehicle, the steering angle value that needs to be corrected is calculated and mapped to the effective range of the steering actuator through the sigmoid function to generate the steering angle correction value.

[0060] Furthermore, the specific formula for the steering angle correction is as follows:

[0061]

[0062] Among them, Δθ is the steering angle correction, Q is the steering gain coefficient, e y is the lateral position error, e x is the longitudinal position error, δ is the sensitivity coefficient, d is the current position error, d0 is the expected position error threshold, τ is the historical correction weight coefficient, γ i is the weight of the i-th historical revision, and M is the number of historical revisions considered.

[0063] Specifically, a speed adjustment coefficient is set according to the position error value. When the position error is greater than or equal to the second threshold, the driving speed is reduced; when the position error is less than the second threshold, the normal speed is restored. The speed adjustment coefficient is converted into a speed adjustment instruction that can be recognized by the engine control unit, and the intermediate layer features of the neural network model are visualized and analyzed to extract the feature combination that contributes most to the prediction accuracy, and the discard rate and attention weight are dynamically adjusted.

[0064] It should be noted that the second threshold is a position error critical value determined based on the dynamic response characteristics of the engineering vehicle steering system and safe driving requirements.

[0065] S6: The controller executes the steering angle correction value and the speed adjustment instruction to achieve trajectory tracking and positioning of the engineering vehicle.

[0066] Specifically, the steering angle correction value is sent to the steering controller via the CAN bus, and the PID algorithm is used to perform closed-loop control on the speed and angle of the steering motor; at the same time, the speed adjustment instruction is sent to the engine control unit via the CAN bus, and the throttle opening and fuel injection amount are adjusted according to the speed adjustment instruction to control the engine output torque.

[0067] Furthermore, the response status of the steering actuator and the power output system is monitored in real time. When it is detected that the execution deviation exceeds the third threshold, an alarm signal is issued and an emergency processing operation is initiated; when it is detected that the execution deviation is lower than the third threshold, the normal control logic is continued to be maintained and the current execution status data is recorded.

[0068] It should be noted that the third threshold is an execution deviation tolerance value determined based on the mechanical tolerance range of the steering actuator and the power output system and the dynamic response characteristics of the controller, and is used to determine whether the system is in a normal working state.

[0069] Furthermore, the emergency handling operation includes using a fuzzy controller to smooth the steering angle correction to eliminate the severe shaking caused by the steering system; at the same time, setting a buffer zone for speed adjustment, using a ramp function to achieve gradual speed adjustment, and recording the state parameters during the execution process into the memory.

[0070] In summary, the present invention realizes the real-time collection and transmission of multi-source data through the vehicle-mounted sensor network, and combines the Kalman filter algorithm to eliminate the errors of GPS and inertial navigation systems, thereby improving positioning accuracy; adopts the adaptive particle filter algorithm to calculate the trajectory deviation, thereby enhancing the environmental adaptability of the system under complex terrain conditions; predicts the vehicle position through the neural network model and dynamically adjusts the model parameters, thereby realizing the forward-looking optimization of the control strategy; finally, the closed-loop feedback control mechanism is used to perform steering angle correction and speed regulation, thereby ensuring the accurate trajectory tracking of the engineering vehicle and improving the efficiency and safety of construction operations.

[0071] This embodiment also provides a computer device suitable for the adaptive tracking and positioning method for engineering vehicles, including a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the adaptive tracking and positioning method for engineering vehicles proposed in the above embodiment.

[0072] The computer device may be a terminal, comprising 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 comprises 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 may 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 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 button, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse.

[0073] This embodiment also provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the following steps: collecting position information and operating parameters of the engineering vehicle while it is traveling, and transmitting them to an on-board control unit via on-board sensors; fusing the position information and the operating parameters via the on-board control unit, and using a Kalman filter algorithm to eliminate GPS signal noise and accumulated inertial navigation errors to generate real-time position trajectory data of the engineering vehicle; based on the real-time position trajectory data and in combination with a pre-stored electronic map of the engineering area, using an adaptive particle filter algorithm to calculate the deviation between the engineering vehicle and a predetermined path; inputting the deviation value and the operating parameters into a constructed neural network model to predict the position of the engineering vehicle at the next moment; dynamically adjusting the parameters of the neural network model based on the error between the predicted position and the actual position of the engineering vehicle to generate a steering angle correction value and a speed adjustment instruction; and executing the steering angle correction value and the speed adjustment instruction via a controller to achieve trajectory tracking and positioning of the engineering vehicle.

[0074] Example 2

[0075] Referring to Table 1, which is a second embodiment of the present invention, this embodiment provides an adaptive tracking and positioning method for an engineering vehicle. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0076] Specifically, a CAT 793F mining dump truck from a mining area was selected as the test subject for adaptive tracking and positioning testing along a pre-set haul route within the mining area. The test vehicle was equipped with a BeiDou / GPS dual-mode receiver (sampling frequency: 20Hz), an MTi-G-710 inertial measurement unit (sampling frequency: 100Hz), a steering angle sensor, and an onboard computing unit. A test route consisting of straight sections, curves, and ramps was set up along a typical 5-kilometer-long haul route within the mining area, collecting eight hours of continuous test data.

[0077] Furthermore, the data from the BeiDou / GPS receiver and the inertial measurement unit are aligned by time stamp, and an adaptive Kalman filter algorithm is used for integrated navigation. The initial value of the process noise covariance matrix is ​​set to a diagonal matrix of [0.01, 0.01, 0.01], and the initial value of the measurement noise covariance matrix is ​​set to a diagonal matrix of [0.1, 0.1, 0.1].

[0078] Furthermore, in the electronic map matching phase, the number of particles is set to 1000, and the particle weight variance threshold is set to 0.6. An improved Monte Carlo positioning algorithm is used, introducing road topology constraints to improve particle sampling efficiency. The neural network model adopts a four-layer structure, with 88 neurons in the input layer, two hidden layers with 64 and 32 neurons respectively, and 6 neurons in the output layer. The hidden layers use the tanh activation function, and the loss rate is initially set to 0.3. The steering control system adopts a hierarchical control strategy based on fuzzy PID. The steering motor uses a permanent magnet synchronous motor with a rated power of 45kW and a maximum steering angle of ±35°. For speed control, the engine speed is controlled within the range of 1200-1800 rpm, and smooth speed changes are achieved by adjusting the throttle opening. The system sets the position error alarm threshold to ±1.5 meters and the execution deviation alarm threshold to ±2.5°.

[0079] Specifically, the method of the present invention offers significant performance advantages over traditional methods. As shown in Table 1, in terms of positioning accuracy, the average positioning error of the method of the present invention is only 0.38-0.51 meters, an improvement of approximately 75% compared to the 1.82 meters of the traditional method. The maximum lateral deviation is also reduced from 2.45 meters to between 0.78-0.92 meters, an improvement of approximately 65%. This is primarily due to the adaptive Kalman filter algorithm's effective suppression of GPS signal noise and inertial device errors, as well as the predictive positioning strategy combining particle filtering with a neural network. In terms of trajectory tracking performance, the tracking accuracy of the method of the present invention is maintained within the range of 0.48-0.58 meters, an improvement of approximately 73% compared to the 1.95 meters of the traditional method. The system response time is reduced from 280 milliseconds to 145-160 milliseconds, an improvement of approximately 45%. This demonstrates that the hierarchical control strategy and adaptive parameter adjustment mechanism of the present invention can effectively improve the system's dynamic tracking performance. Even in complex road conditions, such as curves and slopes, the method of the present invention maintains high positioning accuracy and tracking performance.

[0080] Table 1. Comparison between the present invention and traditional methods

[0081]

[0082] Furthermore, while the positioning error increased slightly (0.51 meters) on curved sections, it was still far superior to traditional methods. This demonstrates the excellent environmental adaptability of the neural network prediction model. Furthermore, by dynamically adjusting speed and steering parameters, the system not only maintains tracking accuracy but also reduces fuel consumption (by an average of approximately 8%) and improves the positioning success rate (from 92.5% to approximately 98%).

[0083] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. 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. An adaptive tracking and positioning method for an engineering vehicle, characterized by: include, Collect the location information and operating parameters of the engineering vehicle while it is driving, and transmit them to the on-board control unit through on-board sensors; The vehicle-mounted control unit fuses the position information and the operating parameters, and uses a Kalman filter algorithm to eliminate GPS signal noise and inertial navigation cumulative errors to generate real-time position trajectory data of the engineering vehicle; Based on the real-time location trajectory data and in combination with a pre-stored electronic map of the engineering area, an adaptive particle filter algorithm is used to calculate the deviation between the engineering vehicle and the predetermined path; Inputting the deviation value and the operating parameters into the constructed neural network model to predict the next position of the engineering vehicle; According to the error between the predicted and actual positions of the engineering vehicle, the neural network model parameters are dynamically adjusted to generate steering angle correction and speed adjustment instructions; The steering angle correction value and the speed adjustment instruction are executed by a controller to achieve trajectory tracking and positioning of the engineering vehicle.

2. The adaptive tracking and positioning method for an engineering vehicle according to claim 1, wherein: The position information includes GPS coordinate data and inertial navigation data; the operating parameters include acceleration, angular velocity, steering angle and vehicle posture data; the method for generating the real-time position trajectory data is: Synchronously aligning the GPS coordinate data and the inertial navigation data according to timestamps, and establishing a state equation and a measurement equation for combined navigation of the GPS coordinate data and the inertial navigation data; Predicting the position state of the engineering vehicle according to the state equation, and calculating a covariance matrix of the state prediction, wherein the covariance matrix includes a process noise covariance matrix and a measurement noise covariance matrix; Calculate the Kalman gain matrix according to the measurement equation, use the GPS coordinate data as the observation quantity, correct the predicted state, and eliminate the accumulated drift of the inertial navigation data; Adopting an adaptive Kalman filter algorithm to dynamically adjust the process noise covariance matrix and the measurement noise covariance matrix, wherein the process noise covariance matrix includes the random error of the inertial sensor; the measurement noise covariance matrix includes the multipath effect and atmospheric delay of the GPS signal; Based on the filtered position coordinates, the motion trajectory of the engineering vehicle is continuously recorded according to the sampling period. The position coordinates between two adjacent sampling points are smoothly connected through the cubic spline interpolation algorithm to form the real-time position trajectory data of the engineering vehicle, and the deviation value between the engineering vehicle and the planned path is calculated.

3. The adaptive tracking and positioning method for an engineering vehicle according to claim 2, wherein: The specific formula of the state equation is as follows: Among them, X k is the state vector at time k, Φ k,k-1 is the state transfer matrix, B k is the control input matrix at time k, u k is the control input vector at time k, α i is the weight coefficient of the i-th inertial sensor, σ i is the standard deviation of the i-th inertial sensor, w i is the measurement value of the i-th inertial sensor, and n is the number of inertial sensors; The specific formula of the measurement equation is as follows: Among them, Z k is the observation vector at time k, H k is the observation matrix at time k, μ i is the mean value of the i-th inertial sensor, β j is the weight coefficient of the jth GPS receiver channel, and m is the number of GPS receiver channels.

4. The adaptive tracking and positioning method for an engineering vehicle according to claim 2, wherein: The calculation method of the deviation value is: Retrieving an electronic map of the project area from a memory and generating N initial particles on the electronic map of the project area, wherein the electronic map of the project area includes road grid data, road attribute data, and a sequence of predetermined path coordinate points; Matching the real-time position trajectory data with the predetermined path coordinate point sequence, calculating the likelihood of particles based on the road grid data, and updating the particle weights; Adaptive resampling threshold is used. When the particle weight variance is greater than the first threshold, the resampling process is triggered, and the optimal estimate of the current position of the engineering vehicle is calculated based on the resampled particle swarm. Subtracting the optimal estimate from the coordinates of the corresponding point on the predetermined path to obtain a deviation value, wherein the deviation value includes a lateral deviation value and a longitudinal deviation value; The deviation value and the operating parameters are input into the constructed neural network model, and the prediction result is output.

5. The adaptive tracking and positioning method for an engineering vehicle according to claim 4, wherein: The neural network model includes an input layer, a hidden layer and an output layer; the input layer receives position information, operating parameters and deviation values; the hidden layer uses a tanh activation function to extract the combination relationship between features; and the output layer outputs the predicted position at the next moment through a fully connected layer.

6. The adaptive tracking and positioning method for an engineering vehicle according to claim 5, wherein: Based on the error between the predicted and actual positions of the engineering vehicle, the neural network model parameters are dynamically adjusted to generate steering angle corrections and speed adjustment instructions, including: Calculate the Euclidean distance between the predicted position and the actual position, and use the calculated position error, error change rate, and error accumulation value as feedback signals; The back-propagation algorithm is used to update the weight parameters of the neural network model, and the Adam optimizer is used to adaptively adjust the learning rate, where the weight update amount is proportional to the position error value; Based on the directional component of the position error and the current heading angle of the engineering vehicle, the steering angle value that needs to be corrected is calculated and mapped to the effective range of the steering actuator through the sigmoid function to generate the steering angle correction value; Setting a speed adjustment coefficient according to the position error value, and reducing the driving speed when the position error is greater than or equal to a second threshold; When the position error is less than the second threshold, the normal speed is restored, the speed adjustment coefficient is converted into a speed adjustment instruction that can be recognized by the engine control unit, and the intermediate layer features of the neural network model are visualized and analyzed to extract the feature combination that contributes most to the prediction accuracy, and dynamically adjust the discard rate and attention weight.

7. The adaptive tracking and positioning method for an engineering vehicle according to claim 6, wherein: The controller executes the steering angle correction value and the speed adjustment instruction to achieve trajectory tracking and positioning of the engineering vehicle, including: The steering angle correction value is sent to the steering controller via the CAN bus, and the speed and angle of the steering motor are closed-loop controlled using a PID algorithm; At the same time, the speed adjustment instruction is sent to the engine control unit via the CAN bus, and the throttle opening and fuel injection amount are adjusted according to the speed adjustment instruction to control the engine output torque; The response status of the steering actuator and the power output system is monitored in real time. When it is detected that the execution deviation exceeds the third threshold, an alarm signal is issued and an emergency processing operation is initiated.

8. The adaptive tracking and positioning method for an engineering vehicle according to claim 7, wherein: The emergency handling operations include: A fuzzy controller is used to smooth the steering angle correction to eliminate severe vibrations generated by the steering system; At the same time, a buffer zone for speed regulation is set, a ramp function is used to implement gradual speed regulation, and state parameters during execution are recorded in a memory.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the adaptive tracking and positioning method for an engineering vehicle according to any one of claims 1 to 8 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 adaptive tracking and positioning method for an engineering vehicle according to any one of claims 1 to 8 are implemented.

Images