A moving target trajectory prediction method and related equipment

Abstract

The present application discloses a method for predicting the trajectory of a moving target and related equipment, relating to the field of trajectory prediction technology. The method comprises: obtaining multi-source dynamic perception data of the moving target, wherein the multi-source dynamic perception data includes target motion state parameters, environmental disturbance parameters, and biological behavior characteristic parameters; generating an initial trajectory prediction result based on the target motion state parameters; dynamically correcting the initial trajectory prediction result based on the environmental disturbance parameters to obtain a disturbance-compensated trajectory; and performing behavioral pattern matching on the disturbance-compensated trajectory based on the biological behavior characteristic parameters to generate a target trajectory prediction result. The present application improves trajectory prediction accuracy in complex scenarios by fusing multi-source dynamic perception data and combining a dual optimization mechanism of dynamic correction of environmental disturbances and matching of biological behavior patterns.

Description

Technical Field

[0001] The present application relates to the field of trajectory prediction technology, and in particular to a moving target trajectory prediction method and related equipment. Background Art

[0002] Existing methods for predicting moving target trajectories rely solely on single-dimensional data. For example, traditional models use radar or visual sensors to obtain the real-time position and velocity of moving targets and then extrapolate their trajectories based on kinematic models. Furthermore, for moving targets with biological characteristics, existing methods lack the ability to accurately predict sudden turns or speed changes due to their inattention to biological behavioral features. This is particularly problematic in long-term tracking scenarios, where error accumulation is a significant issue. Therefore, a moving target trajectory prediction method is urgently needed to address these technical issues. Summary of the Invention

[0003] The Summary of the Invention introduces a series of simplified concepts that will be further described in the Detailed Description of the Invention. The Summary of the Invention of this application is not intended to limit the key features and essential technical features of the claimed technical solution, nor is it intended to determine the scope of protection of the claimed technical solution.

[0004] In a first aspect, the present application provides a moving target trajectory prediction method, comprising:

[0005] Acquiring multi-source dynamic perception data of a moving target, wherein the multi-source dynamic perception data includes target motion state parameters, environmental disturbance parameters, and biological behavior characteristic parameters;

[0006] Generate initial trajectory prediction results based on target motion state parameters;

[0007] Based on the environmental disturbance parameters, the initial trajectory prediction results are dynamically corrected to obtain the disturbance-compensated trajectory;

[0008] Based on the biological behavior characteristic parameters, the disturbance compensation trajectory is matched with the behavior pattern to generate the target trajectory prediction result.

[0009] In some embodiments, obtaining multi-source dynamic perception data of a moving target includes:

[0010] The millimeter-wave radar is used to obtain the real-time speed, acceleration, and direction of movement of the moving target as the target motion state parameters;

[0011] The distributed air pressure sensor array collects the airflow velocity, pressure gradient and terrain undulation around the moving target as environmental disturbance parameters;

[0012] The surface temperature distribution and limb swing frequency of the moving target are extracted by infrared thermal imager as biological behavior characteristic parameters.

[0013] In some embodiments, generating an initial trajectory prediction result based on the target motion state parameters includes:

[0014] Calculate the instantaneous trajectory vector of the moving target based on real-time speed, acceleration and direction of movement;

[0015] Construct the target motion state matrix based on the instantaneous trajectory vector and the preset kinematic model;

[0016] The target motion state matrix is input into the trajectory prediction neural network to generate an initial trajectory prediction result. The trajectory prediction neural network is trained with historical motion data and is used to output a trajectory sequence that matches the current motion state.

[0017] In some embodiments, dynamically correcting the initial trajectory prediction result based on the environmental disturbance parameter to obtain the disturbance-compensated trajectory includes:

[0018] Determine the fluid resistance coefficient of the moving target based on the airflow velocity and movement direction;

[0019] Based on the terrain relief, determine the rate of change of terrain elevation;

[0020] Determine the disturbance compensation model based on the terrain elevation change rate and pressure gradient;

[0021] Based on the fluid resistance coefficient and the disturbance compensation model, the initial trajectory prediction results are dynamically corrected to determine the disturbance compensation trajectory.

[0022] In some embodiments, based on the biological behavioral characteristic parameters, behavioral pattern matching is performed on the disturbance compensation trajectory to generate a target trajectory prediction result, including:

[0023] Determine the energy consumption rate of the moving target based on the body surface temperature distribution;

[0024] Based on the limb swing frequency, a motion rhythm map of the movement target is constructed;

[0025] Determining a trajectory segment correction intensity based on a first comparison result of the energy consumption rate and a preset energy threshold;

[0026] Based on the motion rhythm map, matching the reference trajectory segment in the historical behavior pattern library, wherein the similarity between the reference trajectory segment and the motion rhythm map is greater than a preset similarity threshold;

[0027] Based on the trajectory segment correction strength and the reference trajectory segment, the disturbance compensation trajectory is spatiotemporally superimposed to generate the target trajectory prediction result.

[0028] In some embodiments, further comprising:

[0029] Generate a trajectory visualization map based on the target trajectory prediction results, where the trajectory visualization map includes dynamic risk area markers;

[0030] Based on the dynamic risk area markings, determine the obstacle avoidance instructions or path optimization suggestions corresponding to the moving target.

[0031] In some embodiments, further comprising:

[0032] Based on the target trajectory prediction results, the deviation between the actual motion path of the moving target and the target trajectory prediction results is monitored in real time;

[0033] When the deviation is greater than a first preset threshold, a re-collection instruction of the multi-source dynamic sensing data is triggered;

[0034] Based on the re-collection instruction, multi-source dynamic perception data is re-acquired.

[0035] In a second aspect, the present application proposes a moving target trajectory prediction device, comprising:

[0036] A multi-source data acquisition unit is used to acquire multi-source dynamic perception data of a moving target, wherein the multi-source dynamic perception data includes target motion state parameters, environmental disturbance parameters, and biological behavior characteristic parameters;

[0037] An initial trajectory generation unit generates an initial trajectory prediction result based on the target motion state parameters;

[0038] The compensation trajectory determination unit dynamically corrects the initial trajectory prediction result based on the environmental disturbance parameters to obtain the disturbance compensation trajectory;

[0039] The target trajectory prediction unit performs behavioral pattern matching on the disturbance compensation trajectory based on the biological behavioral characteristic parameters to generate a target trajectory prediction result.

[0040] In a third aspect, an electronic device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor is configured to implement the steps of the motion target trajectory prediction method of any one of the first aspects when executing the computer program stored in the memory.

[0041] In a fourth aspect, the present application proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for predicting the trajectory of a moving target according to any one of the first aspects.

[0042] In summary, this application improves trajectory prediction accuracy in complex scenarios by integrating multi-source dynamic perception data and combining a dual optimization mechanism of dynamic correction of environmental disturbances with matching of biological behavioral patterns. Specifically, a compensation model constructed based on environmental disturbance parameters can offset external interference such as airflow resistance and terrain undulations in real time, while the introduction of biological behavioral characteristics can simulate the physiological constraints and behavioral preferences of the target, thereby achieving trajectory prediction that is more in line with actual motion patterns in fields such as drone navigation and intelligent security, effectively reducing the risk of path deviation and enhancing the system's adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present description. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0044] Figure 1 A schematic diagram of a moving target trajectory prediction method provided in an embodiment of the present application;

[0045] Figure 2 A schematic diagram of the structure of a moving target trajectory prediction device provided in an embodiment of the present application;

[0046] Figure 3 A structural diagram of an electronic device for predicting moving target trajectories provided in an embodiment of the present application. DETAILED DESCRIPTION

[0047] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or that are inherent to these processes, methods, products or devices. The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the embodiments described are only part of the embodiments of the present application, not all of the embodiments.

[0048] See also Figure 1 , which is a flow chart of a moving target trajectory prediction method provided in an embodiment of the present application, which may specifically include:

[0049] S110, acquiring multi-source dynamic perception data of a moving target, wherein the multi-source dynamic perception data includes target motion state parameters, environmental disturbance parameters, and biological behavior characteristic parameters;

[0050] For example, acquiring multi-source dynamic perception data of moving targets is the core foundation of trajectory prediction. It achieves a comprehensive perception of the target's motion characteristics by fusing target motion state parameters, environmental disturbance parameters, and biological behavioral characteristic parameters. Among them, target motion state parameters (such as speed, acceleration, and direction) directly reflect the target's instantaneous motion patterns; environmental disturbance parameters (such as airflow velocity and terrain undulation) characterize the dynamic interference of the external environment on the target's trajectory; and biological behavioral characteristic parameters (such as body surface temperature distribution and limb swing frequency) are used to reveal the target's behavioral patterns under physiological constraints, thereby providing multi-dimensional data support for prediction.

[0051] The synergistic effect of multi-source dynamic perception data overcomes the limitations of single-dimensional data. By combining millimeter-wave radar, a distributed air pressure sensor array, and infrared thermal imagers, the system can simultaneously capture the target's physical motion, environmental disturbances, and biological behavioral characteristics, forming a multi-dimensional data fusion framework. This integrated perception mechanism not only provides high-precision input for subsequent initial trajectory prediction but also lays the data foundation for environmental disturbance compensation and biological behavior pattern matching, thereby improving the robustness and adaptability of trajectory prediction in complex scenarios.

[0052] S120, generating an initial trajectory prediction result based on the target motion state parameters;

[0053] For example, initial trajectory predictions are generated based on the target's motion state parameters, aiming to construct a basic motion model using physical quantities such as the target's real-time velocity, acceleration, and direction of motion. These parameters directly reflect the target's instantaneous motion patterns. Using a preset kinematic model or neural network algorithm, they are mapped into a trajectory sequence for a period of time in the future, forming a preliminary prediction framework. These initial predictions serve as a benchmark for subsequent corrections and optimizations, providing the core input for environmental disturbance compensation and behavioral pattern matching.

[0054] The key to initial trajectory prediction lies in converting discrete motion state parameters into a mathematical representation of a continuous trajectory. By integrating the target's dynamic motion characteristics (such as the direction of acceleration changes), its short-term motion trends can be deduced. Using a neural network trained with historical data, potential correlations within complex motion patterns can be captured. This process not only establishes a physical and logical foundation for trajectory prediction but also provides an interface for subsequent dynamic corrections using multi-source data fusion, ensuring the predictive results are adaptable and scalable in complex scenarios.

[0055] S130, dynamically correcting the initial trajectory prediction result based on the environmental disturbance parameter to obtain a disturbance-compensated trajectory;

[0056] For example, dynamic corrections to the initial trajectory predictions are made based on environmental disturbance parameters. This aims to construct a disturbance compensation model by quantifying the impact of external environmental factors such as airflow velocity, terrain relief, and pressure gradient on target motion. This model calculates key parameters such as the fluid drag coefficient and the rate of change of terrain elevation to adjust the curvature and velocity components of the initial predicted trajectory in real time, thereby offsetting trajectory deviations caused by environmental disturbances and improving the physical plausibility of the prediction results.

[0057] The core of the dynamic correction process lies in the coupled analysis of environmental parameters and the target's motion state. By quantifying dynamic disturbances such as airflow resistance and terrain changes into computable compensation factors, the idealized assumptions in the initial trajectory prediction are iteratively optimized to generate a disturbance-compensated trajectory that better reflects actual environmental conditions. This mechanism effectively addresses the problem of cumulative prediction errors caused by traditional methods that ignore external disturbances, providing environmentally adaptive trajectory input for subsequent behavioral pattern matching.

[0058] S140 , performing behavioral pattern matching on the disturbance compensation trajectory based on the biological behavioral characteristic parameters to generate a target trajectory prediction result.

[0059] For example, behavioral pattern matching is performed on disturbance compensation trajectories based on biobehavioral characteristic parameters, aiming to analyze the target's movement preferences and behavioral patterns through its physiological characteristics (such as energy consumption rate and limb swing frequency). By comparing the body surface temperature distribution with a preset energy threshold, the target's energy reserve state can be inferred, thereby predicting the sustainability or direction of subsequent movement. At the same time, a motion rhythm map constructed based on limb swing frequency can map the target's periodic behavior patterns, providing a physiological constraint basis for trajectory correction.

[0060] The core of behavioral pattern matching lies in linking biological features with historical behavioral data to achieve personalized optimization of trajectory prediction. By matching motion rhythm patterns with highly similar reference trajectory segments from a historical behavioral pattern library, a trajectory correction strategy that aligns with the target's physiological characteristics is extracted. Combined with the correction strength determined by the energy consumption rate, disturbance compensation trajectories are then temporally and spatially superimposed. The resulting trajectory prediction retains the physical rationality of environmental disturbance correction while incorporating the dynamic adaptability driven by biological behavior, improving prediction accuracy in complex scenarios.

[0061] In summary, in the embodiment of the present application, by fusing the target motion state parameters, environmental disturbance parameters and biological behavior characteristic parameters, a multi-dimensional data-driven trajectory prediction framework is constructed, thereby improving the prediction accuracy and adaptability in complex scenarios. The initial trajectory prediction result is dynamically corrected based on the environmental disturbance parameters, effectively offsetting the influence of external interferences such as airflow resistance and terrain undulations on the target motion, and enhancing the robustness of the physical model; the behavioral pattern matching of the compensated trajectory is carried out in combination with the biological behavior characteristic parameters, and physiological constraints such as energy consumption rate and motion rhythm map are integrated into the prediction logic to achieve dual optimization from physical motion laws to biological behavior preferences. The embodiment of the present application breaks through the limitations of the traditional single data source. In the fields of autonomous navigation of drones, intelligent security tracking, etc., it can accurately predict sudden turns or speed changes and reduce the risk of path deviation. At the same time, through dynamic risk area marking and closed-loop feedback mechanism, it provides high-reliability data support for real-time obstacle avoidance decision-making and path planning, and has broad application value and technological innovation.

[0062] In some examples, obtaining multi-source dynamic perception data of a moving target includes:

[0063] The millimeter-wave radar is used to obtain the real-time speed, acceleration, and direction of movement of the moving target as the target motion state parameters;

[0064] The distributed air pressure sensor array collects the airflow velocity, pressure gradient and terrain undulation around the moving target as environmental disturbance parameters;

[0065] The surface temperature distribution and limb swing frequency of the moving target are extracted by infrared thermal imager as biological behavior characteristic parameters.

[0066] For example, a millimeter-wave radar transmits high-frequency electromagnetic waves in a specific frequency band and receives reflected signals from the target. Based on the Doppler effect and signal processing algorithms, it determines the target's real-time velocity, acceleration, and direction of motion. To measure real-time velocity, the radar analyzes the Doppler shift of the reflected signal and, combined with the frequency difference between the transmitted and received waves, directly calculates the target's velocity component in the radar's radial direction. To determine the target's direction of motion in three-dimensional space, the radar uses a multi-antenna array to receive signals. Using angle-of-arrival estimation techniques, it calculates the target's azimuth and elevation angles relative to the radar. Using a coordinate transformation algorithm, it decomposes the radial velocity into a velocity vector in three-dimensional space, thereby determining the target's direction of motion. Acceleration is determined by calculating velocity differences within a continuous time window. This involves capturing the target's instantaneous velocity sequence within a preset sampling period. The velocity rate of change is fitted using a least-squares method or Kalman filter algorithm, eliminating measurement noise, and outputting a smoothed acceleration value. These parameters rely on the timing synchronization and spatial calibration of the radar signal processing chain to ensure consistency of the target's motion state parameters in both time and space, providing input for subsequent trajectory prediction.

[0067] The distributed pressure sensor array uses multiple pressure sensor nodes distributed according to a preset topology to synchronously collect pressure data around the target, and combines the principles of fluid mechanics with spatial interpolation algorithms to determine the airflow velocity. Specifically, each node measures the air pressure value within the same time window, calculates the local pressure gradient through the pressure difference between adjacent nodes, and converts the pressure gradient into an airflow velocity vector based on the Bernoulli equation. To determine the direction of the airflow, the array analyzes the temporal correlation of air pressure changes between multiple nodes and combines the spatial position relationship to reconstruct the movement trend of the airflow in three-dimensional space. The determination of the pressure gradient depends on the spatial distribution density and measurement accuracy of the array nodes. The finite difference method is used to fit the discrete air pressure measurement values into a continuous air pressure field, and then calculate the rate of change of air pressure per unit distance, that is, the pressure gradient. Terrain relief is measured using an integrated laser rangefinder module or inertial measurement unit. The laser rangefinder emits a pulsed laser at the ground and receives the reflected signal. The vertical distance between the sensor node and the ground is calculated using the time-of-flight difference. This information, combined with the relative position data between the nodes, is then used to construct a terrain elevation model. The inertial measurement unit uses accelerometers and gyroscopes to detect changes in the sensor's attitude and, based on the initial elevation reference, deduce the terrain relief. These parameters are determined using a spatiotemporal synchronized calibration of the sensor array and a multi-source data fusion algorithm, ensuring high spatial resolution and physical consistency in the measurements of air velocity, pressure gradient, and terrain relief.

[0068] Infrared thermal imagers generate thermal images reflecting the temperature distribution by receiving infrared energy radiated from the target body surface. The surface temperature distribution is determined based on the radiation intensity data captured by the thermal imager's infrared sensor array. The radiation intensity at each pixel is converted to a corresponding temperature value using the blackbody radiation law, forming a two-dimensional temperature matrix. Discrete temperature points are smoothed using a spatial interpolation algorithm to eliminate environmental noise and measurement errors, and a continuous temperature distribution map is output. A preset physiological model further correlates the temperature distribution with the target metabolic rate. For example, high-temperature areas correspond to areas of high metabolic activity, while low-temperature areas represent areas of low energy consumption, thereby quantifying the spatial heterogeneity of surface temperature.

[0069] The limb sway frequency is determined by analyzing a continuous frame thermal imaging sequence. Within a preset sampling period, the thermal imager captures a dynamic thermal image sequence of the target limb and locates key temperature characteristic areas at the joints or limb extremities using a feature point recognition algorithm. Based on the displacement vectors of the characteristic areas between adjacent frames, the limb's sway amplitude and direction changes per unit time are calculated, and the dominant frequency component of the periodic motion signal is extracted using a Fourier transform or autocorrelation algorithm. By counting the number of sway cycles within a preset time window, the limb sway frequency is ultimately determined, characterizing the target's motion rhythmic characteristics. The analysis of these parameters is based on the thermal imager's temporal resolution and spatial calibration accuracy, ensuring the spatiotemporal consistency of the temperature distribution and sway frequency data.

[0070] The preset physiological model is an empirical or theoretical model based on the correlation between the known physiological characteristics of organisms and the surface temperature distribution, which is used to map the surface temperature data obtained by the thermal imager to the energy metabolism state. The model is set to show a positive correlation between surface temperature and internal metabolic activity, blood circulation intensity and tissue heat production efficiency. For example, high-temperature areas may correspond to areas with frequent muscle activity or high energy consumption (such as limb joints during exercise), while low-temperature areas may represent static tissues or energy reserve areas. Through the correspondence between temperature distribution and the measured value of metabolic rate under specific motion states, the model establishes a quantitative mapping rule between temperature gradient and energy consumption rate, thereby supporting the reverse inference of the target's real-time energy metabolism level from non-invasive temperature measurement, and providing physiological constraint parameters for trajectory prediction.

[0071] In some examples, generating an initial trajectory prediction result based on the target motion state parameters includes:

[0072] Calculate the instantaneous trajectory vector of the moving target based on real-time speed, acceleration and direction of movement;

[0073] Construct the target motion state matrix based on the instantaneous trajectory vector and the preset kinematic model;

[0074] The target motion state matrix is input into the trajectory prediction neural network to generate an initial trajectory prediction result. The trajectory prediction neural network is trained with historical motion data and is used to output a trajectory sequence that matches the current motion state.

[0075] For example, to determine the instantaneous trajectory vector of a moving target based on real-time velocity, acceleration, and direction of motion, the velocity and acceleration vectors are first decomposed and synthesized in a three-dimensional Cartesian coordinate system. The real-time velocity vector direction is determined by the azimuth and pitch angles measured by the millimeter-wave radar, which are converted into velocity components in three-dimensional space using trigonometric functions. The acceleration vector direction is derived based on the instantaneous rate of change of the velocity vector, and the acceleration component is calculated by velocity differentials within a continuous time window. The velocity and acceleration components are superimposed at each time step to generate an instantaneous trajectory vector, specifically represented as a linear combination of the velocity and acceleration vectors within a time increment. This vector is mapped from the current velocity and acceleration to the displacement increment at the next moment through integration or kinematic equations, thereby representing the instantaneous motion trend of the target in three-dimensional space. The instantaneous trajectory vector represents the instantaneous motion trend of the target at the current moment, including the combined effects of linear velocity, angular velocity, and acceleration, providing basic data for the subsequent construction of the motion state matrix.

[0076] When constructing the target motion state matrix, a preset kinematic model defines the target's motion constraints based on Newtonian mechanics or rigid body kinematics. For example, the uniform linear motion model assumes a constant target velocity, the uniformly accelerated motion model introduces an acceleration vector, and the curvilinear motion model combines angular velocity and curvature parameters. A multidimensional state matrix is generated by mapping the real-time velocity, acceleration, and direction of motion into the model parameter space. Specifically, the velocity component is decomposed into X-axis velocity, Y-axis velocity, and Z-axis velocity in a three-dimensional coordinate system; the acceleration component corresponds to the X-axis acceleration, Y-axis acceleration, and Z-axis acceleration; the azimuth angle and pitch angle are represented. Combined with model parameters (such as mass and moment of inertia), the matrix is constructed in a time series. Each row of the matrix corresponds to a sampling time point, and the column vectors sequentially contain the target's position coordinates, velocity components, acceleration components, azimuth angle, and model parameters in three-dimensional space, forming a structured dataset describing the target's motion characteristics.

[0077] The initial trajectory prediction is determined by inputting the target's motion state matrix into a trajectory prediction neural network, a deep learning model based on a long short-term memory network (LSTM) or a convolutional neural network (CNN). The trajectory prediction neural network is trained using historical motion data. The training data consists of the historical motion state matrix and the corresponding real trajectory sequence. The motion state matrix consists of parameters such as velocity, acceleration, and direction, while the trajectory sequence is a time series of position coordinates. During training, the network optimizes weight parameters using a backpropagation algorithm to learn the nonlinear mapping between the motion state parameters and the future trajectory. The loss function uses mean squared error (MSE) or dynamic time warping (DTW) to minimize the deviation between the predicted and actual trajectories. Model hyperparameters (such as the number of network layers, number of neurons, and learning rate) are determined through cross-validation to ensure a balance between generalization and prediction accuracy. During the prediction phase, the current target motion state matrix is used as input. After forward propagation through the neural network, the network outputs a sequence of trajectory point coordinates within a preset future time window, forming the initial trajectory prediction. This result contains the target's predicted position and velocity distribution in consecutive time steps, serving as the baseline trajectory for subsequent correction and optimization.

[0078] In some instances, the initial trajectory prediction result is dynamically corrected based on the environmental disturbance parameters to obtain a disturbance-compensated trajectory, including:

[0079] Determine the fluid resistance coefficient of the moving target based on the airflow velocity and movement direction;

[0080] Based on the terrain relief, determine the rate of change of terrain elevation;

[0081] Determine the disturbance compensation model based on the terrain elevation change rate and pressure gradient;

[0082] Based on the fluid resistance coefficient and the disturbance compensation model, the initial trajectory prediction results are dynamically corrected to determine the disturbance compensation trajectory.

[0083] Exemplarily, the determination of the fluid drag coefficient is based on the vector relationship between the airflow velocity and the target motion direction. The airflow velocity is collected by a distributed air pressure sensor array, and the three-dimensional airflow velocity field is reconstructed through pressure gradient analysis and spatial interpolation algorithm to obtain the airflow velocity vector around the target; the target motion direction is determined by the real-time velocity vector direction measured by the millimeter wave radar. The calculation of the fluid drag coefficient first calculates the angle between the airflow velocity vector and the target motion direction vector using the vector dot product formula, specifically the dot product of the two vectors divided by the arc cosine value of the product of their moduli, expressed as: in, is the angle; is the air flow velocity vector; is the target motion direction vector.

[0084] Based on this angle, combined with the Reynolds number and target geometric characteristic parameters, the fluid resistance coefficient is determined by a preset empirical model or a lookup table method. , expressed as:

[0085] in, is the Reynolds number, given by the air density , air flow velocity , target characteristic length (such as windward diameter or equivalent length) and air viscosity coefficient The target characteristic length is obtained by matching the target type with the preset database, such as the standard geometric parameters of drones, vehicles or organisms; the target geometric characteristics include the windward area, surface roughness and shape factor, which are obtained by matching the target type with the preset database. For example, the windward area of a spherical target is , and the frontal area of a streamlined object needs to be calculated based on a three-dimensional model; the preset empirical model associates the above parameters with the fluid drag coefficient, for example, for a target of a specific shape, the fluid drag coefficient The variation of Reynolds number and angle can be fitted into a piecewise function; the table lookup rule is realized by pre-generated multi-dimensional lookup table. The index in the table is Reynolds number interval, angle interval and target type, which stores the experimentally measured According to the real-time calculated Reynolds number, angle and target type, accurate interpolation or nearest neighbor matching is used to obtain The fluid drag coefficient comprehensively characterizes the obstruction effect of airflow direction, flow state and target geometric characteristics on motion, providing a quantitative resistance parameter input for the subsequent disturbance compensation model.

[0086] When determining the rate of change of terrain elevation based on terrain relief, discrete elevation data along the target's motion path is first acquired using a laser rangefinder module or an inertial measurement unit. The laser rangefinder emits a pulsed laser at the ground and receives the reflected signal, calculating the vertical distance between the sensor node and the ground. This is combined with the relative position coordinates between the nodes to generate a discrete set of elevation points. The inertial measurement unit detects changes in the sensor's own attitude using an accelerometer and gyroscope, and, based on the initial elevation reference, calculates a continuous sequence of elevation changes. This discrete elevation data is then used to construct a three-dimensional point cloud based on spatial coordinates. A digital elevation model (DEM) is generated using the Delaunay triangulation algorithm, forming a continuous representation of the terrain surface.

[0087] The calculation of the terrain elevation change rate is based on the elevation gradient in the target movement direction, the arc length sequence along the target movement path and the corresponding elevation value. The elevation difference and arc length difference of adjacent sampling points are calculated by the sliding window method, which is expressed as:

[0088] in, For the The elevation value of each sampling point; From the starting point to the The cumulative arc length of each sampling point. To eliminate measurement noise, a Kalman filter algorithm is used to smooth the raw differential results, outputting a stable sequence of elevation change rates. This parameter quantifies the dynamic changes in terrain slope along the target's motion direction, directly reflecting the real-time impact of terrain undulation on target motion energy consumption and trajectory curvature, providing precise terrain disturbance quantification input for subsequent disturbance compensation.

[0089] When determining the disturbance compensation model based on the terrain elevation change rate and the pressure gradient, the model is constructed by linearly superimposing the terrain disturbance component and the pressure disturbance component. The calculation of the terrain disturbance component is based on the terrain elevation change rate, which is obtained by the laser ranging module or the inertial measurement unit through discrete elevation data, combined with the digital elevation model and the sliding window difference algorithm to calculate the elevation gradient along the target movement direction. Terrain slope angle Depend on Determine and combine target quality and gravitational acceleration , through the formula Calculating terrain resistance components , characterizes the impact of terrain undulation on the energy consumption of target movement. The calculation of the pressure disturbance component is based on the spatial interpolation results of the distributed pressure sensor array to obtain the pressure gradient , its direction is consistent with the air flow velocity vector. By the formula Calculate, where The target effective force area is obtained by matching the preset database or real-time measurement. Finally, the disturbance compensation model superimposes the terrain resistance component and the pressure gradient force in the direction to generate a comprehensive disturbance compensation vector , expressed as:

[0090] in, is the unit vector of the slope direction; is the unit vector of the airflow direction; the model provides a physical compensation basis for trajectory correction by quantifying the dynamic interference of terrain and air pressure disturbances on target motion.

[0091] When dynamically correcting the initial trajectory prediction results based on the fluid resistance coefficient and disturbance compensation model, the fluid resistance coefficient is first Substitute into the air resistance formula to calculate the airflow resistance , expressed as: in, is the air density; the resistance characterizes the dynamic obstruction effect of airflow on the target movement.

[0092] The comprehensive disturbance compensation vector combined with the output of the disturbance compensation model , calculate the total disturbance force , expressed as: in, It is generated by the linear superposition of the terrain resistance component and the pressure gradient force, corresponding to the influence of the terrain elevation change rate and the pressure gradient respectively.

[0093] The total disturbance force is input into the kinematic equation, and the velocity component and acceleration component of the initial trajectory prediction result are iteratively corrected. Corrected speed and location Expressed as: in, is the initial predicted acceleration; the disturbance compensation trajectory is generated by updating the velocity and position parameters at each time step The trajectory is dynamically superimposed with comprehensive interference compensation for airflow resistance, terrain slope, and pressure gradient to eliminate errors caused by ignoring external disturbances in the initial prediction, thereby improving the physical rationality and prediction accuracy of the trajectory in complex environments.

[0094] In some instances, behavioral pattern matching is performed on the disturbance compensation trajectory based on biological behavioral characteristic parameters to generate a target trajectory prediction result, including:

[0095] Determine the energy consumption rate of the moving target based on the body surface temperature distribution;

[0096] Based on the limb swing frequency, a motion rhythm map of the movement target is constructed;

[0097] Determining a trajectory segment correction intensity based on a first comparison result of the energy consumption rate and a preset energy threshold;

[0098] Based on the motion rhythm map, matching the reference trajectory segment in the historical behavior pattern library, wherein the similarity between the reference trajectory segment and the motion rhythm map is greater than a preset similarity threshold;

[0099] Based on the trajectory segment correction strength and the reference trajectory segment, the disturbance compensation trajectory is spatiotemporally superimposed to generate the target trajectory prediction result.

[0100] For example, when determining the energy consumption rate of a moving target based on the surface temperature distribution, the infrared thermal imager captures the radiation intensity data of the target's body surface through its infrared sensor array. The radiation intensity of each pixel is converted into a corresponding temperature value through the blackbody radiation law to form a two-dimensional temperature matrix. The preset physiological model is constructed based on the experimental calibration data of the body temperature and metabolic rate. The local temperature value is converted into the energy consumption rate through the mapping relationship between the temperature gradient and the metabolic rate. Specifically, the model corresponds to the high-temperature area (such as the limb joints in motion) as the high metabolic activity area, and the low-temperature area (such as static tissue) as the low-energy consumption area. It also establishes a linear or nonlinear functional relationship between the temperature gradient and the local metabolic rate based on the spatial difference of the temperature distribution. After the temperature value of each pixel is mapped to the local energy consumption rate through this functional relationship, the metabolic rate of the temperature area of the entire image is integrated and calculated using a weighted average algorithm, and the overall energy consumption rate of the target is finally output. This parameter provides a physiological constraint basis for subsequent trajectory correction by quantifying the energy consumption intensity of the target in the current motion state.

[0101] When constructing a motion rhythm map of a moving target based on limb swing frequency, the periodic motion characteristics of the target limb are analyzed through a continuous frame thermal imaging sequence. An infrared thermal imager captures a dynamic thermal image sequence of the target limb at a preset sampling frequency, and a feature point recognition algorithm is used to locate high-temperature areas at joints or limb extremities. This high-temperature area is determined by analyzing the thermal imaging pixels of the body surface temperature distribution. In combination with a preset temperature threshold, feature points with temperatures significantly higher than those of surrounding tissue are selected to generate an initial set of feature points containing spatial coordinates.

[0102] Based on the initial set of feature points, the displacement vectors of these feature points between adjacent frames are calculated using optical flow or template matching algorithms. For each feature point, the change in its pixel coordinates across consecutive frames is extracted. By combining the thermal imager's spatial resolution and time interval, the pixel displacements are converted into actual physical displacement vectors. The direction and magnitude of the displacement vectors represent the directional change and intensity of the limb's swing, respectively, generating a time-series displacement vector dataset.

[0103] Based on the displacement vector dataset, a fast Fourier transform (FFT) is used to extract the dominant frequency component of the displacement sequence. The time-domain displacement amplitude sequence is converted into a frequency-domain signal, and the frequency corresponding to the dominant frequency peak is determined through spectral analysis. The dominant frequency extraction must meet a preset signal-to-noise ratio threshold to eliminate spurious peaks caused by noise interference. The dominant frequency value and the corresponding spectral energy distribution are then output.

[0104] Based on the dominant frequency and displacement sequence, an autocorrelation algorithm is used to verify periodicity. An autocorrelation calculation is performed on the displacement amplitude sequence to generate an autocorrelation function curve. Periodicity is determined based on the uniformity of the peak intervals and the decay rate of the peak amplitudes in the curve. If the autocorrelation curve exhibits periodic peaks and the decay rate is below a preset threshold, the displacement sequence is deemed periodic, and a periodicity verification flag and peak interval time parameters are output.

[0105] A motion rhythm spectrum is generated based on the dominant frequency value, periodicity verification results, and displacement vector data. The spectrum is stored as a time and frequency matrix, with the rows corresponding to the time series and the columns containing the dominant frequency value, mean displacement amplitude, and phase angle parameters. The phase angle is calculated from the initial phase offset of the displacement series, and the mean displacement amplitude is obtained by smoothing the modulus of the displacement vector using a sliding window method. The spectrum is constructed by filling data gaps using an interpolation algorithm to ensure continuity in the time dimension, and the output is a standardized motion rhythm spectrum.

[0106] The resulting motion rhythm map fully characterizes the periodicity of the target limb's swing, including frequency stability, amplitude fluctuation range, and phase synchronization. The map is stored as a structured multidimensional array, supporting efficient query and matching operations and providing standardized input data for subsequent behavioral pattern matching.

[0107] When determining the trajectory segment correction intensity based on the first comparison result of the energy consumption rate and the preset energy threshold, the preset energy threshold is set through historical motion data statistics, and different energy threshold intervals are set specifically for different biological types (such as humans and animals). For example, the energy threshold for human jogging is set to 5 kcal / min, while that for dog running is set to 3 kcal / min. The real-time energy consumption rate is obtained by inverse deduction from the body surface temperature distribution through a preset physiological model, and is compared with the corresponding threshold in real time. When the real-time rate exceeds the threshold, it is determined that the target is in a high energy consumption state and the exercise intensity needs to be reduced or the path needs to be adjusted, indicating that the greater the target energy consumption, the more reliance on historical behavior patterns is required; if it does not exceed, the current motion mode is maintained. The correction intensity α is quantified according to the percentage exceeding the threshold, expressed as: α=1-[(real-time energy consumption rate-preset energy threshold) / preset energy threshold]

[0108] If the real-time energy consumption rate is 120% of the threshold, it is linearly mapped to an intensity coefficient of 0.8 (range 0-1), indicating a low correction intensity and preserving the main body of the compensation trajectory. If it exceeds 50%, it is mapped to 0.5. If it exceeds 100%, α corresponds to 0, indicating a high correction intensity, fully referencing the historical trajectory. The intensity of the segmented trajectory correction directly determines the degree of intervention in subsequent trajectory corrections, ensuring the physical plausibility of the predicted trajectory within energy constraints and preventing the trajectory from deviating from the actual motion capability due to excessive correction.

[0109] When determining the reference trajectory segment, the historical motion data stored in the historical behavior pattern library includes the motion rhythm map and its corresponding trajectory sequence. The motion rhythm map records the periodic motion characteristics of the target in the form of a time and frequency matrix, including limb swing frequency, amplitude and phase information. During the matching process, the dynamic time warping (DTW) algorithm is used to calculate the time series alignment error between the current motion rhythm map and the historical rhythm map in the library, specifically by minimizing the cumulative distance between the two sequences on the time axis. A preset similarity threshold (for example, 0.8) is set as a screening condition, and only candidate trajectory segments with an alignment error lower than the threshold are retained.

[0110] To improve matching efficiency, a multidimensional indexing technique (such as an R-tree or hash table) is used to create an index structure for the frequency, amplitude, and phase characteristics of historical rhythm maps, prioritizing the search for entries that are closest to the current map in key dimensions. Finally, through similarity sorting and threshold filtering, the reference trajectory segment with the highest degree of match to the current motion rhythm map is determined, ensuring that the selected segment has reasonable motion rhythm characteristics.

[0111] Disturbance compensation trajectory based on trajectory segment correction strength and reference trajectory segment When performing spatiotemporal superposition, a weighted interpolation algorithm is used to fuse the compensation trajectory and the reference trajectory to generate the target trajectory prediction result. Specifically, the trajectory segment correction strength It is determined by comparing the energy consumption rate with the preset energy threshold, and its value range is 0 to 1. When it approaches 0, it indicates that the current trajectory segment needs to refer to the historical behavior pattern. The weight coefficient is , the dominant trajectory is superimposed; when the trajectory segment correction intensity When it approaches 1, it indicates that the current trajectory segment is more dependent on the environmental disturbance compensation result, and the disturbance compensation trajectory The weight coefficient is , retaining its main structure.

[0112] In the process of space-time superposition, the position, velocity and acceleration parameters of each trajectory point are expressed as: in, The disturbance compensation trajectory is generated by dynamically correcting the initial trajectory through the disturbance compensation model. It includes the compensation results of environmental disturbances such as airflow resistance, terrain undulations, and pressure gradients. The reference trajectory segment in the historical behavior pattern library whose similarity with the current motion rhythm map is greater than a preset threshold; is the time step The corresponding trajectory segment correction intensity is dynamically adjusted by the real-time comparison result of the energy consumption rate and the preset energy threshold. The curvature and velocity parameters of the reference trajectory segment are overwritten by the interpolation algorithm to cover the corresponding parameters of the compensation trajectory, such as directly replacing the curvature radius or adjusting the velocity direction; for low correction intensity For trajectory segments, the physical rationality of the compensation trajectory is retained, and only smooth transition corrections are performed on local path points.

[0113] The superimposed trajectory point sequence is globally smoothed using cubic spline interpolation or Kalman filtering to eliminate the curvature mutation or speed discontinuity caused by segmented correction. The final target trajectory prediction result is It is a time-series three-dimensional coordinate set containing the position, velocity, and acceleration information of each time step. This result not only integrates the physical motion law of environmental disturbance compensation, but also embeds the physiological constraints of historical behavior patterns, achieving dual optimization from physical dynamics to biological behavior, and improving the trajectory prediction accuracy and adaptability in complex scenarios. In UAV navigation, it is necessary to Adjust the flight path in real time to ensure that Avoid dynamic risk areas (such as areas with sudden airflow changes or terrain obstacles).

[0114] In some instances, this also includes:

[0115] Generate a trajectory visualization map based on the target trajectory prediction results, where the trajectory visualization map includes dynamic risk area markers;

[0116] Based on the dynamic risk area markings, determine the obstacle avoidance instructions or path optimization suggestions corresponding to the moving target.

[0117] For example, when generating a trajectory visualization map based on target trajectory prediction results, the three-dimensional coordinate sequence of the predicted trajectory is spatially mapped to dynamic risk zone parameters. Dynamic risk zone markings are determined through real-time analysis of environmental disturbance parameters (such as airflow velocity, terrain relief, and pressure gradient) and biological behavioral characteristic parameters (such as energy consumption rate). Specifically, areas where airflow velocity exceeds a preset safety threshold are marked as "high-risk airflow areas," areas where terrain relief results in a slope angle greater than a preset angle are marked as "terrain obstacle areas," and trajectory segments where energy consumption rates exceed physiological tolerance thresholds are marked as "high-energy risk segments." Risk levels are rendered in a three-dimensional geographic information system using color coding (e.g., red for high risk, yellow for medium risk, and green for safe areas), forming a visual map superimposed on the predicted trajectory. The construction of the map relies on a multi-source data fusion algorithm that spatially aligns data collected by millimeter-wave radar, distributed pressure sensor arrays, and infrared thermal imagers with the DEM to ensure that risk markers are strictly aligned with real geographic coordinates, providing intuitive visualization support for path decision-making.

[0118] When determining obstacle avoidance instructions or path optimization suggestions based on dynamic risk area markings, the trajectory optimization algorithm is used to replan the path of high-risk areas. For "high-risk airflow areas", the obstacle avoidance instruction generation module calculates the detour path based on the direction and intensity of the airflow, giving priority to adjacent areas where the airflow speed is lower than the safety threshold, and verifies the feasibility of the detour path through the kinematic model; for "terrain obstacle areas", the path optimization suggestion module combines the terrain elevation change rate and the target movement ability (such as maximum climbing angle, minimum turning radius) to generate a slope-adapted circuitous path; for "high-energy consumption risk sections", based on the energy consumption rate and remaining energy prediction, a deceleration or pause strategy is recommended to reduce the metabolic load. All instructions and suggestions are comprehensively evaluated through a decision tree algorithm to ensure that the obstacle avoidance strategy meets environmental safety, movement efficiency and physiological constraints at the same time. The final output is a structured control instruction containing coordinate offset, speed adjustment value and direction correction angle, which drives the moving target to perform trajectory adjustment in real time.

[0119] In some instances, this also includes:

[0120] Based on the target trajectory prediction results, the deviation between the actual motion path of the moving target and the target trajectory prediction results is monitored in real time;

[0121] When the deviation is greater than a first preset threshold, a re-collection instruction of the multi-source dynamic sensing data is triggered;

[0122] Based on the re-collection instruction, multi-source dynamic perception data is re-acquired.

[0123] For example, when monitoring the deviation between the actual motion path of a moving target and the predicted result of the target trajectory in real time, the actual motion coordinates of the target are continuously collected through a positioning device (such as a GPS or a visual positioning system), and are aligned in time and space with the coordinate sequence of the predicted trajectory. The deviation is calculated using a dynamic time warping algorithm, which compares the coordinate difference between the actual path and the predicted path point by point, and accumulates the overall deviation value. The specific process is: first, the timestamps of the actual path and the predicted path are aligned through an interpolation algorithm to ensure that each time point has corresponding actual and predicted coordinates; then, the Euclidean distance between each alignment point is calculated, all distances are accumulated and normalized, and a comprehensive deviation is obtained. The deviation reflects the consistency between the actual motion and the predicted trajectory, and the larger the value, the more significant the deviation.

[0124] When the deviation exceeds the first preset threshold, the re-collection instruction of the multi-source dynamic perception data is triggered. The first preset threshold is set according to the application scenario through historical data analysis. For example, it is set to 2 meters for the drone navigation scenario and 5 meters for the wildlife tracking scenario. The trigger logic is: the deviation of real-time monitoring is compared with the preset threshold. If it exceeds the threshold for multiple consecutive time steps (such as 3 sampling cycles), it is determined to be a continuous deviation and the data re-collection process is started immediately. The trigger instruction is sent synchronously to the millimeter wave radar, distributed air pressure sensor array and infrared thermal imager, forcing these sensors to re-collect the target motion state parameters (speed, acceleration, direction), environmental disturbance parameters (airflow velocity, air pressure gradient, terrain undulation) and biological behavior characteristic parameters (body surface temperature distribution, limb swing frequency) within a preset time window (such as 5 seconds before and after the current moment) to ensure that the time interval of sudden environmental changes or abnormal target behavior is covered.

[0125] After reacquiring multi-source dynamic perception data, perform data updates and trajectory corrections, including:

[0126] Millimeter-wave radar analyzes the Doppler shift and angle of arrival of the reflected signal to recalculate the target's real-time three-dimensional velocity, acceleration, and direction of motion, and construct an updated motion state matrix.

[0127] The air pressure sensor array synchronously collects air pressure data from each node, reconstructs the airflow velocity field through a spatial interpolation algorithm, and combines it with a laser ranging module to update the terrain relief and generate corrected environmental disturbance parameters.

[0128] The infrared thermal imager captures the latest surface thermal imaging sequence, extracts the limb swing frequency through feature point tracking, and inverts the energy consumption rate based on the preset physiological model to update the biological behavior characteristic parameters.

[0129] After spatial and temporal alignment and filtering to remove noise, this data is fed into the trajectory prediction model to regenerate the initial trajectory, disturbance-compensated trajectory, and final target trajectory. This closed-loop feedback mechanism effectively addresses dynamic environmental changes or sudden changes in target behavior through real-time data updates and trajectory corrections, improving prediction accuracy and system stability in complex scenarios.

[0130] See also Figure 2 , is a schematic diagram of the structure of a moving target trajectory prediction device provided in an embodiment of the present application, comprising:

[0131] A multi-source data acquisition unit 21 is used to acquire multi-source dynamic perception data of a moving target, wherein the multi-source dynamic perception data includes target motion state parameters, environmental disturbance parameters, and biological behavior characteristic parameters;

[0132] An initial trajectory generating unit 22 generates an initial trajectory prediction result based on the target motion state parameters;

[0133] The compensation trajectory determination unit 23 dynamically corrects the initial trajectory prediction result based on the environmental disturbance parameter to obtain the disturbance compensation trajectory;

[0134] The target trajectory prediction unit 24 performs behavior pattern matching on the disturbance compensation trajectory based on the biological behavior characteristic parameters to generate a target trajectory prediction result.

[0135] See also Figure 3 An embodiment of the present application also provides an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor. When the processor 320 executes the computer program 311, the steps of any method for predicting the trajectory of a moving target are implemented.

[0136] Since the electronic device introduced in this embodiment is a device used to implement a motion target trajectory prediction device in the embodiment of the present application, based on the method introduced in the embodiment of the present application, technical personnel in this field can understand the specific implementation of the electronic device of this embodiment and its various variations. Therefore, how the electronic device implements the method in the embodiment of the present application is no longer introduced in detail here. As long as the equipment used by technical personnel in this field to implement the method in the embodiment of the present application falls within the scope of protection to be protected by this application.

[0137] During the specific implementation process, when the computer program 311 is executed by the processor, any implementation method of the embodiments corresponding to the first aspect can be implemented.

[0138] It should be noted that, in the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0139] Those skilled in the art will appreciate that the embodiments of the present application may provide methods, systems, or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-readable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-readable program code.

[0140] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0141] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0142] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0143] The present application also provides a computer program product, which includes computer software instructions. When the computer software instructions are executed on a processing device, the processing device executes Figure 1 The process of a moving target trajectory prediction method in the corresponding embodiment.

[0144] A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via wired or wireless means. The computer-readable storage medium can be any available medium that a computer can store or a data storage device such as a server or data center that includes one or more available media integrated therein. The available medium can be a magnetic medium, an optical medium or a semiconductor medium, etc.

[0145] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0146] In the several embodiments provided in this application, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0147] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0148] In addition, the functional units in the various embodiments of the present application may be integrated into a processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The above-mentioned integrated units may be implemented in the form of hardware and / or software functional units.

[0149] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for enabling a computer device to execute all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: various media that can store program code, such as a USB flash drive, a mobile hard disk, a read-only memory, a magnetic disk, or an optical disk.

[0150] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

[0151] Although the preferred embodiments of this specification have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concepts. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of this specification.

[0152] Obviously, those skilled in the art may make various changes and modifications to this specification without departing from the spirit and scope of this specification. Thus, if such changes and modifications fall within the scope of the claims of this specification and their equivalents, this specification is intended to include such changes and modifications.

Claims

1. A moving target trajectory prediction method, characterized in that: include: Acquiring multi-source dynamic perception data of a moving target, wherein the multi-source dynamic perception data includes target motion state parameters, environmental disturbance parameters, and biological behavior characteristic parameters, wherein the biological behavior characteristic parameters include body surface temperature distribution and limb swing frequency; generating an initial trajectory prediction result based on the target motion state parameters; Based on the environmental disturbance parameters, dynamically correcting the initial trajectory prediction result to obtain a disturbance-compensated trajectory; Based on the biological behavior characteristic parameters, behavior pattern matching is performed on the disturbance compensation trajectory to generate a target trajectory prediction result; The performing behavior pattern matching on the disturbance compensation trajectory based on the biological behavior characteristic parameters to generate a target trajectory prediction result includes: determining an energy consumption rate of the moving target based on the body surface temperature distribution; constructing a motion rhythm map of the motion target based on the limb swing frequency; determining a trajectory segment correction intensity based on a first comparison result of the energy consumption rate and a preset energy threshold; Based on the motion rhythm map, matching a reference trajectory segment in a historical behavior pattern library, wherein the similarity between the reference trajectory segment and the motion rhythm map is greater than a preset similarity threshold; Based on the trajectory segment correction strength and the reference trajectory segment, the disturbance compensation trajectory is temporally and spatially superimposed to generate a target trajectory prediction result.

2. The method according to claim 1, characterized in that The acquiring of multi-source dynamic perception data of the moving target includes: The real-time speed, acceleration and direction of movement of the moving target are obtained by millimeter-wave radar as the target motion state parameters; The air flow velocity, pressure gradient and terrain undulation around the moving target are collected as the environmental disturbance parameters by a distributed air pressure sensor array; The surface temperature distribution and limb swing frequency of the moving target are extracted by an infrared thermal imager as the biological behavior characteristic parameters.

3. The method according to claim 2, characterized in that The generating of an initial trajectory prediction result based on the target motion state parameter includes: Calculating an instantaneous trajectory vector of the moving target based on the real-time speed, the acceleration, and the moving direction; Constructing a target motion state matrix according to the instantaneous trajectory vector and a preset kinematic model; The target motion state matrix is input into a trajectory prediction neural network to generate the initial trajectory prediction result, wherein the trajectory prediction neural network is trained through historical motion data and is used to output a trajectory sequence that matches the current motion state.

4. The method according to claim 2, characterized in that The dynamically correcting the initial trajectory prediction result based on the environmental disturbance parameter to obtain a disturbance-compensated trajectory includes: determining a fluid resistance coefficient of the moving object based on the airflow velocity and the moving direction; determining a rate of change of terrain elevation based on the terrain relief; determining a disturbance compensation model based on the terrain elevation change rate and the air pressure gradient; Based on the fluid resistance coefficient and the disturbance compensation model, the initial trajectory prediction result is dynamically corrected to determine a disturbance compensation trajectory.

5. The method according to claim 1, wherein Also includes: Based on the target trajectory prediction result, generating a trajectory visualization map, wherein the trajectory visualization map includes dynamic risk area markings; Based on the dynamic risk area mark, an obstacle avoidance instruction or a path optimization suggestion corresponding to the moving target is determined.

6. The method according to claim 1, characterized in that Also includes: Based on the target trajectory prediction result, real-time monitoring of the deviation between the actual motion path of the moving target and the target trajectory prediction result; When the deviation is greater than a first preset threshold, triggering a re-collection instruction of the multi-source dynamic sensing data; Based on the re-acquisition instruction, the multi-source dynamic perception data is reacquired.

7. A moving target trajectory prediction device, characterized in that: include: a multi-source data acquisition unit, configured to acquire multi-source dynamic perception data of a moving target, wherein the multi-source dynamic perception data includes target motion state parameters, environmental disturbance parameters, and biological behavior characteristic parameters, wherein the biological behavior characteristic parameters include body surface temperature distribution and limb swing frequency; An initial trajectory generating unit, which generates an initial trajectory prediction result based on the target motion state parameters; a compensation trajectory determination unit, which dynamically corrects the initial trajectory prediction result based on the environmental disturbance parameter to obtain a disturbance compensation trajectory; a target trajectory prediction unit, which performs behavior pattern matching on the disturbance compensation trajectory based on the biological behavior characteristic parameters to generate a target trajectory prediction result; The performing behavior pattern matching on the disturbance compensation trajectory based on the biological behavior characteristic parameters to generate a target trajectory prediction result includes: determining an energy consumption rate of the moving target based on the body surface temperature distribution; constructing a motion rhythm map of the motion target based on the limb swing frequency; determining a trajectory segment correction intensity based on a first comparison result of the energy consumption rate and a preset energy threshold; Based on the motion rhythm map, matching a reference trajectory segment in a historical behavior pattern library, wherein the similarity between the reference trajectory segment and the motion rhythm map is greater than a preset similarity threshold; Based on the trajectory segment correction strength and the reference trajectory segment, the disturbance compensation trajectory is temporally and spatially superimposed to generate a target trajectory prediction result.

8. An electronic device comprising: A memory, a processor, and a computer program stored in the memory and runnable on the processor, wherein the processor is configured to implement the steps of the moving target trajectory prediction method according to any one of claims 1 to 6 when executing the computer program stored in the memory.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the moving target trajectory prediction method according to any one of claims 1 to 6 is implemented.

Images