Method and system for trajectory prediction of moving objects based on monte carlo method

Abstract

The application discloses a moving object trajectory prediction method and system based on a Monte Carlo method, and belongs to the field of trajectory prediction; the method comprises the following steps: collecting historical trajectory data of a vehicle, acquiring environment information, and performing data cleaning processing and feature extraction; determining a vehicle operation mode, an acceleration model and environment influence factors according to the historical trajectory data and the environment information after feature extraction, and constructing a motion model; based on the motion model, a Monte Carlo method is adopted to perform multiple simulation experiments on a future trajectory of the vehicle, different initial conditions and environment influence factors are considered, and a series of trajectory samples are generated; based on the trajectory samples, a future trajectory distribution of the vehicle is estimated, and a position and a probability distribution of the vehicle possibly appearing at each moment are obtained; a position confidence interval is calculated according to the characteristics of a Gaussian distribution, and the future trajectory of the vehicle is predicted in combination with the estimation result of the future trajectory distribution of the vehicle.

Description

Technical Field

[0001] This invention belongs to the field of trajectory prediction technology, specifically relating to a method and system for predicting the trajectory of moving objects based on the Monte Carlo method. Background Technology

[0002] With the rapid development of intelligent transportation systems and autonomous driving technology, the demand for vehicle trajectory prediction is increasing. Existing trajectory prediction methods are usually based on statistical models or machine learning algorithms, which rely on a large amount of labeled data and prior assumptions about the model structure (Chen Xinping. Research on Trajectory Prediction Method of Moving Object Based on Multi-Feature Fusion of Deep Learning [D]. Beijing University of Posts and Telecommunications, 2021. DOI:10.26969 / d.cnki.gbydu.2021.002529.). Therefore, a new method is needed that can accurately predict the future trajectory of vehicles using historical data.

[0003] Meanwhile, the Monte Carlo method, as a powerful numerical simulation technique, has been widely applied in many fields, including finance, physics, and biology (Zhu Lulu. Monte Carlo Method and Its Applications [D]. Central China Normal University, 2014.). In the field of transportation, the Monte Carlo method has also proven to be an effective tool for handling uncertainty problems and conducting traffic simulation (Wang Y, Liu Z, Zuo Z, et al. Trajectory Planning and Safety Assessment of Autonomous Vehicles Based on Motion Prediction and ModelPredictive Control [J]. IEEE Transactions on Vehicular Technology, 2019, 68(99):8546-8556.DOI:10.1109 / TVT.2019.2930684.). However, applying the Monte Carlo method to vehicle trajectory prediction still presents challenges, requiring solutions on how to effectively utilize historical data and simulation techniques to improve prediction accuracy and efficiency. Summary of the Invention

[0004] To address the shortcomings of existing technologies, the present invention aims to provide a method and system for predicting the trajectory of moving objects based on the Monte Carlo method, thereby solving the problems in the prior art.

[0005] The objective of this invention can be achieved through the following technical solutions: Monte Carlo-based methods for predicting the trajectory of moving objects include: Collect historical trajectory data of vehicles, obtain environmental information, and perform data cleaning and feature extraction; Based on the historical trajectory data and environmental information after feature extraction, the vehicle operation mode, acceleration model and environmental influencing factors are determined, and a motion model is constructed. Based on the motion model, the Monte Carlo method is used to conduct multiple simulation experiments on the future trajectory of the vehicle, taking into account different initial conditions and environmental influencing factors, and generating a series of trajectory samples. Based on trajectory samples, the future trajectory distribution of vehicles is estimated, and the possible locations of vehicles at each time point and their probability distribution are obtained. The location confidence interval is calculated based on the characteristics of the Gaussian distribution, and the future trajectory of the vehicle is predicted by combining the estimation results of the future trajectory distribution of the vehicle.

[0006] Furthermore, the data cleaning process includes: removing outliers and noise from the collected historical trajectory data, and performing spatiotemporal interpolation to fill in the missing data.

[0007] Furthermore, features extracted from historical trajectory data include: location, speed, acceleration, and steering angle; features extracted from environmental information include: road type and traffic flow.

[0008] Furthermore, the motion model is as follows: in, , and Let represent the vehicle's position, speed, and step size at time t, respectively. This represents the vehicle's acceleration, which is a non-linear function. This indicates external input, taking into account environmental factors.

[0009] Furthermore, the steps for simulating the future trajectory of the vehicle include: S31, randomly select an initial state from historical trajectory data. ; S32, randomly sample each parameter from the given parameter space, including the control input u and the process noise w; S33, Based on the aforementioned motion model, the future motion of the vehicle is simulated using the results of parameter sampling: S34. Multiple simulation experiments were conducted, each using different random perturbations and initial states, resulting in multiple trajectory samples. .

[0010] Furthermore, the step of estimating the future trajectory distribution of the vehicle includes: S41, preprocess each trajectory sample, including: outlier detection and data interpolation. S42, Perform statistical analysis on the preprocessed trajectory samples and calculate the mean of each trajectory sample. Covariance ; S43, Based on the statistical analysis results, assume that the preprocessed trajectory samples follow a multivariate Gaussian distribution, and X is the position vector of the trajectory sample. It is the mean vector. Let be the covariance matrix, and d be the trajectory dimension. The Gaussian probability density function is constructed as follows: S44, using maximum likelihood estimation to estimate Gaussian distribution parameters and ; S45, the estimated and Substituting the Gaussian probability density function, we obtain the trajectory distribution for each trajectory sample.

[0011] Furthermore, the process of calculating the location confidence interval includes: Choosing 95% as the confidence level, then the parameter ; Calculate the critical value of the chi-square distribution at the corresponding confidence level. : Where d is the trajectory dimension; ICDF represents the inverse cumulative distribution function of the chi-square distribution, and its mathematical expression is: Where v represents the degrees of freedom, Let x represent the gamma function and x represent the position vector X. t = (x t1 , x t2 , …, x tᵈ A certain observation point; The radius of the confidence interval is: Confidence intervals are defined as all conditions that satisfy the following inequality: Where y represents the position vector X t The observed values.

[0012] A moving object trajectory prediction system based on the Monte Carlo method includes: Data collection and processing module: collects historical trajectory data of vehicles, obtains environmental information, and performs data cleaning and feature extraction; Model building module: Based on historical trajectory data and environmental information after feature extraction, determine the vehicle operation mode, acceleration model and environmental influencing factors, and build a motion model; Trajectory Sample Generation Module: Based on the motion model, the Monte Carlo method is used to conduct multiple simulation experiments on the future trajectory of the vehicle, taking into account different initial conditions and environmental influencing factors, and generating a series of trajectory samples; Trajectory distribution estimation module: Based on trajectory samples, it estimates the distribution of the vehicle's future trajectory, obtaining the possible location of the vehicle at each time point and its probability distribution; Trajectory prediction module: Calculates the position confidence interval based on the characteristics of Gaussian distribution, and predicts the future trajectory of the vehicle by combining the estimation results of the future trajectory distribution of the vehicle.

[0013] A computer storage medium storing a readable program that, when executed, instructs a computing device to perform a moving object trajectory prediction method based on the Monte Carlo method as described above.

[0014] An electronic device includes: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other through the communication bus; The memory is used to store at least one executable instruction that causes the processor to perform operations corresponding to the Monte Carlo method-based moving object trajectory prediction method described above.

[0015] The beneficial effects of this invention are: 1. Through extensive random sampling, the model of this invention can generate a vehicle state space that closely approximates the distribution in the real world, thereby providing high-precision trajectory prediction.

[0016] 2. This invention has a strong uncertainty handling capability. The Monte Carlo method can effectively handle uncertainties such as road conditions, traffic flow and driving behavior, ensuring the reliability of prediction results.

[0017] 3. This invention is easy to integrate multi-source data. The model design allows for easy integration of data from different sensors, such as GPS, radar and cameras, enhancing the comprehensiveness of state estimation.

[0018] 4. This invention does not rely on a large amount of labeled data and is applicable to various traffic scenarios and vehicle types. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart of the moving object trajectory prediction method based on the Monte Carlo method of the present invention; Figure 2 This is a schematic diagram of the original position image of the moving object according to the present invention; Figure 3 This is a trajectory distribution diagram from the Monte Carlo simulation of the present invention; Figure 4 This is a schematic diagram comparing the predicted trajectory with the actual trajectory of the present invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Example 1 In this embodiment, the vehicle is taken as the moving object, such as Figure 1 As shown, the moving object trajectory prediction method based on the Monte Carlo method includes the following steps: S1 collects historical trajectory data of vehicles, obtains environmental information, and performs data cleaning and feature extraction. S11, Historical Trajectory Data Collection: a. Real-time collection of vehicle location data on the road network through onboard sensors, GPS devices, traffic monitoring cameras, and other equipment; b. Data collection should cover different types of roads (such as urban roads, highways, etc.) and different driving scenarios (such as congested road sections, high-speed driving, etc.) to ensure the diversity and representativeness of the data; c. Use in-vehicle sensors to record parameters such as the vehicle's driving status, speed, and acceleration; S12, Obtaining environmental information: a. Collect geographic information about the road network, including road type, number of lanes, speed limit signs, and traffic light locations; b. Obtain traffic flow data, including vehicle density, speed distribution, and other information, through traffic monitoring systems or data provided by traffic management departments; c. Collect weather information, such as rainfall and visibility.

[0023] S13, Data Cleaning Processing: The collected historical trajectory data is cleaned and processed to remove outliers and noise, ensuring data quality and accuracy. Spatiotemporal interpolation is performed on the historical trajectory data to fill in missing data points, meeting the needs of subsequent motion model building and prediction. S14, Feature Extraction Features such as location, speed, acceleration, and steering angle are extracted from historical trajectory data; environmental features such as road type and traffic flow are extracted from environmental information. Feature extraction should consider a comprehensive description of vehicle behavior and environmental impact so that the subsequent motion model can fully account for the influence of various factors on trajectory prediction.

[0024] S2. Based on the historical trajectory data and environmental information after feature extraction, determine the vehicle operation mode, acceleration model and environmental influencing factors, and construct a motion model. The process of building a motion model includes: S21. First, based on historical trajectory data and environmental information, the initial state of the simulation needs to be determined, including the vehicle's initial position, speed, acceleration, etc. Multiple initial states can be set according to actual conditions to cover different driving situations and environmental conditions.

[0025] The motion pattern of a vehicle can usually be represented by a state equation; for example, for uniform linear motion, the state equation can be expressed as: In the formula, , and Let represent the vehicle's position, speed, and step size at time t, respectively; S22, Acceleration Model Settings: To determine the vehicle's acceleration model, one can choose uniform motion, uniformly accelerated motion, or other more complex motion models. These can typically be represented by difference equations; for example, the difference equation for uniformly accelerated motion can be expressed as: Among them, vehicle acceleration It is a non-linear function; S23, consider environmental factors through external input terms in the state equation. For example, if the impact of road conditions is considered, road conditions can be represented as an external input. The state equation then becomes ; S24, Uncertainty Handling: Uncertainty is simulated using random variables. For example, the acceleration of a vehicle. It can be represented as a random variable that follows a certain distribution: in, This represents the expected value of vehicle acceleration. Indicates the standard deviation of acceleration; S3, based on a motion model, uses the Monte Carlo method to conduct multiple simulation experiments on the future trajectory of the vehicle, considering different initial conditions and environmental influencing factors, and generates a series of trajectory samples; Each simulation represents a single vehicle movement, starting from the initial state and proceeding according to a pre-defined motion pattern and acceleration model, with a set time step. Within each time step, based on the current state and environmental information, the vehicle's position, velocity, acceleration, and other parameters for the next moment are calculated. Random perturbations are introduced to simulate environmental uncertainties and data noise. These random perturbations can be represented as either a normal distribution or a uniform distribution. In this embodiment, the original location image of the vehicle is as follows: Figure 2 As shown, a detailed model including road network, traffic signals and surrounding environment is constructed; by randomly selecting an initial state from historical trajectory data and simulating the motion process according to the established motion model, all trajectory samples including noise can be obtained.

[0026] The steps for simulating the future trajectory of a vehicle are as follows: S31, Select initial state: Randomly select an initial state from historical trajectory data. ; S32, Parameter Sampling; Randomly sample each parameter from the given parameter space, including the control input u and the process noise w; S33, Simulation of Motion Process: Based on the established motion model, the future motion of the vehicle is simulated using the results of parameter sampling. S34. Multiple simulation experiments were conducted, each using different random perturbations and initial states. After the simulation experiments were completed, multiple possible trajectory samples were obtained. .

[0027] S4, based on trajectory samples, estimates the distribution of future vehicle trajectories to obtain the possible locations of vehicles at each time point and their probability distribution; The steps for estimating the future trajectory distribution of vehicles are as follows: S41, after N Monte Carlo simulations, a series of possible future trajectory samples of the vehicle can be obtained. Let the trajectory sample of the i-th simulation be... First, it is necessary to process each trajectory sample. Preprocessing operations such as outlier detection and data interpolation are performed to ensure the quality and reliability of the trajectory data. Preprocessing operations include: a. Outlier detection: Use statistical methods to remove outliers.

[0028] in, The mean of the data represents the outlier detection threshold coefficient, which is used to control the sensitivity of outlier detection. This represents the standard deviation of the trajectory data samples; b. Data interpolation: Data interpolation is performed using the basis functions of cubic spline interpolation. Given a set of known data points... ,in If we represent time and location vectors, then cubic spline interpolation can be expressed by the following formula: In the formula, Indicates the first Basis functions for cubic spline interpolation within the interval, , , , The coefficients within this interval need to be solved using boundary and continuity conditions; S42, Trajectory Data Analysis Preprocessed trajectory samples Perform statistical analysis and calculate the mean of each trajectory sample. Covariance Equal statistics.

[0029] S43, Trajectory Distribution Modeling Based on the statistical analysis results, assuming the preprocessed trajectory samples It follows a multivariate Gaussian distribution, where X is the position vector of the trajectory sample. It is the mean vector. Let be the covariance matrix, and d be the trajectory dimension. Then the Gaussian probability density function is as follows: S44, Parameter Estimation Estimating Gaussian distribution parameters using maximum likelihood and Where N is the total number of trajectory points; S45, Trajectory Distribution Estimation: The estimated and Substituting the Gaussian probability density function in S43, we can obtain the trajectory distribution of each trajectory sample, that is, the possible location of the vehicle at each time point and its probability distribution. In this embodiment, the trajectory distribution is as follows: Figure 3 As shown, by Figure 3 It can be seen that the trajectory samples obtained through multiple Monte Carlo simulations exhibit obvious probability clustering characteristics in space. High-probability areas are concentrated near the main path that the vehicle may travel in the future, while low-probability areas are distributed at the edge of the trajectory.

[0030] The above distribution characteristics indicate that the present invention can effectively characterize the uncertainty range of the future trajectory of a vehicle and reflect the probability of different locations by means of a probability distribution, providing a reliable basis for subsequent confidence interval calculation and trajectory prediction.

[0031] Results verification and adjustment: The results were verified and the model was adjusted using residual analysis.

[0032] a. Calculate the residuals: for each observation... and predicted value Calculate the residual This refers to the difference between the observed value and the predicted value. b. Plot the residuals: Plot the residuals... Along the horizontal axis, predicted values Plot a scatter plot along the vertical axis. Identify outliers and exceptions in the plot. c. Based on the results of the residual analysis, adjust or improve the model to enhance its fit and prediction accuracy.

[0033] S5. Calculate the position confidence interval based on the characteristics of the Gaussian distribution, and combine it with the estimation results of the vehicle's future trajectory distribution to predict the vehicle's future trajectory.

[0034] When predicting the future trajectory of a vehicle, the position confidence interval is calculated based on the characteristics of the Gaussian distribution. The calculation process is as follows: Choosing 95% as the confidence level, then the parameter ; Calculate the critical value of the chi-square distribution at the corresponding confidence level. Where d is the trajectory dimension: Wherein, ICDF represents the inverse cumulative distribution function of the chi-square distribution, and its mathematical expression is as follows: Where v represents the degrees of freedom, Let x represent the gamma function and x represent the position vector X. t =(x t1 , x t2, …, x tᵈ A certain observation point; The radius of the confidence interval can be expressed as: The confidence ellipse can be represented by all inequalities that satisfy the following: in, The statistical mean parameter of the trajectory distribution is independent of the aforementioned external input variables in both meaning and function.

[0035] By following the steps above, the position confidence interval at time t can be calculated. This confidence interval represents the possible range of future positions of the moving object, thereby improving the accuracy and reliability of trajectory prediction. Combined with trajectory distribution information, the future driving path (trajectory) of the vehicle can be predicted, such as... Figure 4 As shown, by following the above steps, the position confidence interval at time t can be calculated. The confidence interval is used to represent the range of possible positions of the moving object at future times, thereby providing a probabilistic constraint for trajectory prediction.

[0036] Based on this, and combined with the trajectory distribution estimation results, the future driving path of the vehicle is predicted, and the prediction results are as follows: Figure 4 As shown. By Figure 4 It can be seen that when the vehicle enters the deceleration phase, the spatial change trend of the predicted trajectory remains consistent with the actual motion state of the vehicle, with the trajectory spacing gradually decreasing and the magnitude of positional change slowing down. These results demonstrate that, by introducing trajectory distribution and confidence interval constraints, this invention can reasonably predict the future motion trend of the vehicle under varying speeds, especially in deceleration scenarios, avoiding abrupt trajectory changes or divergence, thereby improving the stability and reliability of the trajectory prediction results.

[0037] Based on a similar inventive concept, embodiments of the present invention also provide a computer storage medium storing a readable program that, when run by a processor, can execute the aforementioned method for predicting the trajectory of a moving object based on the Monte Carlo method.

[0038] Based on a similar inventive concept, this invention provides an electronic device, including: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other through the communication bus; The memory is used to store at least one executable instruction, which causes the processor to perform the operation corresponding to the moving object trajectory prediction method based on the Monte Carlo method described above.

[0039] Based on a similar inventive concept, embodiments of the present invention also provide a computer program product, including computer instructions, which instruct a computing device to perform the operations corresponding to the above-described Monte Carlo method-based moving object trajectory prediction method.

[0040] Example 2 Based on the moving object trajectory prediction method based on the Monte Carlo method proposed in Example 1, this example proposes a moving object trajectory prediction system based on the Monte Carlo method, specifically including: Data collection and processing module: collects historical trajectory data of vehicles, obtains environmental information, and performs data cleaning and feature extraction; Model building module: Based on historical trajectory data and environmental information after feature extraction, determine the vehicle operation mode, acceleration model and environmental influencing factors, and build a motion model; Trajectory Sample Generation Module: Based on the motion model, the Monte Carlo method is used to conduct multiple simulation experiments on the future trajectory of the vehicle, taking into account different initial conditions and environmental influencing factors, and generating a series of trajectory samples; Trajectory distribution estimation module: Based on trajectory samples, it estimates the distribution of the vehicle's future trajectory, obtaining the possible location of the vehicle at each time point and its probability distribution; Trajectory prediction module: Calculates the position confidence interval based on the characteristics of Gaussian distribution, and predicts the future trajectory of the vehicle by combining the estimation results of the future trajectory distribution of the vehicle.

[0041] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0042] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention.

Claims

1. A method for predicting the trajectory of a moving object based on the Monte Carlo method, characterized in that, include: Collect historical trajectory data of vehicles, obtain environmental information, and perform data cleaning and feature extraction; Based on the historical trajectory data and environmental information after feature extraction, the vehicle operation mode, acceleration model and environmental influencing factors are determined, and a motion model is constructed. Based on the motion model, the Monte Carlo method is used to conduct multiple simulation experiments on the future trajectory of the vehicle, taking into account different initial conditions and environmental influencing factors, and generating a series of trajectory samples. Based on trajectory samples, the future trajectory distribution of vehicles is estimated, and the possible locations of vehicles at each time point and their probability distribution are obtained. The location confidence interval is calculated based on the characteristics of the Gaussian distribution, and the future trajectory of the vehicle is predicted by combining the estimation results of the future trajectory distribution of the vehicle.

2. The method for predicting the trajectory of a moving object based on the Monte Carlo method according to claim 1, characterized in that, The data cleaning process includes: removing outliers and noise from the collected historical trajectory data, and performing spatiotemporal interpolation to fill in the missing data.

3. The method for predicting the trajectory of a moving object based on the Monte Carlo method according to claim 1, characterized in that, Features extracted from historical trajectory data include: location, speed, acceleration, and steering angle; features extracted from environmental information include: road type and traffic flow.

4. The method for predicting the trajectory of a moving object based on the Monte Carlo method according to claim 1, characterized in that, The motion model is as follows: in, , and Let represent the vehicle's position, speed, and step size at time t, respectively. This represents the vehicle's acceleration, which is a non-linear function. This indicates external input, taking into account environmental factors.

5. The method for predicting the trajectory of a moving object based on the Monte Carlo method according to claim 4, characterized in that, The steps involved in simulating the future trajectory of a vehicle include: S31, randomly select an initial state from historical trajectory data. ; S32, randomly sample each parameter from the given parameter space, including the control input u and the process noise w; S33, Based on the aforementioned motion model, the future motion of the vehicle is simulated using the results of parameter sampling: S34. Multiple simulation experiments were conducted, each using different random perturbations and initial states, resulting in multiple trajectory samples. .

6. The method for predicting the trajectory of a moving object based on the Monte Carlo method according to claim 1, characterized in that, The steps for estimating the future trajectory distribution of the vehicle include: S41, preprocess each trajectory sample, including: outlier detection and data interpolation. S42, Perform statistical analysis on the preprocessed trajectory samples and calculate the mean of each trajectory sample. Covariance ; S43, Based on the statistical analysis results, assume that the preprocessed trajectory samples follow a multivariate Gaussian distribution, and X is the position vector of the trajectory sample. It is the mean vector. Let be the covariance matrix, and d be the trajectory dimension. The Gaussian probability density function is constructed as follows: S44, using maximum likelihood estimation to estimate Gaussian distribution parameters and ; S45, the estimated and Substituting the Gaussian probability density function, we obtain the trajectory distribution for each trajectory sample.

7. The method for predicting the trajectory of a moving object based on the Monte Carlo method according to claim 6, characterized in that, The process of calculating the location confidence interval includes: Choosing 95% as the confidence level, then the parameter ; Calculate the critical value of the chi-square distribution at the corresponding confidence level. : Where d is the trajectory dimension; ICDF represents the inverse cumulative distribution function of the chi-square distribution, and its mathematical expression is: Where v represents the degrees of freedom, Let x represent the gamma function and x represent the position vector X. t = (x t1 , x t2 , …, x tᵈ A certain observation point; The radius of the confidence interval is: Confidence intervals are defined as all conditions that satisfy the following inequality: Where y represents the position vector X t The observed values.

8. A moving object trajectory prediction system based on the Monte Carlo method, comprising the method described in any one of claims 1-7, characterized in that, include: Data collection and processing module: collects historical trajectory data of vehicles, obtains environmental information, and performs data cleaning and feature extraction; Model building module: Based on historical trajectory data and environmental information after feature extraction, determine the vehicle operation mode, acceleration model and environmental influencing factors, and build a motion model; Trajectory Sample Generation Module: Based on the motion model, the Monte Carlo method is used to conduct multiple simulation experiments on the future trajectory of the vehicle, taking into account different initial conditions and environmental influencing factors, and generating a series of trajectory samples; Trajectory distribution estimation module: Based on trajectory samples, it estimates the distribution of the vehicle's future trajectory, obtaining the possible location of the vehicle at each time point and its probability distribution; Trajectory prediction module: Calculates the position confidence interval based on the characteristics of Gaussian distribution, and predicts the future trajectory of the vehicle by combining the estimation results of the future trajectory distribution of the vehicle.

9. A computer storage medium storing a readable program, characterized in that, When the program runs, it can instruct the computing device to perform the moving object trajectory prediction method based on the Monte Carlo method as described in any one of claims 1-7.

10. An electronic device, characterized in that, include: The processor, memory, communication interface, and communication bus are provided, wherein the processor, memory, and communication interface communicate with each other via the communication bus. The memory is used to store at least one executable instruction, which causes the processor to perform the operation corresponding to the moving object trajectory prediction method based on the Monte Carlo method as described in any one of claims 1-7.

Images