A vehicle driving assistance system and control method based on digital twin

Abstract

The present application discloses a vehicle driving assistance system and a control method based on digital twin. The vehicle driving assistance system provided by the present application includes: a vehicle following subsystem, a vehicle scanning subsystem, and a control subsystem. When the vehicle following subsystem detects the approach of a target vehicle, the vehicle sensing device can be controlled to follow the target vehicle along the target road section through a guide rail and a driving mechanism, so as to realize data tracking and collection of the target vehicle, ensure stable collection of target vehicle information, synchronize the collected vehicle state data and the vehicle contour space data collected by the vehicle scanning subsystem to the control subsystem, and the control subsystem updates the traffic digital twin model corresponding to the target road section. The twin model generates vehicle driving assistance information and feeds it back to the target vehicle, thereby providing a vehicle driving assistance service with higher accuracy and stability for the target vehicle, and solving the technical problem that the existing method is prone to inaccuracy when facing complex road conditions.

Description

Technical Field

[0001] This application relates to the field of digital transportation technologies, and in particular, to a vehicle driving assistance system and a control method based on digital twins. Background Art

[0002] Road traffic is more complex than other modes of transportation. Road traffic usually includes various types of vehicles, and the parallel, lane-changing, and overtaking behaviors of external vehicles will all affect the normally driving vehicles. At the same time, road conditions also have a great impact on the normal operation of vehicles. Factors such as road conditions, quality, and capacity will increase the complexity of road traffic.

[0003] Currently, vehicle data on the road can be collected through sensing devices such as cameras and radars, and the running states of vehicles at the road scene can be simulated using intelligent algorithms or digital twin models, and then driving assistance information is generated to assist drivers in making more appropriate driving decisions. However, the existing methods are prone to the technical problem of inaccurate driving assistance information when facing complex road conditions with heavy traffic. Summary of the Invention

[0004] This application provides a vehicle driving assistance system and a control method based on digital twins, which are used to solve the technical problem that the existing methods are prone to inaccurate driving assistance information when facing complex road conditions with heavy traffic.

[0005] To solve the above technical problem, in the first aspect of this application, a vehicle driving assistance system based on digital twins is provided, including: a vehicle following subsystem, a vehicle scanning subsystem, and a control subsystem;

[0006] The vehicle following subsystem includes: a vehicle sensing device, a guide rail, and a driving mechanism. Among them, the vehicle sensing device is used to communicate wirelessly with a target vehicle and collect the vehicle state information of the target vehicle, and the driving mechanism is used to drive the vehicle sensing device along the guide rail to follow the target vehicle and move;

[0007] The vehicle scanning subsystem is used to scan the vehicle contour space data of the vehicles driving on a target section, where the vehicle contour space data includes the shape of the vehicle contour and the road space occupied by the vehicle contour;

[0008] The control subsystem is used to update the section digital twin model of the target section according to the received vehicle state information and the vehicle contour space data, and generate vehicle driving assistance information based on the section digital twin model and feedback it to the target vehicle.

[0009] Preferably, the guide rail includes: a longitudinal guide rail and a transverse guide rail, wherein the longitudinal guide rail is a guide rail laid along the target road section, and the driving mechanism includes: a first driving structure and a second driving mechanism;

[0010] The vehicle sensing device is assembled on the transverse guide rail and is movably connected to the transverse guide rail;

[0011] The transverse guide rail is assembled on the longitudinal guide rail and is movably connected to the transverse guide rail;

[0012] The first driving mechanism is used to drive the transverse guide rail to move on the longitudinal guide rail, and the second driving mechanism is used to drive the vehicle sensing device to move on the transverse guide rail.

[0013] Preferably, the longitudinal guide rail is a circular guide rail laid based on a vertical plane.

[0014] Preferably, the longitudinal guide rail is a circular guide rail laid based on a horizontal plane.

[0015] Preferably, it further includes: a road surface obstacle recognition subsystem, and the road surface obstacle recognition subsystem includes: an image acquisition module and an image recognition module;

[0016] The image recognition module is used to obtain the obstacle recognition result of the target road section according to the road surface image collected by the image acquisition module, combined with a preset obstacle image recognition model, and then upload the obstacle recognition result to the control subsystem, so that the control subsystem is used to update the digital twin model of the road section of the target road section according to the received obstacle recognition result.

[0017] Preferably, the vehicle following subsystem is buried under the road surface of the target road section.

[0018] Preferably, the vehicle sensing device further includes: a wireless charging transmitting module for wirelessly charging the target vehicle.

[0019] Preferably, it further includes: a tempered glass strip;

[0020] There are road surface through holes on the road surface of the target road section, and the tempered glass strip is embedded and fixed in the road surface through holes.

[0021] Meanwhile, a second aspect of the present application provides a control method for a vehicle driving assistance system based on digital twins, which is applied to the vehicle driving assistance system based on digital twins provided in the first aspect of the present application. The control method includes:

[0022] Real-time collect the vehicle contour space data of the vehicles driving on the target road section through the vehicle scanning subsystem;

[0023] When a vehicle following assistance request is received, the vehicle identification information in the vehicle following assistance request is extracted, and then the vehicle identification information is sent to the vehicle following subsystem, so that when the vehicle following subsystem detects the target vehicle corresponding to the vehicle identification information, it executes the following task for the target vehicle and collects the vehicle state information of the target vehicle in real time;

[0024] According to the received vehicle state information and the vehicle profile space data, the digital twin model of the target road section is updated, and vehicle driving assistance information is generated based on the digital twin model of the road section and fed back to the target vehicle.

[0025] Preferably, after updating the digital twin model of the target road section, it further includes:

[0026] According to the real-time target vehicle state and real-time traffic state in the current digital twin model of the road section, path optimization is performed through a driving path optimization model constructed based on a preset genetic algorithm and a stochastic simulation algorithm to obtain the reference driving path of the target vehicle.

[0027] From the above technical solutions, it can be seen that the present application has the following advantages:

[0028] The vehicle driving assistance system provided by the present application includes: a vehicle following subsystem, a vehicle scanning subsystem, and a control subsystem. When the vehicle following subsystem detects that a target vehicle is approaching, it can control the vehicle sensing device to follow the target vehicle along the target road section through a guide rail and a driving mechanism, realize data tracking and collection of the target vehicle, ensure stable collection of target vehicle information, synchronize the collected vehicle state data and the vehicle profile space data collected by the vehicle scanning subsystem to the control subsystem, and the control subsystem updates the traffic digital twin model corresponding to the target road section. The twin model generates vehicle driving assistance information and feeds it back to the target vehicle, thereby providing a vehicle driving assistance service with higher accuracy and stability for the target vehicle, and solving the technical problem that the existing method is prone to inaccurate driving assistance information in the face of complex road conditions with heavy traffic. Description of the Drawings

[0029] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0030] Figure 1 It is a schematic diagram of the system architecture of a vehicle driving assistance system based on digital twins provided by the present application.

[0031] Figure 2 Schematic diagram of the effect of the embedding method of the vehicle following subsystem provided by this application.

[0032] Figure 3 Schematic diagram of the effect of the overhead laying method of the vehicle following subsystem provided by this application.

[0033] Figure 4 Schematic diagram of a laying path of the longitudinal guide rail of the vehicle following subsystem provided by this application and the movement trajectory of the vehicle sensing device.

[0034] Figure 5 Schematic diagram of another laying path of the longitudinal guide rail of the vehicle following subsystem provided by this application and the movement trajectory of the vehicle sensing device.

[0035] Figure 6 Schematic diagram of the cross-sectional structure of a vehicle driving assistance system and a road surface structure based on digital twin provided by this application.

[0036] Figure 7 Schematic diagram of the road surface gap structure in a vehicle driving assistance system based on digital twin provided by this application.

[0037] Figure 8 Schematic diagram of the flow of a control method for a vehicle driving assistance system based on digital twin provided by this application.

[0038] Among them, the reference numerals are as follows:

[0039] A, vehicle following subsystem; B, vehicle scanning subsystem; C, control subsystem; D, road surface obstacle recognition subsystem; S, server; 1, vehicle sensing device; 11, transverse guide rail; 12, longitudinal guide rail; 13, guide rail support frame; 2, scanning device; 3, road surface; 31, reinforced glass strip. Detailed implementation manners

[0040] In view of the technical defects existing in the prior art, through analysis and research, it is found that the reason for the problem is that most of the existing vehicle sensing devices adopt a fixed setting method. When the traffic flow on the road is large, it is easy to lose the target due to vehicle occlusion, thus affecting the accuracy of simulating the running state of vehicles on the road site by subsequent intelligent algorithms or digital twin models.

[0041] In view of this, this application provides a vehicle driving assistance system and a control method based on digital twin, which are used to solve the technical problem that the existing method is prone to inaccurate driving assistance information in the face of complex road conditions with large traffic flow.

[0042] In order to make the invention objectives, features, and advantages of the present application more obvious and understandable, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the embodiments described below are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative efforts belong to the scope of protection of the present application.

[0043] In the description of the present application, it should be noted that the orientation or positional relationship indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. is based on the orientation or positional relationship shown in the accompanying drawings. It is only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present application. In addition, the terms "first", "second", and "third" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance.

[0044] Unless otherwise clearly defined and limited, the terms "installed", "connected", and "connected" shall be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific circumstances.

[0045] First, a detailed description of an embodiment of a vehicle driving assistance system based on digital twin provided by the present application is as follows:

[0046] Please refer to Figure 1 , a vehicle driving assistance system based on digital twin provided in this embodiment includes: a vehicle following subsystem A, a vehicle scanning subsystem B, and a control subsystem C;

[0047] The vehicle following subsystem A includes: a vehicle sensing device 1, a guide rail, and a driving mechanism. Among them, the vehicle sensing device 1 is used to communicate wirelessly with the target vehicle and collect the vehicle status information of the target vehicle. The driving mechanism is used to drive the vehicle sensing device 1 along the guide rail to follow the target vehicle.

[0048] The vehicle scanning subsystem B is used to scan the vehicle contour space data of the vehicles driving on the target section. The vehicle contour space data includes the shape of the vehicle contour and the road space occupied by the vehicle contour.

[0049] The control subsystem C is used to update the digital twin model of the target road section according to the received vehicle status information and vehicle profile space data, and generate vehicle driving assistance information based on the digital twin model of the road section and feedback it to the target vehicle.

[0050] It should be noted that the vehicle driving assistance system provided in this embodiment includes: a vehicle following subsystem A, a vehicle scanning subsystem B, and a control subsystem C.

[0051] Among them, the vehicle following subsystem A is used to perform following communication on the vehicle that needs to provide driving assistance services. When the vehicle following subsystem A detects that the target vehicle is approaching, it controls the vehicle sensing device 1 to establish a wireless communication connection with the target vehicle, and follows the target vehicle along the target road section through the guide rail and the driving mechanism, so as to realize the data tracking and collection of the target vehicle, ensure that the target vehicle information can be stably collected. The setting method of the vehicle following subsystem A can be buried under the road surface 3, specifically as Figure 2 shown, or it can be set above the road surface 3, for example, set at the tunnel ceiling. If it is an open road scene, in addition to the buried setting method, it can also be adopted as Figure 3 shown in the overhead setting method.

[0052] The moving vehicle is equipped with in-vehicle communication equipment, such as an in-vehicle communication module or an in-vehicle communication system. These devices can receive external data through wireless communication technologies, such as Wi-Fi, Bluetooth, or mobile networks. Roadside devices are installed beside the road or under the road surface 3. These devices can be communication base stations, roadside sensors, or road traffic management systems. These devices can cover the entire road network through wireless networks and communicate with the moving vehicles. Data transmission: The driving guarantee system transmits the prediction result data to the roadside device, and then the roadside device sends it to the moving vehicle through wireless communication technology. The data transmission can use short-distance wireless communication technologies, such as Wi-Fi or Bluetooth, or can be remotely transmitted through the mobile network.

[0053] In this process, generally, the signal of the vehicle following subsystem A is transmitted to the digital twin model through a wireless network link, and the signal characteristics of the vehicle following subsystem A are identified and compared through the driving guarantee model in the digital twin model, thereby obtaining the vehicle's instant state. The signal of the vehicle following subsystem A here can be processed and feature extracted in the digital twin model, or can be done before entering the digital twin model, which can effectively reduce the data transmission volume and improve the update efficiency of the digital twin model.

[0054] The vehicle scanning subsystem B is installed on the road surface 3. Specifically, it can be installed on both sides of the target road section or above the road in an overhead manner. Generally, it includes multiple scanning devices 2, and each scanning device 2 is evenly distributed in the target road section and is used to scan the contour data of the vehicles driving on the target road section, specifically including the contour shape of the vehicle and the space occupied by the contour. When the vehicle following subsystem A adopts an overhead laying mode, the scanning device 2 of the vehicle scanning subsystem B can be installed on the guide rail support frame 13.

[0055] The control subsystem C belongs to the main control and scheduling unit. It can collect the vehicle contour space data of the vehicles driving on the target road section in real time through the vehicle scanning subsystem B on the target road section, can respond to service requests directly sent by users or forwarded via the server S, and schedule the vehicle following subsystem A to execute the following task to collect the vehicle state data of the target vehicle, including but not limited to vehicle speed, acceleration, wheel steering angle, vehicle tilt condition, etc. Then, the collected vehicle state data and vehicle contour space data are uniformly summarized in the control subsystem C. The control subsystem C updates the digital twin model of the target road section carried by it in real time, and generates vehicle driving assistance information based on the road section digital twin model and feeds it back to the target vehicle through the vehicle sensing device 1 or other wireless communication links for the target vehicle to refer to, so as to facilitate the driver of the target vehicle to more comprehensively understand the overall road conditions of the road section and thus make more appropriate driving decisions. The vehicle driving assistance information fed back generally includes the following information: the contour and spatial position of the vehicle itself, the contours of other vehicles, and their spatial distribution on the target road section.

[0056] In some embodiments, the vehicle driving assistance system provided in this embodiment may further include: a road surface 3 obstacle recognition subsystem D, and the road surface 3 obstacle recognition subsystem D includes: an image acquisition module and an image recognition module;

[0057] The image recognition module is used to obtain the obstacle recognition result of the target road section according to the road surface 3 image collected by the image acquisition module in combination with a preset obstacle image recognition model, and then upload the obstacle recognition result to the control subsystem C so that the control subsystem C can update the road section digital twin model of the target road section according to the received obstacle recognition result.

[0058] It should be noted that the road surface 3 obstacle recognition subsystem D collects the road surface 3 image through an image acquisition device such as a camera, inputs it into a pre-trained obstacle image recognition model, obtains the obstacle recognition result of the target road section, and then uploads the obstacle recognition result to the control subsystem C so that the control subsystem C can update the road section digital twin model of the target road section according to the received obstacle recognition result, and can provide driving assistance information including road obstacle information for the target vehicle.

[0059] Further, the guide rail includes: a longitudinal guide rail 12 and a transverse guide rail 11. Among them, the longitudinal guide rail 12 is a guide rail laid along the target section. The driving mechanism includes: a first driving structure and a second driving mechanism;

[0060] The vehicle sensing device 1 is assembled on the transverse guide rail 11 and is movably connected to the transverse guide rail 11;

[0061] The transverse guide rail 11 is assembled on the longitudinal guide rail 12 and is movably connected to the transverse guide rail 11;

[0062] The first driving mechanism is used to drive the transverse guide rail 11 to move on the longitudinal guide rail 12, and the second driving mechanism is used to drive the vehicle sensing device 1 to move on the transverse guide rail 11.

[0063] It should be noted that when the vehicle following subsystem A starts a following task, the first driving mechanism drives the transverse guide rail 11 and the vehicle sensing device 1 along the longitudinal guide rail 12 and follows the target vehicle forward. When it is detected that the target vehicle changes lanes, the second driving mechanism can be used to drive the vehicle sensing device 1 to move on the transverse guide rail 11, so that the vehicle sensing device 1 can shift in the direction of the target vehicle's lane change, thereby reducing the impact of the change in the spacing distance on the communication quality.

[0064] In some special scenarios, such as in a one-way single-lane or two-way single-lane section, that is, a section with only one lane in the same direction, the transverse guide rail 11 and the second driving mechanism can also be removed, the vehicle sensing device 1 can be directly assembled on the longitudinal guide rail 12, and directly driven by the first driving mechanism.

[0065] Further, the longitudinal guide rail 12 mentioned in this embodiment can specifically adopt the structure of a circular guide rail. Whether the circular guide rail is set in a buried or overhead manner, there are two laying methods for the selection of the laying direction of the circular guide rail: one is the laying method based on the vertical plane direction, and the other is the laying method based on the horizontal plane direction.

[0066] Among them, for the first laying method, the laying path of the longitudinal guide rail 12 and the moving trajectory of the vehicle sensing device 1 can be referred to Figure 4 , which is generally used for single-lane sections, but can also be applied to two-way lane sections. Its operating principle is that when the vehicle sensing device 1 ends the following task and reaches the end of the section, the vehicle sensing device 1 continues to move forward under the drive of the driving mechanism, enters the return loop, returns to the starting point of the section through this return loop, and then waits to receive the next following task.

[0067] For the second laying method, the laying path of the longitudinal guide rail 12 and the moving trajectory of the vehicle sensing device 1 can be referred to Figure 5, its operating principle is similar to the first laying method. The difference is that the return loop of this laying method is located in the oncoming lane and is applicable to two-way traffic road sections. When the vehicle sensing device 1 finishes following the task and reaches the end of the road section, the vehicle sensing device 1 continues to move forward driven by the driving mechanism. After passing through the buffer area, it moves to the starting point of the road section in the oncoming lane and then waits to receive the next following task in this direction.

[0068] If the vehicle following subsystem A adopts the buried method, the vehicle sensing device 1 may further include: a wireless charging transmitting module, which can be used to wirelessly charge the target vehicle with wireless charging conditions during the execution of the following task.

[0069] Such as Figure 6 and Figure 7 As shown, if the vehicle following subsystem A adopts the buried method, through holes may also be provided on the road surface 3 of the target road section, and the toughened glass strip 31 is embedded and fixed in the through holes of the road surface 3. The strip has high reflectivity and wear resistance, so that while providing visual guidance, it can withstand frequent traffic passing. The glass material strip can be embedded in or attached to the road surface and fixed by special adhesives or mechanical fixing means to ensure its long-term stability. At the same time, by setting the glass strip on the road surface 3, the attenuation effect of concrete on wireless signals and charging electromagnetic waves can also be reduced. Moreover, due to the strong light transmission ability of glass, when the vehicle sensing device 1 and the target vehicle adopt optical signal communication methods such as infrared rays, their communication quality can be ensured.

[0070] The above is a detailed description of an embodiment of a digital twin-based vehicle driving assistance system provided by this application. The following is a detailed description of an embodiment of a control method of a digital twin-based vehicle driving assistance system provided by this application.

[0071] Please refer to Figure 8 , the second aspect of this application provides a control method of a digital twin-based vehicle driving assistance system, which is applied to the digital twin-based vehicle driving assistance system provided in the above embodiment. This control method includes:

[0072] Step 101: Real-time collect the vehicle contour space data of the vehicles driving on the target road section through the vehicle scanning subsystem;

[0073] Step 102: When receiving a vehicle following assistance request, extract the vehicle identification information in the vehicle following assistance request, and then send the vehicle identification information to the vehicle following subsystem, so that when the vehicle following subsystem detects the target vehicle corresponding to the vehicle identification information, it executes the following task for the target vehicle and real-time collects the vehicle state information of the target vehicle;

[0074] Step 103: Update the digital twin model of the target road section according to the received vehicle status information and vehicle contour space data, and generate vehicle driving assistance information based on the road section digital twin model and feedback it to the target vehicle.

[0075] It should be noted that when constructing a digital twin model, a geometric model is first established, and the assembly connection relationships of the components in the geometric model are the same as those in the physical space. Suppose there are two point cloud data sets: one is the initial scanned point cloud data without processing, representing the original shape and position of the car; the other is the target point cloud data, representing the shape and position of the car after being scanned as expected. The goal is to align the point cloud data obtained from the initial scan to the target point cloud data in order to construct an accurate digital twin model. Taking the ICP algorithm as an example, the following steps can be applied:

[0076] Suppose the unknown to be solved is the transformation matrix T, which includes the rotation matrix R and the translation vector t. The steps of the ICP algorithm can be re-described as follows: Initialization, select an initial transformation matrix T0. Here, T0 is an unknown, representing the transformation in the initial state. Corresponding point matching, for each source point pi (a point in the initial scanned point cloud), find the nearest point qi in the target point cloud data and establish the corresponding relationship between the point pairs. Calculate the transformation, through the matching point pairs, calculate the optimal transformation T to minimize the distance between the initial scanned point cloud data and the target point cloud data. This can be achieved by minimizing the distance between the point pairs, that is, minimizing the error function:

[0077]

[0078] where E(T) is the error function, representing the distance error between the point cloud data. N is the number of point pairs. represents the Euclidean distance. p i is a point in the source point cloud. q i is the point in the target point cloud corresponding to p i . T(p i ) is the point obtained by transforming the point p i through the transformation T. The distance here can use the Euclidean distance or other distance measurement methods. Update the point cloud, apply the obtained optimal transformation T to transform the point cloud data obtained from the initial scan to align it with the target point cloud data. Repeat iteration, repeat the above steps until the stop condition is met, such as reaching the maximum number of iterations or reaching the convergence condition. In each iteration, the transformation matrix T will be updated until the error function E(T) reaches the minimum value or converges to a certain threshold. In this scenario, the ICP algorithm helps to achieve the accurate registration of the car, thus enabling the construction of an accurate digital twin model. Through this model, various simulation, analysis, and design work can be carried out without the actual physical car.

[0079] For the convenience of description, here the physical space refers to the actual space, while the virtual space refers to the space where the digital twin model is located. Sometimes, in order to simplify the digital twin device model, certain restrictions are set for the digital twin device model in the virtual space to make it meet certain production operation functions, forming an integrated digital twin device model with topological characteristics. According to the established geometric model, physical attributes are added to the digital twin model in the virtual space, such as process requirements for some components, constraints on connecting components, material attribute information, etc., to make it more in line with the equipment in the physical space. The outer wall of the digital twin model fully reflects the state of the physical space and is used for the simulation calculation of the 3D geometric scanner. In the digital twin model, the influence of the environment can be fully considered, and then the interactivity between the environment and the equipment in the digital twin model can be realized.

[0080] Further, after updating the road section digital twin model of the target road section, it further includes:

[0081] Step 104: According to the real-time target vehicle state and real-time traffic state in the current road section digital twin model, path optimization is performed through a driving path optimization model constructed based on a preset genetic algorithm and stochastic simulation algorithm to obtain the reference driving path of the target vehicle.

[0082] Among them, the stochastic simulation algorithms mentioned in this embodiment mainly include: decision tree, random forest, Bayesian algorithm, time series prediction algorithm, etc.

[0083] It should be noted that the genetic algorithm is used to optimize the parameters of each component in the digital twin model to ensure that the model can accurately reflect the behavior and performance of the physical system.

[0084]

[0085] Among them: P (set of model parameters), F (fitness function), P optimal is the set of parameters that maximizes the fitness function F, and the fitness function is defined according to the prediction accuracy and efficiency of the model.

[0086] For example, the fitness function: , where 、 , are weight coefficients, reflecting the relative importance of safety, comfort, and energy consumption in the driving strategy.

[0087] Its general process is as follows: Initialize the population, randomly generate a population of the initial parameter set. Each individual represents a parameter set. Evaluate the fitness, use the fitness function F(P) to evaluate each individual in the population, and calculate its fitness value. Selection, according to the fitness value of each individual, use the selection operator to select excellent individuals as parents for crossover and mutation. Crossover, through the crossover operation, combine the genes of the parent individuals to generate new offspring individuals. Mutation, perform mutation operations on the newly generated offspring individuals to introduce some randomness to increase the diversity of the population. Evaluate the new individuals: Apply the fitness function to the newly generated individuals and calculate their fitness values. Update the population, use the selection operator to select the most excellent individuals from the parents and offspring to form the next generation population.

[0088] Then repeat the operations of crossover, mutation, and selection until the termination condition is reached (such as reaching the maximum number of iterations or finding a satisfactory solution). Through this process, the parameter set of the model can be gradually optimized, so that the prediction accuracy and efficiency of the model reach the optimal, while considering the relative importance of safety, comfort, and energy consumption, thus ensuring that the model can accurately reflect the behavior and performance of the entity system.

[0089] Based on decision trees and random forests are used to identify key features and patterns from the imported data for further analysis and prediction. Decision trees are used to classify the driving states of vehicles in real time (such as "normal driving", "emergency braking", "obstacle avoidance") based on real-time sensor data. Let X (feature vector) contain real-time sensor data, and Y (target variable) be the driving state of the vehicle (such as "normal driving", "emergency braking", "obstacle avoidance"). Then there is Y = f(X), where f is a function learned by a decision tree or random forest for predicting or classifying vehicle states.

[0090] The digital twin model performs simulation calculations and training on the input data of the vehicle following subsystem, and constructs the following driving guarantee objective function.

[0091]

[0092] In the formula, y t is the vehicle state predicted by the time series model, is white noise. is the parameter of the autoregressive model. θ 1 ,…,θ q are the parameters of the moving average model

[0093] During the operation of the device, the digital twin model is updated in real-time. Some of its operating states can be directly collected, such as temperature, speed, and power consumption. Its motion state is fed back through the signals of the follow-up vehicle following subsystem. Through the signal feedback of the follow-up vehicle following subsystem, we can obtain the immediate state of the vehicle in a timely and comprehensive manner.

[0094] Based on the probability prediction of the Bayesian network for real-time updating of vehicle states, it processes real-time data with high uncertainty. The posterior probability is calculated by P(X∣Y)=P(Y)*P(Y∣X) to update the uncertainty assessment of the vehicle state. The Bayesian network is used to handle and reason about situations with higher uncertainty, such as driving under poor visibility conditions, and calculates the success probabilities of various driving strategies based on historical data and current perception data.

[0095] In this embodiment, by performing simulation calculations on the target digital twin model to obtain simulation operation data, compared with the method of obtaining data based on experiments in the actual physical space, it can reduce the time cost of data acquisition. Time series analysis, such as the autoregressive moving average (ARMA) model, is used to process and analyze the time series data of the vehicle driving state to optimize the prediction accuracy of the digital twin model.

[0096] In an implementable way, the model is trained based on experimental data and simulation data, including:

[0097] Comparing the experimental data and the simulation data to obtain error data;

[0098] Correcting the target digital twin model according to the error data;

[0099] Constructing a sample set based on the experimental data and the simulation data;

[0100] Training the corrected target digital twin model according to the sample set.

[0101] In this embodiment, correcting the target digital twin model first during the model training process can ensure that the digital twin model is consistent with the functional characteristics of the device and the environment in the physical space, ensure the correct association of the sensing device and the vehicle in the digital twin model, and is beneficial to improving the training accuracy.

[0102] In model training, an efficient and applicable intelligent prediction model can be obtained based on a method that combines a judgment model and a generation model. Here, it refers to performing evolutionary computation on a suitable model based on operation data and simulation data to obtain appropriate model parameters. The judgment model can be a combination of one or more of linear regression, logarithmic regression, linear discriminant analysis, support vector machine, boosting, conditional random field, neural network, etc. The generation model can be a collection of one or more models, such as Gaussian mixture model, hidden Markov model, and naive Bayes model, etc.

[0103] The generated driving guarantee model can give the driving state of the vehicle according to the signals of the vehicle following subsystem. Usually, while giving the driving state result, it can also perform predictive analysis on future safe driving, so as to help relevant personnel discover problems in advance and take measures to deal with them, in order to reduce or lower losses and casualties.

[0104] It should be understood that in this application, "at least one (item)" means one or more, and "a plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that there can be three relationships. For example, "A and / or B" can mean: only A exists, only B exists, and both A and B exist at the same time. Among them, A and B can be singular or plural. The character " / " generally means that the associated objects before and after are in an "or" relationship. "At least one (one) of the following" or its similar expression refers to any combination of these items, including any combination of single item (one) or plural items (ones). For example, at least one (one) of a, b, or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0105] The unit described as a separate component may or may not be physically separated. The component shown as a unit may or may not be a physical unit, that is, it can be located in one place, or it can be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0106] In addition, in each embodiment of the present invention, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0107] When 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 invention, in essence, 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 several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs.

[0108] As described above, the above embodiments are only used to illustrate the technical solutions of the present application and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present application.

Claims

1. A vehicle driving assistance system based on digital twins, characterized in that: include: A vehicle following subsystem, a vehicle scanning subsystem and a control subsystem, wherein the vehicle following subsystem and the vehicle scanning subsystem are both communicatively connected with the control subsystem; The vehicle following subsystem includes: a vehicle sensing device, a guide rail and a driving mechanism, wherein the vehicle sensing device is used to wirelessly communicate with a target vehicle and collect vehicle status information of the target vehicle, and the driving mechanism is used to drive the vehicle sensing device to move along the guide rail to follow the target vehicle, and the guide rail includes: a longitudinal guide rail laid along the target road section, and the longitudinal guide rail is specifically a ring guide rail structure; The vehicle scanning subsystem is used to scan the vehicle contour space data of the vehicle traveling on the target road section, wherein the vehicle scanning subsystem is used to scan the vehicle contour space data of the vehicle traveling on the target road section, and the vehicle contour space data includes the shape of the vehicle contour and the road space occupied by the vehicle contour; The control subsystem is used to update the road section digital twin model of the target road section according to the received vehicle status information and the vehicle contour spatial data, and generate vehicle driving assistance information based on the road section digital twin model and feed it back to the target vehicle through the vehicle sensing device.

2. The vehicle driving assistance system based on digital twin according to claim 1, characterized in that: The guide rail further comprises: a transverse guide rail, and the driving mechanism comprises: a first driving structure and a second driving mechanism; The vehicle sensing device is mounted on the transverse guide rail and is movably connected to the transverse guide rail; The transverse guide rail is assembled on the longitudinal guide rail and is movably connected to the transverse guide rail; The first driving mechanism is used to drive the transverse guide rail to move on the longitudinal guide rail, and the second driving mechanism is used to drive the vehicle sensing device to move on the transverse guide rail.

3. The vehicle driving assistance system based on digital twin according to claim 1, characterized in that: The longitudinal guide rail is an annular guide rail laid on a vertical plane.

4. The vehicle driving assistance system based on digital twin according to claim 1, characterized in that: The longitudinal guide rail is an annular guide rail laid on the basis of a transverse plane.

5. The vehicle driving assistance system based on digital twin according to claim 1, characterized in that: Also includes: A road obstacle recognition subsystem, the road obstacle recognition subsystem comprising: an image acquisition module and an image recognition module; The image recognition module is used to obtain the obstacle recognition result of the target road section based on the road surface image captured by the image acquisition module in combination with a preset obstacle image recognition model, and then upload the obstacle recognition result to the control subsystem, so that the control subsystem is used to update the road section digital twin model of the target road section according to the received obstacle recognition result.

6. The vehicle driving assistance system based on digital twin according to claim 1, characterized in that: The vehicle following subsystem is buried under the road surface of the target road section.

7. The vehicle driving assistance system based on digital twin according to claim 6, characterized in that: The vehicle sensing device also includes: a wireless charging transmitting module, which is used to wirelessly charge the target vehicle.

8. The vehicle driving assistance system based on digital twin according to claim 6, characterized in that: Also includes: tempered glass strips; A road surface through hole is arranged on the road surface of the target road section, and the tempered glass strip is embedded in and fixed in the road surface through hole.

9. A control method for a vehicle driving assistance system based on digital twins, applied to a vehicle driving assistance system based on digital twins as claimed in any one of claims 1 to 8, characterized in that: The control method comprises: The vehicle scanning subsystem collects the vehicle profile spatial data of the vehicles traveling on the target road section in real time; When a vehicle following assistance request is received, the vehicle identification information in the vehicle following assistance request is extracted, and the vehicle identification information is then sent to the vehicle following subsystem, so that when the vehicle following subsystem detects a target vehicle corresponding to the vehicle identification information, it performs a following task for the target vehicle and collects vehicle status information of the target vehicle in real time; According to the received vehicle status information and the vehicle contour spatial data, the road section digital twin model of the target road section is updated, and vehicle driving assistance information is generated based on the road section digital twin model and fed back to the target vehicle through the vehicle sensing device in the vehicle following subsystem.

10. The control method of a vehicle driving assistance system based on digital twin according to claim 9, characterized in that: After updating the road segment digital twin model of the target road segment, the method further includes: According to the real-time target vehicle status and real-time traffic status in the current digital twin model of the road section, path optimization is performed through a driving path optimization model constructed based on a preset genetic algorithm and a random simulation algorithm to obtain a reference driving path for the target vehicle.

Images