Vehicle obstacle avoidance method and system, computer-readable storage medium and electronic device

Abstract

The present disclosure relates to the field of Internet technology, and to a vehicle obstacle avoidance method and system, a computer-readable storage medium, and an electronic device. The vehicle obstacle avoidance system includes a twin edge layer of a digital twin space and an edge service layer of a physical space. The method includes: the twin edge layer trains an obstacle avoidance decision model according to first operation data and second operation data, the first operation data is the operation data of a physical vehicle in the physical space, the second operation data is the operation data of a twin vehicle in the digital twin space, and the twin vehicle is a twin model of the physical vehicle in the digital twin space; the twin edge layer sends the obstacle avoidance decision model to the edge service layer, so that the twin vehicle and the physical vehicle simultaneously make obstacle avoidance decisions based on the obstacle avoidance decision model, so as to control the physical vehicle to perform an obstacle avoidance process according to the obtained decision result. The present disclosure can reduce the occupation of edge computing resources and realize effective obstacle avoidance of vehicles.

Description

Technical Field

[0001] The present disclosure relates to the field of Internet technology, and more specifically, to a vehicle obstacle avoidance method, a vehicle obstacle avoidance system, a computer-readable storage medium, and an electronic device. Background Art

[0002] With the development of the Internet technology field, the Internet of Things based on information carriers such as the Internet and traditional telecommunications networks has enabled all ordinary physical objects that can be independently addressed to form an interconnected network, while the Internet of Vehicles is a large system network based on the intra-vehicle network, inter-vehicle network and in-vehicle mobile Internet, and performs wireless communication and information exchange between vehicles and X (X: vehicles, roads, pedestrians and the Internet, etc.) according to the agreed communication protocols and data interaction standards. Among them, vehicle obstacle avoidance is one of the cutting-edge research directions in the field of Internet of Vehicles and has become a core issue in the field of autonomous driving technology.

[0003] At present, a large amount of edge computing resources are needed to achieve vehicle obstacle avoidance, and the edge computing cost and network load pressure are high.

[0004] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0005] The purpose of the present disclosure is to provide a vehicle obstacle avoidance method and system, a computer-readable storage medium and an electronic device, thereby reducing the occupation of edge computing resources during the vehicle obstacle avoidance process, reducing edge computing costs and network load pressure.

[0006] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by the practice of the present disclosure.

[0007] According to one aspect of the present disclosure, a vehicle obstacle avoidance method is provided, which is applied to a vehicle obstacle avoidance system, wherein the vehicle obstacle avoidance system includes a twin edge layer of a digital twin space and an edge service layer of a physical space, and the method includes:

[0008] The twin edge layer trains an obstacle avoidance decision model according to first operating data and second operating data, wherein the first operating data is operating data of a physical vehicle in the physical space, the second operating data is operating data of a twin vehicle in the digital twin space, and the twin vehicle is a twin model of the physical vehicle in the digital twin space;

[0009] The twin edge layer sends the obstacle avoidance decision model to the edge service layer, so that the twin vehicle and the physical vehicle simultaneously make obstacle avoidance decisions based on the obstacle avoidance decision model, so as to control the physical vehicle to perform the obstacle avoidance process according to the obtained decision results.

[0010] In an exemplary embodiment of the present disclosure, there is a real-time data channel between the twin edge layer and the edge service layer;

[0011] During the simultaneous movement of the physical vehicle and the twin vehicle, the edge service layer uploads the first operating data to the twin edge layer based on the real-time data channel, wherein the twin edge layer includes the corresponding second operating data.

[0012] In an exemplary embodiment of the present disclosure, before the twin edge layer sends the obstacle avoidance decision model to the edge service layer, the method further includes:

[0013] The twin edge layer uses a test sample to verify and evaluate the obstacle avoidance decision model to determine whether the obstacle avoidance decision model meets the model issuance conditions;

[0014] The first operating data and the second operating data are divided into the test samples and the training samples, and the obstacle avoidance decision model is trained based on the training samples.

[0015] In an exemplary embodiment of the present disclosure, the twin edge layer sends the obstacle avoidance decision model to the edge service layer, so that the twin vehicle and the physical vehicle simultaneously make obstacle avoidance decisions based on the obstacle avoidance decision model, so as to control the physical vehicle to perform an obstacle avoidance process according to the obtained decision result, including:

[0016] The process of any obstacle avoidance vehicle making an obstacle avoidance decision based on the obstacle avoidance decision model and obtaining the decision result includes:

[0017] The obstacle avoiding vehicle makes an obstacle avoiding decision according to the obstacle avoiding decision model to obtain first obstacle avoiding movement information, and determines preview point information based on the first obstacle avoiding movement information and in accordance with a dynamic preview search algorithm, wherein the preview point information is position information of the obstacle avoiding vehicle after the obstacle avoiding movement is performed according to the first obstacle avoiding movement information, and the obstacle avoiding vehicle includes the twin vehicle and the physical vehicle;

[0018] Calculating second obstacle avoidance movement information according to the pre-motion parameter information corresponding to the first obstacle avoidance movement information and according to a preset lateral path verification model;

[0019] Calculating third obstacle avoidance movement information according to the pre-motion parameter information corresponding to the first obstacle avoidance movement information, the current vehicle performance information of the obstacle avoidance vehicle, and the current road condition information according to a preset longitudinal path verification model;

[0020] The preview point information is verified in combination with the second obstacle avoidance movement information and the third obstacle avoidance movement information, and after the preview point information passes the verification, the preview point information is used as the decision result.

[0021] In an exemplary embodiment of the present disclosure, the preview point information includes horizontal position information and vertical position information;

[0022] The conditions for the preview point information to pass verification include:

[0023] The deviation of the lateral position information from the second obstacle avoidance movement information is less than a first deviation threshold;

[0024] The deviation of the longitudinal position information from the third obstacle avoidance movement information is less than a second deviation threshold.

[0025] In an exemplary embodiment of the present disclosure, the step of controlling the physical vehicle to perform an obstacle avoidance process according to the obtained decision result includes:

[0026] The twin vehicle moves along an obstacle avoidance path according to the decision result, obtains first obstacle avoidance result information and sends it to the physical vehicle;

[0027] When the first obstacle avoidance result information indicates that the twin vehicle successfully avoids the obstacle, the physical vehicle moves along the obstacle avoidance path according to the decision result to obtain second obstacle avoidance result information.

[0028] In an exemplary embodiment of the present disclosure, the number of the twin vehicles in the twin edge layer is multiple, and each of the twin vehicles moves along an obstacle avoidance path according to its corresponding decision result, and obtains multiple pieces of the first obstacle avoidance result information;

[0029] When the first obstacle avoidance result information indicates that the twin vehicle successfully avoids the obstacle, the physical vehicle moves along the obstacle avoidance path according to the decision result, and before obtaining the second obstacle avoidance result information, the method further includes:

[0030] From the plurality of first obstacle avoidance result information, obtaining target first obstacle avoidance result information indicating that the corresponding twin vehicle successfully avoided an obstacle;

[0031] The target first obstacle avoidance result information is sent to the physical vehicle, so that the physical vehicle moves along an obstacle avoidance path according to the decision result corresponding to the target first obstacle avoidance result information, and obtains the second obstacle avoidance result information.

[0032] In an exemplary embodiment of the present disclosure, the method further includes:

[0033] The twin edge layer trains and updates the obstacle avoidance decision model using the updated first operation data and the updated second operation data according to a set period;

[0034] Among them, the updated first operating data and the updated second operating data are data generated during the synchronous operation of the physical vehicle and the twin vehicle.

[0035] According to one aspect of the present disclosure, there is provided a vehicle obstacle avoidance system, the vehicle obstacle avoidance system comprising a twin edge layer of a digital twin space and an edge service layer of a physical space;

[0036] The twin edge layer is used to train an obstacle avoidance decision model according to first operating data and second operating data, wherein the first operating data is operating data of a physical vehicle in the physical space, the second operating data is operating data of a twin vehicle in the twin edge layer, and the twin vehicle is a twin model of the physical vehicle in the digital twin space;

[0037] The twin edge layer is used to send the obstacle avoidance decision model to the edge service layer, so that the twin vehicle and the physical vehicle can make obstacle avoidance decisions based on the obstacle avoidance decision model at the same time, so as to control the physical vehicle to perform the obstacle avoidance process according to the obtained decision results.

[0038] In an exemplary embodiment of the present disclosure, a decision module is used for an obstacle avoidance vehicle to make an obstacle avoidance decision according to the obstacle avoidance decision model to obtain first obstacle avoidance movement information, wherein the obstacle avoidance vehicle includes the twin vehicle and the physical vehicle;

[0039] A dynamic preview point search module, configured to determine preview point information based on the first obstacle avoidance movement information and in accordance with a dynamic preview search algorithm, wherein the preview point information is position information of the obstacle avoidance vehicle after the obstacle avoidance movement is performed in accordance with the first obstacle avoidance movement information;

[0040] a first vehicle control module, configured to calculate second obstacle avoidance movement information according to the pre-motion parameter information corresponding to the first obstacle avoidance movement information and in accordance with a preset lateral path verification model;

[0041] a second vehicle control module, configured to calculate third obstacle avoidance movement information according to the pre-motion parameter information corresponding to the first obstacle avoidance movement information, the current vehicle performance information of the obstacle avoidance vehicle, and the current road condition information, in accordance with a preset longitudinal path verification model;

[0042] A decision result determination module is used to verify the preview point information in combination with the second obstacle avoidance movement information and the third obstacle avoidance movement information, and after the preview point information passes the verification, use the preview point information as the decision result.

[0043] According to one aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, any one of the above methods is implemented.

[0044] According to one aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute any one of the above-mentioned methods by executing the executable instructions.

[0045] In the vehicle obstacle avoidance method in the exemplary embodiment of the present disclosure, first, the edge service layer collects the first operating data of the physical vehicle, and the twin edge layer collects the second operating data of the twin vehicle. Then, the twin edge layer trains and updates the obstacle avoidance decision model according to the first operating data and the second operating data, and sends the obtained obstacle avoidance decision model to the edge service layer, so that the twin vehicle and the physical vehicle can make obstacle avoidance decisions based on the obstacle avoidance decision model respectively, and finally the physical vehicle is controlled to avoid obstacles according to the obtained decision result. On the one hand, the edge twin network is introduced, and the operating data of the physical vehicle and the twin vehicle are collected for model training through cloud-edge collaboration. After the model is trained in the twin edge layer, the obstacle avoidance decision model is sent to the edge service layer, which reduces the occupation of edge computing resources, reduces edge computing costs and network loads; on the other hand, both the twin vehicle and the physical vehicle participate in the obstacle avoidance decision, thereby improving the accuracy of the decision result; on the other hand, vehicle obstacle avoidance is achieved with less edge computing resource occupation, which is conducive to promoting the popularization and intelligent development of vehicle networks.

[0046] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The above and other objects, features and advantages of the exemplary embodiments of the present disclosure will become readily understood by reading the detailed description below with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present disclosure are shown in an exemplary and non-limiting manner, in which:

[0048] Figure 1 An application scenario diagram according to an exemplary embodiment of the present disclosure is shown;

[0049] Figure 2 A flow chart of a vehicle obstacle avoidance method according to an exemplary embodiment of the present disclosure is shown;

[0050] Figure 3 A structural schematic diagram of a digital twin edge network according to an exemplary embodiment of the present disclosure is shown;

[0051] Figure 4 A flowchart showing an obstacle avoidance decision-making process performed by an obstacle avoidance vehicle based on an obstacle avoidance decision-making model according to an exemplary embodiment of the present disclosure is shown;

[0052] Figure 5 A schematic diagram showing the interaction of various parts of a vehicle obstacle avoidance system involved in a vehicle obstacle avoidance application scenario according to an exemplary embodiment of the present disclosure is shown;

[0053] Figure 6 A flowchart of a vehicle obstacle avoidance planning decision based on DTEN is shown according to an exemplary embodiment of the present disclosure, based on an obstacle avoidance decision model, a physical vehicle in a physical space and a twin vehicle in a digital twin space;

[0054] Figure 7 A schematic diagram of the architecture of a vehicle obstacle avoidance system according to an exemplary embodiment of the present disclosure is shown;

[0055] Figure 8 A block diagram of an electronic device according to an exemplary embodiment of the present disclosure is shown.

[0056] In the drawings, the same or corresponding reference numerals represent the same or corresponding parts. DETAILED DESCRIPTION

[0057] The exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, the exemplary embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that the present disclosure will be more comprehensive and complete and the concepts of the exemplary embodiments will be fully conveyed to those skilled in the art. The same reference numerals in the figures represent the same or similar structures, and thus their detailed description will be omitted.

[0058] In addition, the described features, structures or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced without one or more of the specific details, or other methods, components, devices, steps, etc. may be adopted. In other cases, known structures, methods, devices, implementations or operations are not shown or described in detail to avoid blurring the various aspects of the present disclosure.

[0059] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or these functional entities or parts of functional entities may be implemented in one or more software hardened modules, or these functional entities may be implemented in different networks and / or processor devices and / or microcontroller devices.

[0060] In the related technologies in this field, intelligent vehicles, as intelligent robots with autonomous decision-making capabilities, need to obtain information from the external environment and make decisions based on the information, so as to carry out global path planning and obstacle avoidance in local dangerous conditions. The vehicle obstacle avoidance algorithm of the related technology requires a large amount of computing resources to achieve the function of local obstacle avoidance, and the edge computing cost and network load pressure are high, which to some extent affects the popularization and intelligent development of the Internet of Vehicles.

[0061] Based on this, in an exemplary embodiment of the present disclosure, a vehicle obstacle avoidance method is first provided. Figure 1 FIG. 1 is an application scenario diagram involved in the exemplary embodiment of the present disclosure. Figure 1 As shown, when the vehicle 110 is driving, if there is an obstacle 120 (such as a vehicle, obstacle, etc.) in the driving direction, the vehicle 110 needs to make a timely decision on the driving path to avoid the obstacle 120 to avoid collision and continue driving. The obstacle avoidance process of the vehicle 110 may include overtaking, braking, and lane keeping, etc. The vehicle 110 may be a fully automatic driving vehicle without human intervention for obstacle avoidance. The vehicle 110 may also be driven by a human, but when encountering an obstacle 120, the vehicle may assist in making decisions to avoid the obstacle, further ensuring the safety of vehicle driving, and there is no special restriction on this.

[0062] It is worth mentioning that Figure 1 The number of obstacle objects 120 in the figure is merely illustrative. Depending on the actual vehicle obstacle avoidance scenario, there may be any number of obstacle objects, and the present disclosure does not make any special limitation on this.

[0063] like Figure 2 A flow chart of a vehicle obstacle avoidance method according to an embodiment of the present disclosure is shown, which is applied to a vehicle obstacle avoidance system, and the vehicle obstacle avoidance system includes a twin edge layer of a digital twin space and an edge service layer of a physical space. Figure 2 As shown, the vehicle obstacle avoidance method of the embodiment of the present disclosure may include step S210 and step S220:

[0064] Step S210: The twin edge layer trains an obstacle avoidance decision model according to the first operating data and the second operating data, where the first operating data is the operating data of the physical vehicle in the physical space, and the second operating data is the operating data of the twin vehicle in the digital twin space, and the twin vehicle is a twin model of the physical vehicle in the digital twin space;

[0065] Step S220: The twin edge layer sends the obstacle avoidance decision model to the edge service layer, so that the twin vehicle and the physical vehicle make obstacle avoidance decisions based on the obstacle avoidance decision model at the same time, so as to control the physical vehicle to perform the obstacle avoidance process according to the obtained decision results.

[0066] According to the vehicle obstacle avoidance method of the embodiment of the present invention, an edge twin network is introduced, and the operating data of the physical vehicle and the twin vehicle are collected through cloud-edge collaboration. After the model is trained in the twin edge layer, the obstacle avoidance decision model is sent to the edge service layer, which reduces the occupation of edge computing resources, reduces edge computing costs and network loads; both the twin vehicle and the physical vehicle participate in the obstacle avoidance decision, thereby improving the accuracy of the decision results; in addition, the embodiment of the present invention achieves vehicle obstacle avoidance with less edge computing resource occupancy, which is conducive to promoting the popularization and intelligent development of vehicle networks.

[0067] Combine the following Figure 2 The vehicle obstacle avoidance method of the embodiment of the present disclosure is described in detail.

[0068] In step S210, the twin edge layer trains an obstacle avoidance decision model based on the first operating data and the second operating data, where the first operating data is the operating data of the physical vehicle in the physical space, and the second operating data is the operating data of the twin vehicle in the digital twin space. The twin vehicle is a twin model of the physical vehicle in the digital twin space.

[0069] In the exemplary embodiment of the present disclosure, the structure of the Digital Twins Edge Network (DTEN) is as follows: Figure 3 As shown, the physical space includes an edge service layer and physical vehicles running in the physical space. The edge service layer includes edge service nodes, and the edge service nodes can deploy MEC (Multi-access Edge Computing) devices, RSU (Road Side Unit) devices, etc. The embodiment of the present disclosure does not impose any special restrictions on the number of edge service nodes in the edge service layer and the type of devices deployed by the edge service nodes.

[0070] Edge service nodes and physical vehicles in the physical space all have their own twin images in the digital twin edge network, such as twin edge nodes, twin vehicles, etc. Twin data such as twin edge nodes and twin vehicles together constitute the twin edge layer of the digital twin space. Among them, there is a real-time data channel between the twin edge layer and the edge service layer. The physical vehicles in the physical space and the network entities of the edge service layer send the current operating status to the digital twin space in real time. The digital twin space stores the historical operating data of the physical vehicles and the network entities of the edge service layer. At the same time, the current operating data of the physical vehicles and the network entities of the edge service layer are also collected to monitor the overall operation of the physical space.

[0071] The obstacle avoidance vehicle can perform obstacle avoidance planning based on the obstacle avoidance decision model. The obstacle avoidance decision model of the embodiment of the present disclosure is trained in the twin edge layer.

[0072] In the vehicle obstacle avoidance system of the disclosed embodiment, the physical vehicle generates first operation data when running in the physical space, and the twin vehicle runs simultaneously in the digital twin space and generates second operation data, and the twin edge network trains the obstacle avoidance decision model based on the first operation data and the second operation data. By collecting the operation data of the twin vehicle and the physical vehicle as samples, the twin data and actual data are used to increase the number of samples for the learning algorithm, thereby improving the stability and generalization ability of the obstacle avoidance decision model.

[0073] It is worth noting that the first operation data and the second operation data used in the training of the obstacle avoidance decision model in the embodiment of the present disclosure are from the data of the same operation time. For example, the first operation data is from the data of the physical vehicle running from 10:00 to 12:00, and correspondingly, the second operation data is from the data of the twin vehicle of the physical vehicle in the digital twin space running from 10:00 to 12:00, that is, the collection time of the first operation data and the second operation data is synchronized.

[0074] In an exemplary embodiment, the first operating data and the second operating data may be stored in a storage module of the twin edge layer. The twin edge layer may also include a twin server and a twin edge node, wherein the twin edge node includes a twin simulation platform, the twin simulation platform includes a twin car and a twin environment corresponding to the twin car, and based on the twin simulation platform, the twin server may collaborate with the twin edge node to train and update the obstacle avoidance decision model.

[0075] It should be noted that the obstacle avoidance decision model of the embodiment of the present disclosure may be an A3C (Asynchronous Advantage Actor-Critic) model, a decision tree model or other deep learning models, etc., and there is no special limitation on this.

[0076] In step S220, the twin edge layer sends the obstacle avoidance decision model to the edge service layer, so that the twin vehicle and the physical vehicle make obstacle avoidance decisions based on the obstacle avoidance decision model at the same time, so as to control the physical vehicle to perform the obstacle avoidance process according to the obtained decision results.

[0077] In an exemplary embodiment of the present disclosure, the training and calculation process of the obstacle avoidance decision model is executed in the twin edge layer. After completing the training and update of the obstacle avoidance decision model, the twin edge layer sends the obstacle avoidance decision model to the edge service layer. Based on the obstacle avoidance decision model, the physical vehicle in the physical space and the twin vehicle in the digital twin space can make obstacle avoidance rule decisions based on DTEN.

[0078] The obstacle avoidance decision model is sent to the edge service layer, and the physical vehicle makes obstacle avoidance rule decisions based on the obstacle avoidance decision model. While ensuring the data synchronization between the physical space and the digital twin space, double verification can also be performed based on the decision results of the physical space and the digital twin space.

[0079] In an exemplary embodiment, there is a real-time data channel between the twin edge layer and the edge service layer. During the simultaneous movement of the physical vehicle and the twin vehicle, the edge service layer uploads the first operating data to the twin edge layer based on the real-time data channel, wherein the twin edge layer includes corresponding second operating data.

[0080] In the physical space, the first operation data of the physical vehicle can be collected through the edge node road data collection module, and the first operation data can be uploaded to the digital twin space, for example, to the data storage module of the digital twin space. At the same time, since the movement of the physical vehicle and the twin vehicle is synchronized, the twin vehicle generates second operation data in the twin edge layer.

[0081] The disclosed embodiment stores the operating data generated by the twin vehicle and the physical vehicle in the digital twin space, reducing the storage pressure on the edge service layer.

[0082] In an exemplary embodiment, in order to ensure the integrity and accuracy of the obstacle avoidance decision model sent to the edge service layer, the obstacle avoidance decision model may also be verified and evaluated before the twin edge layer sends the obstacle avoidance decision model to the edge service layer.

[0083] The twin edge layer uses test samples to verify and evaluate the obstacle avoidance decision model to determine whether the obstacle avoidance decision model meets the model issuance conditions. The first operation data and the second operation data are divided into test samples and training samples, and the obstacle avoidance decision model is trained based on the training samples.

[0084] In actual implementation, after the twin edge layer obtains the first operating data and the second operating data, the first operating data and the second operating data are divided into test samples and training samples. For example, 80% of the first operating data and the second operating data are used as training samples, and the remaining 20% ​​are used as test samples. The embodiment of the present disclosure does not impose any special restrictions on the division ratio of training samples and test samples.

[0085] After obtaining the training samples and test samples, the obstacle avoidance decision model can be trained by the twin server of the twin edge layer using the training samples. After the training is completed, the twin server sends the obstacle avoidance decision model to the twin edge node so that the twin edge node can test the obstacle avoidance decision model using the test samples.

[0086] During the test, the twin edge node can use the test samples to test whether the decision accuracy of the current obstacle avoidance decision model is better than the decision accuracy of the previous obstacle avoidance decision model. The twin edge node can also use the test samples to test the integrity of the current obstacle avoidance decision model, for example, whether the obstacle avoidance decision model can fully complete the obstacle avoidance decision and obtain the decision result, etc.

[0087] The model delivery conditions are set according to actual application requirements. For example, the model delivery condition may be that the obstacle avoidance decision model is complete and the decision accuracy reaches a set accuracy threshold. The present disclosure does not make special requirements for this and can be adjusted accordingly according to actual requirements.

[0088] It should be noted that by verifying and evaluating the obstacle avoidance decision model through the twin edge nodes, the accuracy of the obstacle avoidance decision model sent to the physical space can be guaranteed. The embodiments of the present disclosure may also include other model evaluation methods, and no special requirements are made for this.

[0089] In an exemplary embodiment, the twin vehicle in the digital twin space and the real vehicle in the physical space are used as obstacle avoidance vehicles, and both can use the obstacle avoidance decision model to perform the obstacle avoidance process. Figure 4 A flowchart of an obstacle avoidance vehicle making an obstacle avoidance decision based on an obstacle avoidance decision model according to an exemplary embodiment of the present disclosure is shown, referring to Figure 4 As shown, the obstacle avoidance vehicle may make an obstacle avoidance decision based on the obstacle avoidance decision model, including steps S410 to S440:

[0090] Step S410: the obstacle avoiding vehicle makes an obstacle avoiding decision according to the obstacle avoiding decision model to obtain first obstacle avoiding movement information, and determines preview point information based on the first obstacle avoiding movement information and in accordance with the dynamic preview search algorithm.

[0091] In an exemplary embodiment of the present disclosure, the pre-aiming point information is the position information of the obstacle avoiding vehicle after the obstacle avoiding movement is performed according to the first obstacle avoiding movement information, such as three-dimensional coordinate information. The obstacle avoiding vehicle includes a twin vehicle and a physical vehicle, that is, the twin vehicle performs an obstacle avoidance decision-making process in the digital twin space, and the physical vehicle performs an obstacle avoidance decision-making process in the physical space. Among them, the first obstacle avoidance movement information includes relevant information such as overtaking movement, braking, and lane keeping, including but not limited to pre-motion parameter information such as acceleration, movement time, and movement speed. The pre-motion parameter information refers to the relevant motion parameters of the obstacle avoiding vehicle's pre-aiming point information when performing obstacle avoidance movement, such as movement time, acceleration, speed, etc.

[0092] The dynamic preview search algorithm may be, for example, a pure tracking control algorithm such as Pure Pursuit, etc., and there is no special restriction on this. Based on the dynamic preview search algorithm, the obstacle avoidance vehicle can detect the path change ahead in advance according to the road conditions to predict the possible future position information of the obstacle avoidance vehicle.

[0093] Step S420: Calculate the second obstacle avoidance movement information according to the pre-motion parameter information corresponding to the first obstacle avoidance movement information and according to the preset lateral path verification model.

[0094] In an exemplary embodiment of the present disclosure, in addition to determining the preview point information according to the dynamic preview search algorithm, in order to further verify the accuracy of the preview point information, the second obstacle avoidance movement information may also be calculated according to a preset lateral path verification model.

[0095] The second obstacle avoidance movement information can be calculated using the following formula (1):

[0096] Y 横向 =0.5×a×T1 2 +v 横向 ×T1 (1)

[0097] Among them, Y 横向 is the second obstacle avoidance movement information, a is the lateral acceleration in the pre-motion parameter information, v 横向 is the lateral speed in the pre-motion parameter information, and T1 is the lateral motion time in the pre-motion parameter information.

[0098] The second obstacle avoidance movement information is the predicted position information obtained according to the above formula (1), that is, the lateral displacement of the obstacle avoidance vehicle. According to the movement of the obstacle avoidance vehicle according to the lateral displacement, the lateral turning angle of the front wheel can be further obtained.

[0099] It should be noted that the real vehicle and the twin vehicle respectively use formula (1) to calculate the second obstacle avoidance movement information, so that the information between the real vehicle and the twin vehicle is synchronized.

[0100] Step S430: Calculate the third obstacle avoidance movement information according to the pre-motion parameter information corresponding to the first obstacle avoidance movement information, the current vehicle performance information of the obstacle avoidance vehicle and the current road condition information, in accordance with a preset longitudinal path verification model.

[0101] Among them, the current performance information of the obstacle avoidance vehicle is relevant information about the vehicle performance, such as engine parameters, etc., and the current road condition information refers to the information on the degree of influence of the current road condition on the performance of the obstacle avoidance vehicle. The current vehicle performance information and current road condition information of the obstacle avoidance vehicle will affect the longitudinal movement of the obstacle avoidance vehicle, for example, engine parameters, vehicle torque, the influence of the current road condition on the performance of the obstacle avoidance vehicle, etc. affect the longitudinal movement. The embodiment of the present disclosure can use the following formula (2) to calculate the third obstacle avoidance movement information:

[0102] Y 纵向 =λ p ×λ d ×Δt 纵向 T v (2)

[0103] In formula (2), Δt 纵向 is the longitudinal motion time, T v is the current torque of the obstacle avoidance vehicle; p and λ d is a scaling factor, which indicates the influence of engine performance and current road conditions on the longitudinal movement of the vehicle, where λ p is an influencing factor related to engine performance, which is determined according to the engine performance of the obstacle avoidance vehicle, such as engine parameters; d is an influencing factor related to road conditions, which is determined according to the influence of the current road conditions on the performance of the obstacle avoidance vehicle. The better the road conditions, the higher the d For example, the λ of a vehicle traveling on a city road is d Greater than the λ of vehicles traveling on off-road roads d .

[0104] The third obstacle avoidance movement information is the predicted position information obtained according to the above formula (2), that is, the longitudinal displacement of the obstacle avoidance vehicle.

[0105] It should be noted that the real vehicle and the twin vehicle respectively use formula (2) to calculate the third obstacle avoidance movement information, so that the information between the real vehicle and the twin vehicle is synchronized.

[0106] Step S440: Verify the preview point information in combination with the second obstacle avoidance movement information and the third obstacle avoidance movement information. After the preview point information passes the verification, use the preview point information as a decision result.

[0107] The preview aiming point information includes lateral position information and longitudinal position information. In order to further ensure the accuracy of the decision result, the second obstacle avoidance movement information and the third obstacle avoidance movement information can also be used to verify the aiming point information.

[0108] The conditions for the preview point information to pass verification include:

[0109] The deviation of the lateral position information from the second obstacle avoidance movement information is less than a first deviation threshold;

[0110] The deviation of the longitudinal position information from the third obstacle avoidance movement information is less than a second deviation threshold.

[0111] The first deviation threshold and the second deviation threshold are pre-determined allowable deviation values, and the smaller the allowable deviation value is, the higher the required accuracy of the obstacle avoidance movement of the obstacle avoidance vehicle is. If the preview point information passes the verification, the preview point information is used as the decision result of the obstacle avoidance vehicle.

[0112] It should be noted that the twin vehicle and the physical vehicle execute the decision-making process of steps S410 to S440 in their respective spaces and obtain decision results respectively. This process can ensure the information synchronization between the physical vehicle and the twin vehicle.

[0113] Through the process of obtaining the decision result of the embodiment of the present disclosure, a dynamic preview search algorithm is used to determine the preview point information, and the second obstacle avoidance movement information and the third obstacle avoidance movement information are used to verify the decision accuracy of the preview point information, thereby further improving the accuracy of the decision result.

[0114] In an exemplary embodiment, controlling the physical vehicle to perform an obstacle avoidance process according to the obtained decision result may include:

[0115] The twin vehicle moves along the obstacle avoidance path according to the decision result, obtains the first obstacle avoidance result information and sends it to the physical vehicle;

[0116] When the first obstacle avoidance result information indicates that the twin vehicle successfully avoids the obstacle, the physical vehicle moves along the obstacle avoidance path according to the decision result to obtain the second obstacle avoidance result information.

[0117] After the twin edge node in the digital twin space obtains the decision result, the twin vehicle moves along the obstacle avoidance path according to the decision result, simulating the physical vehicle to move along the obstacle avoidance path according to the decision result, and obtaining the first obstacle avoidance result information. The first obstacle avoidance result information may include successful obstacle avoidance and failed obstacle avoidance.

[0118] The twin edge node feeds back the first obstacle avoidance result information to the edge service node, so that the physical vehicle obtains the first obstacle avoidance result information. Since the purpose of feeding back the first obstacle avoidance result information to the physical vehicle is to instruct the physical vehicle to correctly avoid obstacles, only when the first obstacle avoidance result information indicates that the twin vehicle successfully avoids obstacles, the physical vehicle moves along the obstacle avoidance path according to the decision result, and obtains the second obstacle avoidance result information, thereby ensuring that the physical vehicle correctly avoids obstacles.

[0119] In an exemplary embodiment, in order to further ensure that the physical vehicle successfully avoids obstacles, the number of twin vehicles corresponding to the physical vehicle in the twin edge layer is multiple, and each twin vehicle moves along the obstacle avoidance path according to its corresponding decision result to obtain multiple first obstacle avoidance result information. In order for the physical vehicle to successfully avoid obstacles, the first obstacle avoidance result sent by the digital twin space to the physical vehicle indicates that the twin vehicle successfully avoids obstacles. Before the physical vehicle moves along the obstacle avoidance path according to the decision result to obtain the second obstacle avoidance result information, the following process can be performed:

[0120] The twin edge node obtains target first obstacle avoidance result information indicating that the corresponding twin vehicle successfully avoided an obstacle from the plurality of first obstacle avoidance result information;

[0121] The target first obstacle avoidance result information is sent to the physical vehicle, so that the physical vehicle moves along the obstacle avoidance path according to the decision result corresponding to the target first obstacle avoidance result information, and obtains the second obstacle avoidance result information.

[0122] In actual implementation, the digital twin space simulates obstacle avoidance movement through multiple twin vehicles according to their respective decision results. The first obstacle avoidance result information of each twin vehicle may include successful obstacle avoidance and failed obstacle avoidance. The target first obstacle avoidance result information indicating that the twin vehicle has successfully avoided the obstacle is fed back to the physical vehicle, so that the physical vehicle performs the obstacle avoidance process according to the decision result corresponding to the twin vehicle, and the physical vehicle can successfully avoid the obstacle.

[0123] In addition, the obstacle avoidance vehicle is in continuous motion in the physical space and the twin space. After the physical vehicle successfully avoids the obstacle and obtains the second obstacle avoidance result information, the second obstacle avoidance result information can be synchronized to the digital twin space in real time. While the digital twin space stores the historical operation data of the physical vehicle, it can also obtain the obstacle avoidance movement data of the physical vehicle in real time, which is convenient for subsequent obstacle avoidance decisions.

[0124] In an exemplary embodiment, a method for updating the obstacle avoidance decision model is also provided. The twin edge layer trains and updates the obstacle avoidance decision model using the updated first operation data and the updated second operation data according to a set period. The updated first operation data and the updated second operation data are data generated during the synchronous operation of the physical vehicle and the twin vehicle.

[0125] The set period can be one week, one day or one month, which can be selected according to actual needs. Since there is a real-time data channel between the twin edge layer and the edge service layer, the digital twin space obtains the motion data of the physical vehicle and the twin vehicle in real time. In order to maintain the accuracy of the obstacle avoidance decision model, the obstacle avoidance decision model can be regularly trained and updated using the data of the physical vehicle and the twin vehicle during the movement process. The specific details of the update process are the same as the model training and verification evaluation process mentioned above, so they will not be repeated here.

[0126] like Figure 5 The interactive schematic diagram of the various parts of the vehicle obstacle avoidance system involved in the vehicle obstacle avoidance application scenario is shown. The digital twin space includes a data pool for storing the first operating data and the second operating data, a twin server and a twin edge node for training the obstacle avoidance decision model. Among them, the twin car and the twin environment in the twin edge node constitute a twin simulation platform. The physical space includes an edge node road data acquisition module, an edge service node and a physical vehicle.

[0127] Among them, the edge node road data collection module collects the first operating data of the physical vehicle and uploads it to the data pool, and the twin edge node sends the second operating data of the twin vehicle to the data pool. The twin server trains the obstacle avoidance decision model based on the first operating data and the second operating data.

[0128] The twin server sends the trained obstacle avoidance decision model to the twin edge node, verifies and evaluates the obstacle avoidance decision model through the twin edge node, and when it is determined that the obstacle avoidance decision model meets the model delivery conditions, sends the obstacle avoidance decision model to the edge service node, so that the physical vehicle can use the obstacle avoidance decision model to perform the obstacle avoidance process.

[0129] Figure 6 The flowchart of the vehicle obstacle avoidance planning decision based on the obstacle avoidance decision model, the physical vehicle in the physical space and the twin vehicle in the digital twin space are shown. Figure 6 As shown, the physical vehicle 1 performs obstacle avoidance on the physical vehicle 2. The twin model of the physical vehicle 1 in the digital twin space is the twin vehicle 1, and the twin model of the physical vehicle 2 in the digital twin space is the twin vehicle 2. The vehicle obstacle avoidance system also includes a decision module, a dynamic preview point search module, a first vehicle control module, a second vehicle control module, and a decision result determination module. Figure 6 The vehicle obstacle avoidance method according to an embodiment of the present disclosure is described.

[0130] In the digital twin space, twin vehicle 1 makes an obstacle avoidance decision for twin vehicle 2 according to the obstacle avoidance decision model to obtain a decision process. The process is shown in steps S410 to S440 and will not be repeated here.

[0131] Among them, the decision module is used for the twin vehicle 1 to make an obstacle avoidance decision according to the obstacle avoidance decision model to obtain the first obstacle avoidance movement information; the dynamic preview point search module is used to determine the preview point information based on the first obstacle avoidance movement information and according to the dynamic preview search algorithm; the first vehicle control module is used to calculate the second obstacle avoidance movement information according to the pre-motion parameter information corresponding to the first obstacle avoidance movement information and according to the preset lateral path verification model; the second vehicle control module is used to calculate the third obstacle avoidance movement information according to the pre-motion parameter information corresponding to the first obstacle avoidance movement information, the current vehicle performance information of the obstacle avoidance vehicle and the current road condition information according to the preset longitudinal path verification model; the decision result determination module is used to verify the preview point information in combination with the second obstacle avoidance movement information and the third obstacle avoidance movement information, and after the preview point information passes the verification, the preview point information is used as the decision result.

[0132] In the physical space, the twin vehicle 1 makes an obstacle avoidance decision for the twin vehicle 2 according to the obstacle avoidance decision model to obtain a decision result, and the process is shown in steps S410 to S440. For the detailed details of each module in the physical space, please refer to the digital twin space, which will not be repeated here.

[0133] In the digital twin space, twin vehicle 1 successfully avoids twin vehicle 2 according to the decision result, and sends the first obstacle avoidance result information obtained to physical vehicle 1. The first obstacle avoidance result information indicates that twin vehicle 1 successfully avoids the obstacle. Then physical vehicle 1 performs the obstacle avoidance process according to the decision result to avoid physical vehicle 2.

[0134] After the physical vehicle 1 successfully avoids an obstacle, the obstacle avoidance result information is sent to each model, and the cycle is repeated. Then, when the physical vehicle 1 is running in the physical space, it can avoid obstacles encountered.

[0135] From the above, it can be seen that the vehicle obstacle avoidance method of the embodiment of the present invention uses digital twin technology to build a virtual twin, realizes twin vision of twin vehicles, twin servers, etc., implements model training and most decision-making processes in the digital twin space, reduces the occupation of edge computing resources, and reduces the computing cost and load pressure of the edge network. It has practical significance for companies or enterprises to explore and apply the direction of vehicle obstacle avoidance in the Internet of Vehicles.

[0136] In addition, according to an exemplary embodiment of the present disclosure, a vehicle obstacle avoidance system is also provided, such as Figure 7 As shown, the system 700 includes a twin edge layer 710 of a digital twin space and an edge service layer 720 of a physical space.

[0137] The twin edge layer 710 is used to train the obstacle avoidance decision model according to the first operation data and the second operation data, the first operation data is the operation data of the physical vehicle in the physical space, the second operation data is the operation data of the twin vehicle in the twin edge layer 710, and the twin vehicle is a twin model of the physical vehicle in the digital twin space;

[0138] The twin edge layer 710 is also used to send the obstacle avoidance decision model to the edge service layer 720, so that the twin vehicle and the physical vehicle can make obstacle avoidance decisions based on the obstacle avoidance decision model at the same time, so as to control the physical vehicle to perform the obstacle avoidance process according to the obtained decision results.

[0139] In an exemplary embodiment of the present disclosure, the vehicle obstacle avoidance system 700 further includes:

[0140] A decision module, used for the obstacle avoidance vehicle to make an obstacle avoidance decision according to the obstacle avoidance decision model to obtain first obstacle avoidance movement information, the obstacle avoidance vehicle including the twin vehicle and the physical vehicle;

[0141] A dynamic preview point search module is used to determine preview point information based on the first obstacle avoidance movement information and in accordance with a dynamic preview search algorithm, wherein the preview point information is position information of the obstacle avoidance vehicle after performing obstacle avoidance movement according to the first obstacle avoidance movement information;

[0142] a first vehicle control module, configured to calculate second obstacle avoidance movement information according to a preset lateral path verification model based on pre-motion parameter information corresponding to the first obstacle avoidance movement information;

[0143] a second vehicle control module, configured to calculate third obstacle avoidance movement information according to the pre-motion parameter information corresponding to the first obstacle avoidance movement information, the current vehicle performance information of the obstacle avoidance vehicle, and the current road condition information, in accordance with a preset longitudinal path verification model;

[0144] The decision result determination module is used to verify the preview point information in combination with the second obstacle avoidance movement information and the third obstacle avoidance movement information, and after the preview point information passes the verification, use the preview point information as the decision result.

[0145] Since the specific details of each functional module (unit) of the vehicle obstacle avoidance system of the exemplary embodiment of the present disclosure have been described in detail in the inventive embodiment of the above-mentioned vehicle obstacle avoidance method, they will not be repeated here.

[0146] It should be noted that although several modules or units of the vehicle obstacle avoidance system are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided into multiple modules or units to be embodied.

[0147] In addition, in the exemplary embodiment of the present disclosure, a computer storage medium capable of implementing the above method is also provided. A program product capable of implementing the above method of this specification is stored thereon. In some possible embodiments, various aspects of the present disclosure may also be implemented in the form of a program product, which includes a program code, and when the program product is run on a terminal device, the program code is used to enable the terminal device to perform the steps according to various exemplary embodiments of the present disclosure described in the above "Exemplary Method" section of this specification.

[0148] The embodiment of the present disclosure also provides a program product for implementing the above method, which can adopt a portable compact disk read-only memory (CD-ROM) and include program code, and can be run on a terminal device, such as a personal computer. However, the program product of the present disclosure is not limited thereto. In this document, a readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, an apparatus or a device.

[0149] The program product may use any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0150] Computer readable signal media may include data signals propagated in baseband or as part of a carrier wave, in which readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Readable signal media may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0151] The program code embodied on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the foregoing.

[0152] Program code for performing the operations of the present disclosure may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).

[0153] In addition, in an exemplary embodiment of the present disclosure, an electronic device capable of implementing the above method is also provided. It will be appreciated by those skilled in the art that various aspects of the present disclosure may be implemented as a system, method, or program product. Therefore, various aspects of the present disclosure may be specifically implemented in the following forms, namely: a complete hardware embodiment, a complete software embodiment (including firmware, microcode, etc.), or an embodiment combining hardware and software aspects, which may be collectively referred to herein as a "circuit", "module", or "system".

[0154] Refer to the following Figure 8 800 according to this embodiment of the present disclosure is described. Figure 8 The electronic device 800 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0155] like Figure 8 As shown, the electronic device 800 is in the form of a general computing device. The components of the electronic device 800 may include, but are not limited to: the at least one processing unit 810, the at least one storage unit 820, a bus 830 connecting different system components (including the storage unit 820 and the processing unit 810), and a display unit 840.

[0156] The storage unit stores program codes, which can be executed by the processing unit 810, so that the processing unit 810 executes the steps according to various exemplary embodiments of the present disclosure described in the above “Exemplary Method” section of this specification.

[0157] The storage unit 820 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 821 and / or a cache memory unit 822 , and may further include a read-only memory unit (ROM) 823 .

[0158] The storage unit 820 may also include a program / utility 824 having a set (at least one) of program modules 825, such program modules 825 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0159] Bus 830 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0160] The electronic device 800 may also communicate with one or more external devices 900 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 800, and / or communicate with any device that enables the electronic device 800 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 850. Furthermore, the electronic device 800 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 860. As shown, the network adapter 860 communicates with other modules of the electronic device 800 via a bus 830. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 800, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0161] Through the description of the above embodiments, it is easy for those skilled in the art to understand that the example embodiments described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the embodiment of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the embodiment of the present disclosure.

[0162] In addition, the above-mentioned figures are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present disclosure, and are not intended to be limiting. It is easy to understand that the processes shown in the above-mentioned figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be performed synchronously or asynchronously, for example, in multiple modules.

[0163] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. The present disclosure is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary technical means in the art that are not disclosed in the present disclosure. The description and examples are to be considered exemplary only, and the true scope and spirit of the present disclosure are indicated by the claims.

Claims

1. A vehicle obstacle avoidance method, characterized in that: Applied to a vehicle obstacle avoidance system, the vehicle obstacle avoidance system includes a twin edge layer of a digital twin space and an edge service layer of a physical space, and the method includes: The twin edge layer trains an obstacle avoidance decision model according to first operating data and second operating data, wherein the first operating data is operating data of a physical vehicle in the physical space, the second operating data is operating data of a twin vehicle in the digital twin space, and the twin vehicle is a twin model of the physical vehicle in the digital twin space; The twin edge layer sends the obstacle avoidance decision model to the edge service layer, so that the twin vehicle and the physical vehicle simultaneously make obstacle avoidance decisions based on the obstacle avoidance decision model, so as to control the physical vehicle to perform the obstacle avoidance process according to the obtained decision results; Wherein, any obstacle avoidance vehicle makes an obstacle avoidance decision based on the obstacle avoidance decision model to obtain the decision result, including: The obstacle avoiding vehicle makes an obstacle avoiding decision according to the obstacle avoiding decision model to obtain first obstacle avoiding movement information, and determines preview point information based on the first obstacle avoiding movement information and in accordance with a dynamic preview search algorithm, wherein the preview point information is position information of the obstacle avoiding vehicle after the obstacle avoiding movement is performed according to the first obstacle avoiding movement information, and the obstacle avoiding vehicle includes the twin vehicle and the physical vehicle; Calculating second obstacle avoidance movement information according to the pre-motion parameter information corresponding to the first obstacle avoidance movement information and according to a preset lateral path verification model; Calculating third obstacle avoidance movement information according to the pre-motion parameter information corresponding to the first obstacle avoidance movement information, the current vehicle performance information of the obstacle avoidance vehicle, and the current road condition information according to a preset longitudinal path verification model; The preview point information is verified in combination with the second obstacle avoidance movement information and the third obstacle avoidance movement information, and after the preview point information passes the verification, the preview point information is used as the decision result.

2. The method according to claim 1, characterized in that There is a real-time data channel between the twin edge layer and the edge service layer; During the simultaneous movement of the physical vehicle and the twin vehicle, the edge service layer uploads the first operating data to the twin edge layer based on the real-time data channel, wherein the twin edge layer includes the corresponding second operating data.

3. The method according to claim 1, characterized in that Before the twin edge layer sends the obstacle avoidance decision model to the edge service layer, the method further includes: The twin edge layer uses a test sample to verify and evaluate the obstacle avoidance decision model to determine whether the obstacle avoidance decision model meets the model issuance conditions; The first operating data and the second operating data are divided into the test samples and the training samples, and the obstacle avoidance decision model is trained based on the training samples.

4. The method according to claim 1, characterized in that: The preview point information includes horizontal position information and vertical position information; The conditions for the preview point information to pass verification include: The deviation of the lateral position information from the second obstacle avoidance movement information is less than a first deviation threshold; The deviation of the longitudinal position information from the third obstacle avoidance movement information is less than a second deviation threshold.

5. The method according to claim 1, characterized in that The process of controlling the physical vehicle to avoid obstacles according to the obtained decision result includes: The twin vehicle moves along an obstacle avoidance path according to the decision result, obtains first obstacle avoidance result information and sends it to the physical vehicle; When the first obstacle avoidance result information indicates that the twin vehicle successfully avoids the obstacle, the physical vehicle moves along the obstacle avoidance path according to the decision result to obtain second obstacle avoidance result information.

6. The method according to claim 5, characterized in that The number of the twin vehicles in the twin edge layer is multiple, and each of the twin vehicles moves along an obstacle avoidance path according to its corresponding decision result, so as to obtain multiple pieces of the first obstacle avoidance result information; When the first obstacle avoidance result information indicates that the twin vehicle successfully avoids the obstacle, the physical vehicle moves along the obstacle avoidance path according to the decision result, and before obtaining the second obstacle avoidance result information, the method further includes: From the plurality of first obstacle avoidance result information, obtaining target first obstacle avoidance result information indicating that the corresponding twin vehicle successfully avoided an obstacle; The target first obstacle avoidance result information is sent to the physical vehicle, so that the physical vehicle moves along an obstacle avoidance path according to the decision result corresponding to the target first obstacle avoidance result information, and obtains the second obstacle avoidance result information.

7. The method according to any one of claims 1 to 6, characterized in that The method further comprises: The twin edge layer trains and updates the obstacle avoidance decision model using the updated first operation data and the updated second operation data according to a set period; Among them, the updated first operating data and the updated second operating data are data generated during the synchronous operation of the physical vehicle and the twin vehicle.

8. A vehicle obstacle avoidance system, characterized in that: The vehicle obstacle avoidance system includes a twin edge layer of a digital twin space and an edge service layer of a physical space; The twin edge layer is used to train an obstacle avoidance decision model according to first operating data and second operating data, wherein the first operating data is operating data of a physical vehicle in the physical space, the second operating data is operating data of a twin vehicle in the twin edge layer, and the twin vehicle is a twin model of the physical vehicle in the digital twin space; The twin edge layer is used to send the obstacle avoidance decision model to the edge service layer, so that the twin vehicle and the physical vehicle simultaneously make obstacle avoidance decisions based on the obstacle avoidance decision model, so as to control the physical vehicle to perform the obstacle avoidance process according to the obtained decision results; Wherein, the vehicle obstacle avoidance system further includes: A decision module, used for the obstacle avoidance vehicle to make an obstacle avoidance decision according to the obstacle avoidance decision model to obtain first obstacle avoidance movement information, wherein the obstacle avoidance vehicle includes the twin vehicle and the physical vehicle; A dynamic preview point search module, configured to determine preview point information based on the first obstacle avoidance movement information and in accordance with a dynamic preview search algorithm, wherein the preview point information is position information of the obstacle avoidance vehicle after performing obstacle avoidance movement according to the first obstacle avoidance movement information; a first vehicle control module, configured to calculate second obstacle avoidance movement information according to the pre-motion parameter information corresponding to the first obstacle avoidance movement information and in accordance with a preset lateral path verification model; a second vehicle control module, configured to calculate third obstacle avoidance movement information according to the pre-motion parameter information corresponding to the first obstacle avoidance movement information, the current vehicle performance information of the obstacle avoidance vehicle, and the current road condition information, in accordance with a preset longitudinal path verification model; A decision result determination module is used to verify the preview point information in combination with the second obstacle avoidance movement information and the third obstacle avoidance movement information, and after the preview point information passes the verification, use the preview point information as the decision result.

Images