Vehicle passing decision generation method and device, equipment and storage medium

Abstract

The invention discloses a vehicle passing decision generation method, device and equipment and a storage medium, and relates to the technical field of vehicle communication control, and the vehicle passing decision generation method comprises the steps: obtaining local environment perception information, road network perception information and first vehicle cooperative communication information sent by a plurality of surrounding vehicles; performing information fusion on the local environment perception information, the road network perception information and the cooperative communication information of each first vehicle to obtain unified view state information; and generating a vehicle passing decision corresponding to the target vehicle based on the unified view state information. According to the method, the cooperative communication information of the surrounding vehicles is received and processed, and the local environment sensing information and the road network sensing information are integrated, so that efficient cooperative communication among multiple vehicles is realized, the vehicles can obtain a comprehensive view of the surrounding environment, a more reasonable and optimized vehicle passing decision is made, and the vehicle passing efficiency is improved. And the road network traffic capacity is effectively improved.

Description

Technical Field

[0001] The present application relates to the technical field of vehicle communication control, and particularly to a method, device, equipment and storage medium for generating vehicle passing decisions. Background Art

[0002] With the rapid development of vehicle networking technology and fifth-generation communication technology, vehicles on the transportation network have the ability to conduct low-latency, highly reliable, and high-bandwidth communication, providing a basic guarantee for communication between vehicles.

[0003] However, most current technologies still remain in single-vehicle intelligent autonomous driving. Due to the limited observation range of single intelligent vehicles, they are unable to perceive global state information and the driving dynamic information of other vehicles, and thus cannot achieve cooperative driving with other vehicles. As a result, the improvement of the road network passing capacity (i.e., the maximum number of vehicles that the road network can safely and efficiently pass through within a certain period of time) is very limited, leading to the inability to achieve efficient cooperative communication between multiple vehicles, the generated vehicle passing decisions are limited, and it is impossible to accurately judge the next action decision of the vehicle, thereby reducing the road network passing capacity. Summary of the Invention

[0004] The main purpose of the present application is to provide a method, device, equipment and storage medium for generating vehicle passing decisions, aiming to solve the problem that single intelligent vehicles have a limited observation range and are unable to perceive global state information and the driving dynamic information of other vehicles, and thus cannot achieve cooperative driving with other vehicles.

[0005] To achieve the above purpose, the present application proposes a method for generating vehicle passing decisions, and the method includes:

[0006] Obtain local environment perception information, road network perception information, and first vehicle cooperative communication information sent by several surrounding vehicles;

[0007] Perform information fusion on the local environment perception information, the road network perception information, and each of the first vehicle cooperative communication information to obtain unified view state information;

[0008] Generate a vehicle passing decision corresponding to the target vehicle based on the unified view state information.

[0009] In an embodiment, the performing information fusion on the local environment perception information, the road network perception information, and each of the first vehicle cooperative communication information to obtain unified view state information includes:

[0010] Perform data preprocessing on the local environment perception information, the road network perception information, and each of the first vehicle cooperative communication information;

[0011] Extract features from the preprocessed local environment perception information, road network perception information, and the collaborative communication information of each first vehicle to obtain a number of target feature information;

[0012] Perform information fusion on each of the target feature information to obtain the unified view state information.

[0013] In one embodiment, the performing information fusion on each of the target feature information to obtain the unified view state information includes:

[0014] Determine the association relationship, time relationship, and spatial relationship between each of the target feature information;

[0015] Perform context understanding on each of the target feature information to determine the logical relationship corresponding to each of the target feature information;

[0016] Perform information fusion based on the association relationship, time relationship, spatial relationship, and logical relationship corresponding to each of the target feature information to obtain the unified view state information.

[0017] In one embodiment, the generating a vehicle passing decision corresponding to the target vehicle based on the unified view state information includes:

[0018] Obtain the vehicle collaboration granularity, vehicle destination information, and vehicle real-time driving information;

[0019] Perform route analysis on the vehicle destination information and the unified view state information to determine the optimal driving route and the current traffic condition corresponding to the target vehicle;

[0020] Generate a vehicle passing decision corresponding to the target vehicle based on the vehicle collaboration granularity, the optimal driving route, the current traffic condition, and the vehicle real-time driving information.

[0021] In one embodiment, the generating a vehicle passing decision corresponding to the target vehicle based on the vehicle collaboration granularity, the optimal driving route, the current traffic condition, and the vehicle real-time driving information includes:

[0022] Input the vehicle collaboration granularity, the optimal driving route, the current traffic condition, and the vehicle real-time driving information into the upper layer network of the vehicle decision-making generation network to obtain a number of vehicle passing sub-goals output by the upper layer network;

[0023] Input each of the vehicle passing sub-goals into the lower layer network of the vehicle decision-making generation network to obtain a number of vehicle control actions corresponding to each of the vehicle passing sub-goals output by the lower layer network;

[0024] Generate the vehicle passing decision based on each of the vehicle passing sub-goals and the vehicle control actions corresponding to each of the vehicle passing sub-goals.

[0025] In one embodiment, before obtaining the local environment perception information, the road network perception information, and the first vehicle cooperative communication information sent by a plurality of surrounding vehicles, it further includes:

[0026] Obtain the current vehicle state information, the bandwidth resource limit, and the communication medium resource limit, and identify the value information in the current vehicle state information;

[0027] Based on the bandwidth resource limit, compress the value information to generate the second vehicle cooperative communication information;

[0028] Based on the communication medium resource limit, send the second vehicle cooperative communication information to a plurality of surrounding vehicles.

[0029] In one embodiment, after generating the vehicle passing decision corresponding to the target vehicle based on the unified view state information, it further includes:

[0030] Push the vehicle passing decision to the target user for the target user to confirm;

[0031] If a decision execution instruction from the target user is received, convert the vehicle passing decision into a vehicle control instruction;

[0032] Based on the vehicle control instruction, perform vehicle control on the target vehicle.

[0033] In addition, to achieve the above object, the present application further provides a vehicle passing decision generation device, and the vehicle passing decision generation device includes:

[0034] An information acquisition module, configured to acquire local environment perception information, road network perception information, and first vehicle cooperative communication information sent by a plurality of surrounding vehicles;

[0035] An information fusion module, configured to perform information fusion on the local environment perception information, the road network perception information, and each of the first vehicle cooperative communication information to obtain unified view state information;

[0036] A decision generation module, configured to generate a vehicle passing decision corresponding to the target vehicle based on the unified view state information.

[0037] In addition, to achieve the above object, the present application further provides a vehicle passing decision generation device, and the device includes: a memory, a processor, and a computer program stored on the memory and executable on the processor, and the computer program is configured to implement the steps of the vehicle passing decision generation method as described above.

[0038] In addition, to achieve the above object, the present application further provides a storage medium, which is a computer-readable storage medium. A computer program is stored on the storage medium, and when the computer program is executed by a processor, the steps of the vehicle passing decision generation method described above are implemented.

[0039] In addition, to achieve the above object, the present application further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps of the vehicle passing decision generation method described above are implemented.

[0040] The present application provides a vehicle passing decision generation method, device, equipment and storage medium. The vehicle passing decision generation method obtains local environment perception information, road network perception information and first vehicle cooperative communication information sent by a plurality of surrounding vehicles, and then performs information fusion on the local environment perception information, the road network perception information and each first vehicle cooperative communication information to obtain unified view state information. Based on the unified view state information, a vehicle passing decision corresponding to the target vehicle is generated, thereby realizing efficient cooperative communication between multiple vehicles, enabling the vehicle to obtain a comprehensive view of the surrounding environment, formulating more reasonable and optimized vehicle passing decisions, and effectively improving the road network passing capacity. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The accompanying drawings herein are incorporated into the specification and form a part of the specification, showing embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application.

[0042] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0043] Figure 1 It is a schematic flowchart provided for Embodiment 1 of the vehicle passing decision generation method of the present application;

[0044] Figure 2 It is a schematic diagram of a multi-agent cooperation framework in a vehicle-road cooperation scenario provided for the vehicle passing decision generation method of the present application;

[0045] Figure 3 It is a schematic flowchart provided for Embodiment 2 of the vehicle passing decision generation method of the present application;

[0046] Figure 4 It is a schematic flowchart provided for Embodiment 3 of the vehicle passing decision generation method of the present application;

[0047] Figure 5 It is a schematic diagram of the module structure of the vehicle passing decision generation device according to an embodiment of the present application;

[0048] Figure 6 It is a schematic diagram of the device structure of the hardware operating environment involved in the vehicle passing decision generation method according to an embodiment of the present application.

[0049] The realization of the purpose, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. Detailed implementation manners

[0050] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0051] In order to better understand the technical solutions of the present application, the following will be described in detail in conjunction with the accompanying drawings of the specification and specific implementation manners.

[0052] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device, a big data service platform, a vehicle passing decision generation system, etc. that can implement the above functions. Hereinafter, taking the vehicle passing decision generation system as an example, this embodiment and the following embodiments will be described.

[0053] Based on this, an embodiment of the present application provides a vehicle passing decision generation method, referring to Figure 1 , Figure 1 It is a schematic flowchart provided for Embodiment 1 of the vehicle passing decision generation method of the present application.

[0054] In this embodiment, the vehicle passing decision generation method includes steps S11 to S13:

[0055] Step S11, obtaining local environment perception information, road network perception information, and first vehicle cooperative communication information sent by a plurality of surrounding vehicles;

[0056] It should be noted that the local environment perception information refers to the surrounding environment information directly obtained by the vehicle through its own mounted sensors, including the position, size and speed of obstacles; the position and possible movement trajectories of pedestrians and bicycles; the state of lane lines, traffic signs and traffic lights; the road surface condition; the distance from surrounding vehicles, etc., which are not limited here. Additionally, the road network perception information refers to the information about the entire road network state obtained by the vehicle, usually from roadside devices or traffic management systems, including traffic flow and congestion conditions, road construction or closure information, the timing and state of traffic lights, statistical information of other road users, etc.

[0057] It should be further noted that the surrounding vehicles refer to other vehicles within a certain range near the vehicle, such as surrounding autonomous vehicles, etc. Herein, the range is determined by the communication medium resource limitation. The vehicle is regarded as an agent, so as to achieve the purpose of saving communication medium resources and improving the effectiveness of multi-agent communication. The first vehicle collaborative communication information refers to the information exchanged between vehicles through vehicle networking technology, which is used for collaborative driving and improving driving safety, including the current states of surrounding vehicles, such as speed, position, and direction, etc.; the driving plans of the vehicle, such as upcoming turns or lane changes, etc.; the sensor data of the vehicle, such as the detection results of obstacles, etc., which are not limited herein.

[0058] Specifically, local environment perception information, road network perception information, and the first vehicle collaborative communication information sent by several surrounding vehicles are obtained. In one embodiment, information acquisition can be performed through an environment perception layer. The environment perception layer mainly includes an agent communication module, a local environment perception module, and a road network information perception module. The three modules obtain the environmental information of the target vehicle in different ways. Among them, the agent communication module exchanges local vehicle information in the digital space, so as to obtain the first vehicle collaborative communication information sent by several surrounding vehicles. Reference can be made to Figure 2 , Figure 2 which is a schematic diagram of a multi-agent collaboration framework in a vehicle-road collaboration scenario provided for the vehicle passing decision generation method of this application; the local environment perception module perceives the local surrounding environment through the sensors of the target vehicle, so as to obtain the local environment perception information around the target vehicle, such as information about surrounding vehicles, roads, pedestrians, and non-motor vehicles, etc., enabling the intelligent vehicle to perceive the surrounding environment, ensuring driving safety, and at the same time sharing it as local observation information with other intelligent vehicles; the road network information perception module is mainly responsible for perceiving road network information based on a high-precision map, so as to obtain the detailed information of the road on which the intelligent vehicle travels (i.e., road network perception information), so that the target vehicle can make action decisions based on local observation information and multi-agent collaborative communication shared information, so as to achieve the safe passing of the intelligent vehicle on various roads such as non-traffic-light intersections, and at the same time the target vehicle can optimize its driving behavior by integrating overall path planning information and traffic flow information, improving the passing efficiency of the target vehicle.

[0059] Step S12: Perform information fusion on the local environment perception information, the road network perception information, and each first vehicle collaborative communication information to obtain unified view state information;

[0060] It should be noted that the unified view state information refers to the complete view of the vehicle's surrounding environment formed by fusing local environment perception information, road network perception information, and first vehicle cooperative communication information, including comprehensive vehicle surrounding obstacle information, comprehensive traffic conditions and signal information, and comprehensive other vehicle state and intention information. For reference, Figure 2 the perspective view of the virtual vehicle group in Figure 2 is a schematic diagram of the multi-agent cooperation framework in the vehicle-road cooperation scenario provided for the vehicle passing decision generation method of this application.

[0061] Specifically, data preprocessing is performed on the local environment perception information, the road network perception information, and each first vehicle cooperative communication information, and then feature extraction is performed on the preprocessed local environment perception information, road network perception information, and each first vehicle cooperative communication information to obtain a number of target feature information. Thus, each target feature information is fused to obtain the unified view state information.

[0062] Step S13: Generate a vehicle passing decision corresponding to the target vehicle based on the unified view state information.

[0063] It should be noted that the target vehicle refers to the vehicle that needs to make a passing decision in the vehicle-road cooperation scenario, that is, the vehicle currently performing information processing and decision-making, such as an autonomous vehicle.

[0064] Furthermore, it should be noted that the vehicle passing decision refers to the driving decision made by the target vehicle based on the unified view state information, including selecting a driving route, such as going straight, turning left, or turning right; adjusting the driving speed to adapt to the traffic flow or avoid obstacles; performing obstacle avoidance operations, such as decelerating or changing lanes to avoid collisions; preparing to pass through an intersection, including stopping and waiting or passing according to the signal lights. For example, if the target vehicle detects a pedestrian crossing the road ahead and receives the information that the traffic signal is about to turn red, the vehicle passing decision will include decelerating and stopping, waiting for the pedestrian to pass, and then continuing to drive according to the signal light state. The specific vehicle passing decision can be generated according to the actual situation and is not limited here.

[0065] Specifically, obtain the vehicle cooperation granularity, vehicle destination information, and vehicle real-time driving information, and then perform route analysis on the vehicle destination information and the unified view state information to determine the optimal driving path corresponding to the target vehicle and the current traffic conditions. Thus, based on the vehicle cooperation granularity, the optimal driving path, the current traffic conditions, and the vehicle real-time driving information, generate the vehicle passing decision corresponding to the target vehicle.

[0066] In this embodiment, by obtaining local environment perception information, road network perception information, and first vehicle cooperative communication information sent by a number of surrounding vehicles, and then performing information fusion on the local environment perception information, the road network perception information, and each of the first vehicle cooperative communication information to obtain unified view state information, and thus generating a vehicle passing decision corresponding to the target vehicle based on the unified view state information, and further realizing efficient cooperative communication among multiple vehicles, so that the vehicle can obtain a comprehensive view of the surrounding environment, formulate more reasonable and optimized vehicle passing decisions, and effectively improve the road network passing capacity.

[0067] In a feasible implementation manner, the performing information fusion on the local environment perception information, the road network perception information, and each of the first vehicle cooperative communication information to obtain unified view state information includes:

[0068] Step S21, performing data preprocessing on the local environment perception information, the road network perception information, and each of the first vehicle cooperative communication information;

[0069] It should be noted that since the local environment perception information, the road network perception information, and each of the first vehicle cooperative communication information come from different sensors and data sources, it is necessary to integrate this information, such as steps of time synchronization, spatial alignment, and data format unification, to ensure that all information is analyzed under the same reference framework.

[0070] Specifically, performing data preprocessing on the local environment perception information, the road network perception information, and each of the first vehicle cooperative communication information, wherein the preprocessing operations include cleaning: removing noise and incorrect data; standardization: converting data from different sources into a unified format and unit; calibration: ensuring that data from different sensors is aligned in time and space, and there is no limitation here, and it can be set according to the actual situation.

[0071] Step S22, performing feature extraction on the preprocessed local environment perception information, road network perception information, and each of the first vehicle cooperative communication information to obtain a number of target feature information;

[0072] It should be noted that the target feature information refers to multi-dimensional information extracted from the original local environment perception information, road network perception information, and vehicle cooperative communication information, which is of great significance for the vehicle to make passing decisions, such as the speed, acceleration, direction of surrounding vehicles, the distance and size of obstacles, etc., thus representing the essential attributes of the vehicle's surrounding environment and providing a comprehensive, accurate, and real-time information basis for the vehicle's decision control layer.

[0073] Specifically, perform feature recognition on the preprocessed local environment perception information, road network perception information, and each first vehicle cooperative communication information to determine which information is most critical for the vehicle's decision-making process. This involves knowledge base recognition of domain experts, historical data analysis, or machine learning techniques, etc., to identify the feature information most relevant to vehicle behavior and environmental interaction. Then, apply a feature extraction algorithm to extract feature information from the preprocessed data. The feature extraction algorithms include statistical methods, signal processing techniques, image processing techniques, or machine learning algorithms, such as principal component analysis, independent component analysis, or autoencoders, etc., which are not limited here.

[0074] Furthermore, since there may be redundant or irrelevant features in the extracted feature information, through feature selection techniques, such as recursive feature elimination, model-based feature selection, or heuristic-based methods, select the features that are most helpful for vehicle decision-making, and perform further processing on the selected features to improve their usefulness for the decision-making process, including steps such as feature scaling, feature encoding, feature construction, or dimensionality reduction, etc., so as to obtain several target feature information, and extract the target feature information crucial for vehicle decision-making from complex multi-source data to support the vehicle to make more intelligent and safe driving decisions.

[0075] Step S23, perform information fusion on each of the target feature information to obtain the unified view state information.

[0076] Specifically, determine the association relationship, time relationship, and spatial relationship between each of the target feature information, and then perform context understanding on each of the target feature information to determine the logical relationship corresponding to each of the target feature information. Thus, perform information fusion based on the association relationship, time relationship, spatial relationship, and logical relationship corresponding to each of the target feature information to obtain the unified view state information.

[0077] In this embodiment, through data preprocessing of the local environment perception information, the road network perception information, and each first vehicle cooperative communication information, then perform feature extraction on the preprocessed local environment perception information, road network perception information, and each first vehicle cooperative communication information to obtain several target feature information. Thus, perform information fusion on each of the target feature information to obtain the unified view state information, and then remove noise and inaccurate data, unify the data format, thereby simplifying the data, reducing the burden of subsequent calculations, and improving the calculation efficiency, thus improving the quality and reliability of the information. At the same time, integrate duplicate or redundant data, reduce the data volume, lower the processing complexity, and identify the information most useful for decision-making, improve the accuracy of decision-making, and then improve the cooperative efficiency between vehicles to achieve better vehicle driving group behavior.

[0078] Based on this, an embodiment of the present application provides a method for generating a vehicle passing decision. Refer to Figure 3 , Figure 3 which is a schematic flowchart provided for Embodiment 2 of the method for generating a vehicle passing decision of the present application.

[0079] In a feasible implementation manner, the information fusion of each of the target feature information to obtain the unified view state information includes:

[0080] Step S31, determining the association relationship, time relationship, and spatial relationship between each of the target feature information;

[0081] It should be noted that the association relationship refers to the mutual connection and dependence between different feature information. For example, if vehicle A is accelerating, then it will affect the driving strategies of surrounding vehicles B and C. The time relationship refers to the relationship between feature information changing over time, including the timeliness and dynamic changes of the information. For example, the speed and acceleration of a vehicle change over time, and their historical values and predicted values need to be considered. Additionally, the spatial relationship refers to the distribution and positional relationship of feature information in space. For example, the distance and relative position between vehicle A and vehicle B will affect their interaction.

[0082] Specifically, determining the association relationship, time relationship, and spatial relationship between each of the target feature information, where it can be determined through a machine learning model. For example, using algorithms such as the Apriori algorithm to discover frequent item sets and association rules between features, using clustering algorithms (such as K-means (K-means clustering), DBSCAN (Density-Based Spatial Clustering of Applications with Noise)) to identify the spatial relationship between features, and using time series models (such as ARIMA (AutoRegressive Integrated Moving Average), LSTM (Long Short-Term Memory)) to analyze the changes of features over time.

[0083] Step S32, performing context understanding on each of the target feature information to determine the logical relationship corresponding to each of the target feature information;

[0084] It should be noted that the logical relationship refers to the causal relationship and conditional relationship between target feature information, which is usually used to describe decision-making logic. For example, if vehicle A detects a pedestrian crossing the road ahead (condition), then it should decelerate or stop (result); if the traffic signal is red (condition), then the vehicle must stop and wait (result).

[0085] Specifically, machine learning algorithms such as association rule learning, clustering analysis, and causal inference are applied to discover potential patterns and relationships between features. At the same time, a knowledge graph is constructed to represent the logical connections and semantic relationships between features, so as to understand the context of each target feature information, and the rule engine is used to implement feature relationship reasoning based on predefined rules, thereby determining the logical relationship corresponding to each target feature information, and further achieving an in-depth understanding of the context meaning of the target feature information, accurately grasping the logical relationship between features, thereby providing strong information support for vehicle decision-making, improving the intelligence level and operation efficiency of the traffic system.

[0086] Step S33, perform information fusion based on the association relationship, time relationship, space relationship, and logical relationship corresponding to each target feature information to obtain the unified view state information.

[0087] Specifically, data fusion techniques such as Kalman filtering, particle filtering, or Bayesian filtering are used to perform information fusion on the association relationship, time relationship, space relationship, and logical relationship corresponding to each target feature information to obtain unified view state information, so as to obtain more accurate and reliable information.

[0088] In this embodiment, by determining the association relationship, time relationship, and space relationship between each target feature information, and then understanding the context of each target feature information, determining the logical relationship corresponding to each target feature information, and thus performing information fusion based on the association relationship, time relationship, space relationship, and logical relationship corresponding to each target feature information to obtain the unified view state information, and then understanding and analyzing the complex association relationships between different features, so as to more effectively allocate the vehicle's attention and processing resources, prioritize the information that has the greatest impact on the current decision, thereby more accurately predicting traffic conditions and potential risks, improving the accuracy of decision-making and the reliability of the decision-making process, reducing misjudgments caused by errors in a single information source, and further better understanding the traffic situation where the target vehicle is located, including the intentions and possible actions of surrounding vehicles, so as to support more complex decision-making in complex traffic scenarios such as intersection passage and lane change and overtaking, and finally generate a comprehensive, accurate, and real-time unified view state information, providing strong information support for the decision-making control layer of the vehicle, and thus improving the intelligence level of the vehicle and the operation efficiency of the traffic system.

[0089] Based on this, an embodiment of the present application provides a method for generating a vehicle passing decision. Referring to Figure 4 , Figure 4 which is a schematic flowchart provided for Embodiment 3 of the vehicle passing decision generation method of the present application.

[0090] In a feasible implementation manner, generating a vehicle passing decision corresponding to a target vehicle based on the unified view state information includes:

[0091] Step S41, obtaining the vehicle cooperation granularity, vehicle destination information, and vehicle real-time driving information;

[0092] It should be noted that the vehicle cooperation granularity refers to the degree of detail of the information involved and the refinement degree of the decision-making when information exchange and decision-making are carried out between intelligent agents (multiple vehicles) during the multi-agent (multiple vehicles) cooperation process. Among them, the larger the granularity, the more detailed the information and the more refined the decision-making; the smaller the granularity, the more simplified the information and the more rough the decision-making. For example, the granularity can be coarse (such as only considering the main roads and key intersections) or fine-grained (such as considering the specific states of each lane and each traffic signal), which determines the degree of detail of the information exchanged between vehicles and between vehicles and infrastructure, as well as the complexity of the cooperative decision-making.

[0093] Furthermore, it should be noted that the vehicle destination information refers to the final position that the vehicle plans to go to, usually a specific address or geographical coordinate. The vehicle real-time driving information refers to including the current speed, acceleration, driving direction, position, and possible driving intentions of the vehicle, which helps to make real-time driving decisions.

[0094] Specifically, for a specific cooperation task, selecting a suitable vehicle cooperation granularity is crucial. When the vehicle cooperation granularity is too large, the intelligent agent cannot learn effective task decomposition strategies, and the cooperation efficiency will be greatly reduced; when the vehicle cooperation granularity is too small, it will lead to too high a communication cooperation frequency between intelligent agents, and the saving of communication resources is not obvious. Only by setting a reasonable cooperation granularity for the upper-layer network can the communication resources and cooperation efficiency be optimized simultaneously. However, finding the optimal cooperation granularity requires a large number of experiments for verification, consuming a large amount of computing costs and labor costs. At the same time, in the real traffic scenario, the complexity of the vehicle-road cooperation task is not always single and unchanged but dynamically changing. Many factors such as the shape of the traffic road network, the number of vehicles in the cooperation domain, the passing goals of vehicles, and the aggregation degree of vehicles will affect the complexity of the cooperation task, and these factors are all dynamically changing. This poses a need for adaptive cooperation granularity control for the hierarchical communication cooperation algorithm (i.e., the upper-layer network and the lower-layer network in the following text). Thus, through adaptive cooperation granularity control, the cooperation granularity can be dynamically adjusted to ensure high-efficiency cooperation in different traffic scenarios and improve the road network passing capacity.

[0095] Furthermore, a unified vehicle cooperation granularity convention is carried out through the central controller, that is, messages from all agents are received through the cooperation control network to form a perception of the global traffic complexity, and then a cooperation control variable is generated to control the vehicle cooperation granularity of all agents. Thus, through this cooperation granularity generation method, a large number of pre-experiments required for manual setting can be avoided, enabling agents to learn the optimal cooperation granularity setting while training the cooperation strategy. At the same time, when the cooperation control network is well-trained, the cooperation cycle can be set according to the task complexity of the current cooperation cycle to achieve complexity-adaptive swarm intelligence cooperation, enhancing the flexibility of cooperation.

[0096] Even further, vehicle destination information is obtained through user input, and vehicle real-time driving information is obtained through local sensors. There is no limitation here and it can be set according to the actual situation.

[0097] Step S42: Analyze the route between the vehicle destination information and the unified view state information to determine the optimal driving path corresponding to the target vehicle and the current traffic condition;

[0098] It should be noted that the optimal driving path refers to the most effective route calculated based on the current traffic condition, vehicle destination, and vehicle cooperation granularity, aiming to minimize driving time, distance, or energy consumption, while considering traffic rules, road conditions, and potential traffic congestion.

[0099] Further, it should be noted that the current traffic condition includes real-time information such as vehicle density, speed, accidents, construction, traffic signal status, etc. on the road, which helps to evaluate the congestion degree of the road and predict future traffic changes.

[0100] Specifically, analyze the route between the vehicle destination information and the unified view state information to determine the optimal driving path corresponding to the target vehicle and the current traffic condition. Among them, path planning algorithms such as A*, Dijkstra, etc. can be applied to calculate the optimal path from the current position to the destination and determine the current traffic condition corresponding to the target vehicle. In addition, the system also needs to monitor and analyze the current traffic condition in real time to predict possible changes and dynamically adjust the path planning, so as to guide the vehicle to reach the destination safely and efficiently.

[0101] Step S43: Generate a vehicle passing decision corresponding to the target vehicle based on the vehicle cooperation granularity, the optimal driving path, the current traffic condition, and the vehicle real-time driving information.

[0102] Specifically, input the vehicle collaboration granularity, the optimal driving path, the current traffic condition, and the vehicle real-time driving information into the upper network of the vehicle decision-making generation network to obtain several vehicle passing sub-goals output by the upper network. Then, input each of the vehicle passing sub-goals into the lower network of the vehicle decision-making generation network to obtain several vehicle control actions corresponding to each of the vehicle passing sub-goals output by the lower network. Thus, based on each of the vehicle passing sub-goals and the vehicle control actions corresponding to each of the vehicle passing sub-goals, generate the vehicle passing decision.

[0103] In this embodiment, by obtaining the vehicle collaboration granularity, vehicle destination information, and vehicle real-time driving information, and then performing route analysis on the vehicle destination information and the unified view state information to determine the optimal driving path and the current traffic condition corresponding to the target vehicle. Thus, based on the vehicle collaboration granularity, the optimal driving path, the current traffic condition, and the vehicle real-time driving information, generate the vehicle passing decision corresponding to the target vehicle. Furthermore, by planning the optimal driving path for the vehicle, the driving distance and time can be reduced, and it helps to avoid congested sections, improve the road use efficiency and vehicle passing speed. At the same time, it helps the vehicle avoid potential collision risks and improve driving safety. Among them, by considering the traffic condition and collaboration granularity, it helps to balance the traffic flow, reduce traffic congestion, and helps to achieve effective collaboration between vehicles and improve the collaboration efficiency of the entire traffic system. In addition, by optimizing the driving path and speed, unnecessary acceleration and braking can be reduced, thereby improving the energy utilization efficiency. It can also dynamically adjust the driving path and passing decision according to the real-time change of the traffic condition, and then improve the adaptability of the system.

[0104] In a feasible implementation manner, the generating the vehicle passing decision corresponding to the target vehicle based on the vehicle collaboration granularity, the optimal driving path, the current traffic condition, and the vehicle real-time driving information includes:

[0105] Step S51, input the vehicle collaboration granularity, the optimal driving path, the current traffic condition, and the vehicle real-time driving information into the upper network of the vehicle decision-making generation network to obtain several vehicle passing sub-goals output by the upper network;

[0106] It should be noted that the vehicle decision-making generation network refers to a system or model for generating vehicle driving decisions, which usually adopts artificial intelligence and machine learning technologies. This network can process various input information, such as vehicle destinations, real-time traffic conditions, vehicle states, etc., and output corresponding driving strategies and actions. Among them, the upper-layer network refers to a part of the vehicle decision-making generation network, which is responsible for formulating macroscopic or coarse-grained decisions, usually dealing with higher-level tasks, such as path planning, decision-making, and determining the general direction of the driving strategy.

[0107] It should be further noted that the vehicle passing sub-goal refers to the intermediate goal output by the upper-layer network, which is used to guide the vehicle's behavior in specific stages or specific scenarios during driving, reflecting the specific driving requirements that the vehicle needs to achieve in different stages, and may include short-term driving directions, speed adjustments, expected driving actions, etc.

[0108] Specifically, input the vehicle collaboration granularity, the optimal driving path, the current traffic condition, and the vehicle's real-time driving information into the upper-layer network in the vehicle decision-making generation network to obtain several vehicle passing sub-goals output by the upper-layer network. Thus, since only the decisions of the upper-layer network need to be communicated and coordinated, the communication resources and computing resources required for collaborative computing are reduced.

[0109] Step S52: Input each of the vehicle passing sub-goals into the lower-layer network in the vehicle decision-making generation network to obtain several vehicle control actions corresponding to each of the vehicle passing sub-goals output by the lower-layer network;

[0110] It should be noted that the lower-layer network refers to another part of the vehicle decision-making generation network, which is responsible for converting the sub-goals of the upper-layer network into specific vehicle control actions. The lower-layer network deals with lower-level tasks, such as precise speed control, steering angle adjustment, etc., to ensure that the vehicle can drive safely and effectively according to the guidance of the upper-layer network.

[0111] It should be further noted that the vehicle control action refers to the specific instruction output by the lower-layer network, which is used to directly control the vehicle's hardware system, such as the engine, braking system, steering system, etc. Vehicle control actions include but are not limited to accelerating, decelerating, stopping, steering, etc.

[0112] Step S53: Generate the vehicle passing decision based on each of the vehicle passing sub-goals and the vehicle control actions corresponding to each of the vehicle passing sub-goals.

[0113] Specifically, by integrating the vehicle passing sub-goals from the upper-layer network and the vehicle control actions from the lower-layer network, a comprehensive and hierarchical vehicle passing decision is generated, which not only considers the immediate driving needs of the vehicle but also the safety, efficiency, and comfort during the entire driving process, ensuring that the vehicle can intelligently and automatically navigate from the current position to the destination.

[0114] In this embodiment, the vehicle cooperation granularity, the optimal driving path, the current traffic condition, and the vehicle real-time driving information are input into the upper-layer network in the vehicle decision-making generation network to obtain several vehicle passing sub-goals output by the upper-layer network. Then, each vehicle passing sub-goal is input into the lower-layer network in the vehicle decision-making generation network to obtain several vehicle control actions corresponding to each vehicle passing sub-goal output by the lower-layer network. Based on each vehicle passing sub-goal and the vehicle control actions corresponding to each vehicle passing sub-goal, the vehicle passing decision is generated. Furthermore, information is processed in parallel at different levels. The upper-layer network is responsible for formulating coarse-grained passing sub-goals, and the lower-layer network is responsible for specific vehicle control actions. Since only the decisions of the upper-layer network need to communicate and cooperate, the communication resources and computing resources required for cooperative computing are reduced, achieving the simultaneous optimization of communication resources and cooperative efficiency, making the decision more refined, improving the decision-making efficiency, enhancing the adaptability to complex traffic environments, thus more accurately predicting and responding to changes in the traffic environment, and improving the accuracy of the decision.

[0115] In a feasible implementation manner, before obtaining the local environment perception information, the road network perception information, and the first vehicle cooperation communication information sent by several surrounding vehicles, it further includes:

[0116] Step S61, obtaining the vehicle current state information, the bandwidth resource limit, and the communication medium resource limit, and identifying the value information in the vehicle current state information;

[0117] It should be noted that the vehicle current state information refers to various real-time operation parameters and state data of the target vehicle, including but not limited to the position, speed, acceleration, driving direction, braking state, steering angle, vehicle type, vehicle identifier (such as license plate number), etc. of the target vehicle.

[0118] Furthermore, it should be noted that the bandwidth resource limit refers to the limit of the data transmission capacity that the wireless communication network can provide, which affects the amount of data that can be transmitted between vehicles and between vehicles and infrastructure. Additionally, the communication medium resource limit refers to the limit related to the actual physical medium (such as radio waves, infrared rays, etc.) used for vehicle communication, including frequency range, signal strength, transmission distance, etc., which affects the reliability and effectiveness of communication between vehicles, especially when multiple vehicles attempt to communicate simultaneously.

[0119] Furthermore, the value information refers to information that is of great significance to vehicle decision-making and coordination, such as potential collision risks, emergency braking signals, road condition changes, traffic signal states, etc. Value information is the part that should be preferentially retained when the vehicle compresses information to ensure that key information can be effectively transmitted and received.

[0120] Specifically, to achieve efficient communication between multiple vehicles, there are three aspects of challenges: (1) The vehicle's current state information contains redundant information, and it is difficult for the communication system to identify valuable information from the original message and retain valuable information during compression; (2) The vehicle will receive a large number of messages during communication. Therefore, the communication method must extract valuable information from all the received messages to help the vehicle make driving decisions. It is difficult for the communication system to quickly and efficiently integrate the large number of received messages to generate a valuable information vector for the current vehicle; (3) Transmitting worthless messages will waste limited communication resources, and it is difficult for the communication system to evaluate the importance of messages for intelligent vehicle coordination. Therefore, generating value information reduces the complexity of information extraction for the receiver; for bandwidth limitations, an efficient communication method needs to effectively compress the message to be sent before sending the message, so that the message to be sent meets the bandwidth limitations, and at the same time retain valid information during the compression process to ensure the quality of multi-agent coordination; for communication medium resource limitations, an efficient communication method needs to control the number of agents sending messages simultaneously, make message sending decisions based on message evaluation, save communication medium resources, and improve the effectiveness of multi-agent communication.

[0121] Therefore, obtaining the vehicle's current state information, bandwidth resource limitations, and communication medium resource limitations through local sensors is not restricted here. Further, identify the value information in the vehicle's current state information, where the state information most relevant to driving decisions is identified through a vehicle value data model. For example, the speed, direction, and expected trajectory of the target vehicle are more important and of higher value than the vehicle's color or model.

[0122] Furthermore, sort the importance of the state information, and then generate value information according to the sorting. For example, if an agent is approaching an intersection, then information about the traffic signal state is more important than information about vehicles in the distance.

[0123] Step S62, based on the bandwidth resource limitations, compress the value information to generate second vehicle collaborative communication information;

[0124] It should be noted that the second vehicle collaborative communication information refers to the information used for collaborative communication between vehicles after the value information corresponding to the target vehicle is compressed and processed. It takes into account the limitations of bandwidth and communication medium resources and only includes the information that is most valuable for collaborative decision-making with other vehicles, aiming to achieve effective information sharing between vehicles while minimizing the consumption of communication resources.

[0125] Specifically, according to the bandwidth resource limitation, the value information is compressed to generate the second vehicle collaborative communication information. Among them, data compression techniques such as Huffman coding and Lempel-Ziv coding can be applied for information compression to identify patterns in the data and represent these patterns with shorter codes, thereby reducing data redundancy.

[0126] Step S63: Based on the communication medium resource limitation, send the second vehicle collaborative communication information to a number of surrounding vehicles.

[0127] Specifically, according to the communication medium resource limitation, determine the surrounding vehicles that need to conduct collaborative communication within the surrounding range, and then send the second vehicle collaborative communication information to a number of surrounding vehicles.

[0128] In this embodiment, by obtaining the vehicle's current state information, bandwidth resource limitation, and communication medium resource limitation, and identifying the value information in the vehicle's current state information, and then based on the bandwidth resource limitation, compressing the value information to generate the second vehicle collaborative communication information, and thus based on the communication medium resource limitation, sending the second vehicle collaborative communication information to a number of surrounding vehicles. By considering the bandwidth resource limitation and communication medium resource limitation while identifying and compressing the value information, effective information compression is achieved, enabling the message to be sent to meet the bandwidth limitation. At the same time, effective information is retained during the compression process to ensure the quality of multi-agent collaboration, and the number of agents sending messages simultaneously is controlled, thereby saving communication medium resources and improving the effectiveness of multi-agent communication. Especially in the case of limited bandwidth resources, it ensures that key information can be preferentially transmitted, thereby enhancing information sharing between vehicles and achieving efficient collaborative communication between multiple vehicles, which is helpful for achieving better collaborative driving and promoting the development of intelligent transportation systems.

[0129] In a feasible implementation manner, after generating the vehicle passing decision corresponding to the target vehicle based on the unified view state information, the following steps are further included:

[0130] Step S71: Push the vehicle passing decision to the target user for the target user to confirm;

[0131] Specifically, push the vehicle traffic decision to the target user, which can be presented to the target user through a user interface, a visualization tool, voice broadcast, etc. for the target user to confirm.

[0132] Step S72, if a decision execution instruction from the target user is received, convert the vehicle traffic decision into a vehicle control instruction.

[0133] It should be noted that the decision execution instruction refers to an instruction issued by the target user (which can be a driver or a remote operator) to confirm the traffic decision generated by the vehicle decision-making system. This instruction indicates that the user agrees to execute a specific vehicle traffic decision. For example, if the vehicle decision-making system recommends turning right at the next intersection, the decision execution instruction can be for the user to confirm this recommendation and instruct the vehicle to execute.

[0134] Furthermore, it should be noted that the vehicle control instruction refers to an instruction that directly controls the vehicle's hardware system, such as the engine, braking system, steering system, etc. The vehicle control instruction is based on the vehicle traffic decision and converts the macroscopic driving strategy into specific operation actions. For example, if the vehicle traffic decision requires the vehicle to decelerate and turn right at an intersection, the corresponding vehicle control instruction may include reducing the engine power, activating the braking system to decelerate, and adjusting the steering system to execute the right turn.

[0135] Specifically, if a decision execution instruction from the target user is received, convert the vehicle traffic decision into a vehicle control instruction according to the decision execution instruction. In addition, a decision adjustment instruction issued by the target user for the vehicle traffic decision can also be received, and then the vehicle traffic decision is adjusted according to the decision adjustment instruction, and vehicle control is performed after the adjustment is successful.

[0136] Step S73, perform vehicle control on the target vehicle based on the vehicle control instruction.

[0137] Specifically, perform vehicle control on each control module in the target vehicle according to the vehicle control instruction.

[0138] In this embodiment, by pushing the vehicle traffic decision to the target user for the target user to confirm, and then if a decision execution instruction from the target user is received, convert the vehicle traffic decision into a vehicle control instruction, so as to perform vehicle control on the target vehicle based on the vehicle control instruction, thereby avoiding potential risks caused by errors in the automatic decision-making system, reducing misoperations caused by system misjudgment or sensor errors, improving the accuracy of decision-making, increasing safety, and thus ensuring that while the autonomous driving system provides convenience, it can also maintain a high level of safety and adaptability, meeting the needs and expectations of users.

[0139] It should be understood that the sequence numbers of the steps in the above embodiments do not imply the order of execution. The order of execution of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0140] This application also provides a vehicle passing decision generation device. Please refer to Figure 5 The vehicle passing decision generation device includes:

[0141] An information acquisition module, configured to acquire local environment perception information, road network perception information, and first vehicle cooperative communication information sent by a plurality of surrounding vehicles;

[0142] An information fusion module, configured to perform information fusion on the local environment perception information, the road network perception information, and each of the first vehicle cooperative communication information to obtain unified view state information;

[0143] A decision generation module, configured to generate a vehicle passing decision corresponding to a target vehicle based on the unified view state information.

[0144] The vehicle passing decision generation device is further configured to:

[0145] Perform data preprocessing on the local environment perception information, the road network perception information, and each of the first vehicle cooperative communication information;

[0146] Extract features from the preprocessed local environment perception information, road network perception information, and each of the first vehicle cooperative communication information to obtain a plurality of target feature information;

[0147] Perform information fusion on each of the target feature information to obtain the unified view state information.

[0148] The vehicle passing decision generation device is further configured to:

[0149] Determine the association relationship, time relationship, and spatial relationship between each of the target feature information;

[0150] Perform context understanding on each of the target feature information to determine the logical relationship corresponding to each of the target feature information;

[0151] Perform information fusion based on the association relationship, time relationship, spatial relationship, and logical relationship corresponding to each of the target feature information to obtain the unified view state information.

[0152] The vehicle passing decision generation device is further configured to:

[0153] Acquire vehicle cooperation granularity, vehicle destination information, and vehicle real-time driving information;

[0154] Perform route analysis on the vehicle destination information and the unified view state information to determine the optimal driving path corresponding to the target vehicle and the current traffic conditions;

[0155] Generate a vehicle passing decision corresponding to the target vehicle based on the vehicle collaboration granularity, the optimal driving path, the current traffic conditions, and the vehicle real-time driving information.

[0156] The vehicle passing decision generation device is further configured to:

[0157] Input the vehicle collaboration granularity, the optimal driving path, the current traffic conditions, and the vehicle real-time driving information into the upper network of the vehicle decision generation network to obtain a number of vehicle passing sub-goals output by the upper network;

[0158] Input each of the vehicle passing sub-goals into the lower network of the vehicle decision generation network to obtain a number of vehicle control actions corresponding to each of the vehicle passing sub-goals output by the lower network;

[0159] Generate the vehicle passing decision based on each of the vehicle passing sub-goals and the vehicle control actions corresponding to each of the vehicle passing sub-goals.

[0160] The vehicle passing decision generation device is further configured to:

[0161] Obtain the current vehicle state information, the bandwidth resource limit, and the communication medium resource limit, and identify the value information in the current vehicle state information;

[0162] Perform information compression on the value information based on the bandwidth resource limit to generate second vehicle collaboration communication information;

[0163] Send the second vehicle collaboration communication information to a number of surrounding vehicles based on the communication medium resource limit.

[0164] The vehicle passing decision generation device is further configured to:

[0165] Push the vehicle passing decision to the target user for the target user to confirm;

[0166] If a decision execution instruction from the target user is received, convert the vehicle passing decision into a vehicle control instruction;

[0167] Perform vehicle control on the target vehicle based on the vehicle control instruction.

[0168] The vehicle passing decision generation device provided by the present application adopts the vehicle passing decision generation method in the above embodiment, and can solve the technical problems in the background art. Compared with the prior art, the beneficial effects of the vehicle passing decision generation device provided by the present application are the same as those of the vehicle passing decision generation method provided by the above embodiment, and other technical features in the vehicle passing decision generation device are the same as the features disclosed in the method of the above embodiment, which will not be elaborated herein.

[0169] The present application provides a vehicle passing decision generation device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the vehicle passing decision generation method in the first embodiment above.

[0170] Reference is made below to Figure 6 , which shows a schematic structural diagram of a vehicle passing decision generation device suitable for implementing the embodiments of the present application. The vehicle passing decision generation device in the embodiments of the present application may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), in-vehicle terminals (such as in-vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 6 The vehicle passing decision generation device shown is only an example, and should not impose any limitations on the functions and usage scopes of the embodiments of the present application.

[0171] As Figure 6As shown, the vehicle passing decision generation device may include a processing device 1001 (such as a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM: Read Only Memory) 1002 or the program loaded from the storage device 1003 into the random access memory (RAM: Random Access Memory) 1004. In the RAM 1004, various programs and data required for the operation of the vehicle passing decision generation device are also stored. The processing device 1001, the ROM 1002, and the RAM 1004 are connected to each other through a bus 1005. The input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the vehicle passing decision generation device to communicate with other devices wirelessly or wiredly to exchange data. Although the figure shows a vehicle passing decision generation device with various systems, it should be understood that it is not required to implement or have all the shown systems. Instead, more or fewer systems can be implemented or had.

[0172] In particular, according to the embodiments disclosed in the present application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program codes for executing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device, or installed from the storage device 1003, or installed from the ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the methods of the embodiments disclosed in the present application are executed.

[0173] The vehicle passing decision generation device provided by the present application adopts the vehicle passing decision generation method in the above embodiments and can solve the technical problems in the background art. Compared with the prior art, the beneficial effects of the vehicle passing decision generation device provided by the present application are the same as those of the vehicle passing decision generation method provided by the above embodiments, and other technical features in the vehicle passing decision generation device are the same as those disclosed in the method of the previous embodiment, and will not be elaborated here.

[0174] It should be understood that each part disclosed in this application can be implemented by hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in a suitable manner in any one or more embodiments or examples.

[0175] As described above, only the specific embodiments of this application are provided, but the protection scope of this application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application, and all should be covered by the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

[0176] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., computer programs) stored thereon, and the computer-readable program instructions are used to execute the vehicle passing decision generation method in the above embodiments.

[0177] The computer-readable storage medium provided by this application can be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM) or flash memory, optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above. In this embodiment, the computer-readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, device, or device. The program code contained on the computer-readable storage medium can be transmitted by any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

[0178] The above computer-readable storage medium can be included in the vehicle passing decision generation device; it can also exist independently and not be assembled into the vehicle passing decision generation device.

[0179] The above computer-readable storage medium carries one or more programs. When the one or more programs are executed by the vehicle passing decision generation device, the vehicle passing decision generation device is caused to:

[0180] Obtain local environment perception information, road network perception information, and first vehicle cooperative communication information sent by a number of surrounding vehicles;

[0181] Perform information fusion on the local environment perception information, the road network perception information, and each of the first vehicle cooperative communication information to obtain unified view state information;

[0182] Generate a vehicle passing decision corresponding to the target vehicle based on the unified view state information.

[0183] Computer program code for performing the operations of the present application can be written in one or more programming languages or combinations thereof. The above programming languages include object-oriented programming languages - such as Java, Smalltalk, C++, and also include conventional procedural programming languages - such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, executed as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network - including a local area network (LAN: Local Area Network) or a wide area network (WAN: Wide Area Network) - or, it can be connected to an external computer (for example, by using an Internet service provider to connect through the Internet).

[0184] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0185] The modules involved in the embodiments described in the present application can be implemented in software or in hardware. Among them, the name of the module does not constitute a limitation to the unit itself in some cases.

[0186] The readable storage medium provided by this application is a computer-readable storage medium. The computer-readable storage medium stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned vehicle passing decision generation method, and can solve the technical problems in the background art. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by this application are the same as those of the vehicle passing decision generation method provided by the above embodiment, and will not be elaborated here.

[0187] An embodiment of this application provides a computer program product, including a computer program, and when the computer program is executed by a processor, it implements the steps of the vehicle passing decision generation method as described above.

[0188] The computer program product provided by this application can solve the technical problems in the background art. Compared with the prior art, the beneficial effects of the computer program product provided by the embodiment of this application are the same as those of the vehicle passing decision generation method provided by the above embodiment, and will not be elaborated here.

[0189] The above are only partial embodiments of this application, and do not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application by using the content of the specification and drawings of this application, or directly / indirectly applied in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A vehicle traffic decision generation method, characterized in that: include: Acquire local environment perception information, road network perception information, and first vehicle cooperative communication information sent by a plurality of surrounding vehicles; fusing the local environment perception information, the road network perception information, and the first vehicle cooperative communication information to obtain unified view state information; Based on the unified view state information, a vehicle passage decision corresponding to the target vehicle is generated.

2. The vehicle traffic decision generation method according to claim 1, characterized in that: The fusing the local environment perception information, the road network perception information and the first vehicle cooperative communication information to obtain unified view state information includes: Performing data preprocessing on the local environment perception information, the road network perception information, and the first vehicle cooperative communication information; Extracting features of the preprocessed local environment perception information, road network perception information, and cooperative communication information of each first vehicle to obtain a plurality of target feature information; The target feature information is fused to obtain the unified view state information.

3. The vehicle traffic decision generation method according to claim 2, characterized in that: The step of fusing the target feature information to obtain the unified view state information includes: Determine the association relationship, time relationship and spatial relationship between each of the target feature information; Performing contextual understanding on each of the target feature information to determine the logical relationship corresponding to each of the target feature information; The unified view state information is obtained by performing information fusion based on the association relationship, time relationship, spatial relationship and logical relationship corresponding to each of the target feature information.

4. The vehicle traffic decision generation method according to claim 1, characterized in that: The generating a vehicle passage decision corresponding to the target vehicle based on the unified view state information includes: Obtain vehicle coordination granularity, vehicle destination information, and vehicle real-time driving information; Perform route analysis on the vehicle destination information and the unified view state information to determine the optimal driving path and current traffic conditions corresponding to the target vehicle; Based on the vehicle coordination granularity, the optimal driving path, the current traffic conditions and the real-time driving information of the vehicle, a vehicle passage decision corresponding to the target vehicle is generated.

5. The vehicle traffic decision generation method according to claim 4, characterized in that: The generating a vehicle passage decision corresponding to the target vehicle based on the vehicle cooperation granularity, the optimal driving path, the current traffic condition and the real-time driving information of the vehicle includes: Inputting the vehicle coordination granularity, the optimal driving path, the current traffic conditions and the real-time driving information of the vehicle into an upper network in a vehicle decision-making network, and obtaining a plurality of vehicle passage sub-goals output by the upper network; Inputting each of the vehicle passage sub-goals into the bottom layer network in the vehicle decision generation network, and obtaining a plurality of vehicle control actions corresponding to each of the vehicle passage sub-goals output by the bottom layer network; The vehicle passage decision is generated based on each of the vehicle passage sub-goals and the vehicle control actions corresponding to each of the vehicle passage sub-goals.

6. The vehicle traffic decision generation method according to claim 1, characterized in that: Before acquiring the local environment perception information, the road network perception information, and the first vehicle cooperative communication information sent by a plurality of surrounding vehicles, the method further includes: Obtaining vehicle current status information, bandwidth resource limitations, and communication medium resource limitations, and identifying valuable information in the vehicle current status information; Based on the bandwidth resource limitation, compressing the value information to generate second vehicle cooperative communication information; Based on the communication medium resource limitation, the second vehicle cooperative communication information is sent to a plurality of surrounding vehicles.

7. The vehicle traffic decision generation method according to claim 1, characterized in that: After generating a vehicle passage decision corresponding to the target vehicle based on the unified view state information, the method further includes: Pushing the vehicle passage decision to the target user for confirmation by the target user; If a decision execution instruction of the target user is received, converting the vehicle passage decision into a vehicle control instruction; Based on the vehicle control instruction, the target vehicle is controlled.

8. A vehicle traffic decision-making device, characterized in that: include: An information acquisition module, used to acquire local environment perception information, road network perception information, and first vehicle cooperative communication information sent by a plurality of surrounding vehicles; An information fusion module, used to fuse the local environment perception information, the road network perception information and the first vehicle cooperative communication information to obtain unified view state information; The decision generation module is used to generate a vehicle passage decision corresponding to the target vehicle based on the unified view state information.

9. A vehicle traffic decision-making device, characterized in that: The vehicle passage decision generating device comprises: a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the vehicle passage decision generating method as described in any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the vehicle passage decision generation method according to any one of claims 1 to 7 are implemented.

Images