Collaborative path planning method and system for intelligent networked automobile

Abstract

The invention provides a cooperative path planning method and system for intelligent connected automobiles, and relates to the technical field of traffic optimization, and the method comprises the steps: determining multiple groups of initial planning paths of multiple intelligent connected automobiles in a preset region, recognizing a first connected automobile set located at a first intersection at a first prediction moment, and obtaining a first predicted vehicle set; and performing shunting characteristic prediction of the traditional automobile at the first intersection, performing intersection network connection automobile shunting optimization in combination with first shunting prediction information, the driving purpose deterministic label and an initial planning path corresponding to the first network connection automobile set, and returning an optimization result to the corresponding intelligent network connection automobile control terminal. The problems of path conflicts, low shunting efficiency, potential traffic hazards and congestion caused by decision conflicts between intelligent network connection automobiles and traditional vehicles are solved, and the effects of improving the traffic fluency and the overall traffic efficiency of intersections, reducing traffic congestion and conflicts and improving traffic safety and efficiency are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic optimization, and particularly to a collaborative path planning method and system for intelligent connected vehicles. Background Art

[0002] With the continuous development of intelligent connected vehicle technology, information exchange between vehicles has become an important means to improve road traffic safety and efficiency. As a key technology, V2X (Vehicle-to-Everything) communication can achieve real-time data interaction between intelligent connected vehicles and other vehicles, road infrastructure, pedestrians, networks, etc., thus making the traffic system more intelligent and automated. Important applications in V2X communication technology include vehicle-to-vehicle communication (V2V), vehicle-to-infrastructure communication (V2I), vehicle-to-pedestrian communication (V2P), and information interaction between vehicles and networks, cloud platforms (V2N, V2C). Among them, V2V communication can effectively reduce traffic accidents and improve traffic flow by sharing vehicle position information, speed, driving intention, etc.

[0003] Although the application of intelligent connected vehicles has improved traffic management and driving safety to a certain extent, current technologies mainly focus on how to optimize the driving path of a single vehicle or how to improve traffic flow through information sharing. However, for the collaborative path planning problem between traditional vehicles, i.e., non-connected vehicles and intelligent connected vehicles, existing technologies do not fully consider the behavior and reaction patterns of traditional vehicles, nor do they have a deep integration of path optimization for connected vehicles in complex traffic scenarios such as intersections. This leads to decision-making conflicts and path conflicts between intelligent connected vehicles and traditional vehicles in scenarios with high traffic flow and frequent intersections, thus affecting the overall efficiency and safety of the traffic system. Moreover, existing technologies mainly focus on shortest path planning or traffic flow optimization based on static data, lacking technical solutions for intersection diversion optimization, and failing to effectively predict the diversion behavior of traditional vehicles at intersections and incorporate it into the path planning decision-making of connected vehicles. Therefore, the reaction and path planning of traditional vehicles have not been fully coordinated with intelligent connected vehicles, resulting in the potential of traffic mobility and safety not being fully exploited.

[0004] In summary, there is often a lack of collaborative traffic management between intelligent connected vehicles and traditional vehicles in the prior art, which leads to decision-making conflicts between intelligent connected vehicles and traditional vehicles, resulting in path conflicts and low diversion efficiency, causing traffic hazards and congestion problems. Summary of the Invention

[0005] The present application provides a collaborative path planning method and system for intelligent connected vehicles, which are used to solve the technical problems in the prior art that there is a lack of collaborative traffic management between intelligent connected vehicles and traditional vehicles, resulting in decision-making conflicts between intelligent connected vehicles and traditional vehicles, leading to path conflicts and low diversion efficiency, and causing traffic hazards and congestion.

[0006] In view of the above problems, the present application provides a collaborative path planning method and system for intelligent connected vehicles.

[0007] In a first aspect, the present application provides a collaborative path planning method for intelligent connected vehicles, the method comprising: Connect to the V2X communication system, determine multiple sets of initial planned paths of multiple intelligent connected vehicles within a preset area, wherein any intelligent connected vehicle carries a corresponding driving purpose certainty label; analyze the multiple sets of initial planned paths, identify a first set of connected vehicles located at a first intersection at a first prediction moment; predict the diversion characteristics of traditional vehicles at the first intersection based on the first prediction moment, and generate first diversion prediction information; perform cross-intersection connected vehicle diversion optimization by combining the first diversion prediction information, the driving purpose certainty label, and the initial planned paths corresponding to the first set of connected vehicles, and generate a first path optimization result set corresponding to the first set of connected vehicles; return the first path optimization result set to the control terminal of the corresponding intelligent connected vehicle.

[0008] In a second aspect, the present application provides a collaborative path planning system for intelligent connected vehicles, the system comprising: An initial planned path acquisition module, configured to connect to the V2X communication system and determine multiple sets of initial planned paths of multiple intelligent connected vehicles within a preset area, wherein any intelligent connected vehicle carries a corresponding driving purpose certainty label; a path analysis module, configured to analyze the multiple sets of initial planned paths and identify a first set of connected vehicles located at a first intersection at a first prediction moment; a diversion characteristic prediction module, configured to predict the diversion characteristics of traditional vehicles at the first intersection based on the first prediction moment and generate first diversion prediction information; a diversion optimization module, configured to perform cross-intersection connected vehicle diversion optimization by combining the first diversion prediction information, the driving purpose certainty label, and the initial planned paths corresponding to the first set of connected vehicles, and generate a first path optimization result set corresponding to the first set of connected vehicles; an optimization control module, configured to return the first path optimization result set to the control terminal of the corresponding intelligent connected vehicle.

[0009] One or more technical solutions provided in the present application have at least the following technical effects or advantages: The collaborative path planning method for intelligent connected vehicles provided by this application determines multiple groups of initial planned paths of multiple intelligent connected vehicles within a preset area by connecting to a V2X communication system. Among them, any intelligent connected vehicle carries a corresponding driving purpose certainty tag; parses the multiple groups of initial planned paths, and identifies a first set of connected vehicles located at a first intersection at a first prediction moment; predicts the diversion characteristics of traditional vehicles at the first intersection based on the first prediction moment to generate first diversion prediction information; combines the first diversion prediction information, the driving purpose certainty tag, and the initial planned paths corresponding to the first set of connected vehicles to perform diversion optimization of connected vehicles at the intersection, and generates a first set of path optimization results corresponding to the first set of connected vehicles; returns the first set of path optimization results to the control terminals of the corresponding intelligent connected vehicles, solving the technical problem in the prior art of lacking collaborative traffic management between intelligent connected vehicles and traditional vehicles, resulting in decision conflicts between intelligent connected vehicles and traditional vehicles, leading to path conflicts and low diversion efficiency, causing traffic hazards and congestion, and achieving the technical effect of collaborative path planning of intelligent connected vehicles and traditional vehicles, thereby improving the traffic fluency and overall traffic efficiency at intersections, reducing traffic congestion and conflicts, and enhancing traffic safety and efficiency. Description of the Drawings

[0010] Figure 1 This application provides a schematic flowchart of the collaborative path planning method for intelligent connected vehicles.

[0011] Figure 2 This application provides a schematic structural diagram of the collaborative path planning system for intelligent connected vehicles.

[0012] Description of the reference numerals: Initial planned path acquisition module 11, path parsing module 12, diversion characteristic prediction module 13, diversion optimization module 14, optimization control module 15. Detailed Embodiments

[0013] In order to make the objectives, technical solutions and advantages of this application clearer, the following further elaborates on this application in conjunction with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not used to limit this application.

[0014] Example 1, as Figure 1 shown, this application provides a collaborative path planning method for intelligent connected vehicles, and the method includes: Connect to the V2X communication system to determine multiple groups of initial planned paths for multiple intelligent connected vehicles within a preset area. Among them, any intelligent connected vehicle carries a corresponding driving purpose certainty label. Further, the driving purpose certainty label includes a destination determination label and a destination uncertainty label. Among them, the destination determination label is the label when the vehicle has a definite driving task and task destination, and the destination uncertainty label is the label when the vehicle does not have a definite driving task and task destination.

[0015] Specifically, the V2X communication system (Vehicle-to-Everything communication system) refers to the communication system between vehicles, road infrastructure, pedestrians, networks and other traffic-related entities. The V2X technology enables vehicles to perceive the surrounding traffic environment through real-time information exchange, so as to optimize driving decisions and improve traffic safety, fluency and intelligence level. The V2X communication system can collect and share data such as the location information, speed, and driving direction of multiple vehicles in real time, so as to generate multiple feasible initial planned paths for each intelligent connected vehicle. The generation basis of these paths includes the current traffic conditions of the vehicle, driving goals, and the system's prediction of road conditions, etc. Among them, the initial planned path of each intelligent connected vehicle is not single, but forms a set of multiple feasible route alternatives through the analysis of different possible paths, and the optimal path will be selected through further optimization analysis for execution later.

[0016] To ensure the accuracy and pertinence of path planning, each intelligent connected vehicle carries a driving purpose certainty label, and the role of this label is to clarify the current driving task and destination of the vehicle, which is an important basis for path planning. According to different task requirements, the driving purpose certainty label is divided into two types, namely "destination determination label" and "destination uncertainty label". Specifically, the destination determination label is used to represent the state when the vehicle has a definite driving task and target destination. For example, a vehicle may be performing a passenger-carrying task, and its destination is a specific address, and the system will optimize the path according to this destination information. In this case, the vehicle's path planning will focus on reaching the destination quickly and directly while avoiding potential traffic congestion and delays. In contrast, the destination uncertainty label indicates that the vehicle does not have a clear destination or task, which usually appears in the autonomous driving scenario, such as when a driverless taxi is in the "empty car" state without a fixed passenger. For example, in a shared mobility platform, an autonomous vehicle may be in a state of looking for passengers. At this time, the vehicle's task and destination have not been determined, and it may be dynamically adjusted according to real-time traffic flow, surrounding demand and other information. Therefore, in path planning, vehicles with a destination uncertainty label will consider more variables, such as the behavior of other vehicles, changes in traffic signals and other factors, to flexibly adapt to the changing traffic environment.

[0017] By assigning appropriate driving purpose certainty tags to each intelligent connected vehicle, the system can design personalized path planning schemes for each vehicle, ensuring that in different traffic scenarios, both vehicles with a determined destination and vehicles in a destinationless state can obtain the optimal driving path. The core of this process lies in achieving information intercommunication between vehicles through V2X communication technology, so as to perform intelligent path adjustment and optimization in a multi-vehicle collaborative environment. Through the combination of multiple key steps such as the V2X communication system and driving purpose tags, the efficiency and accuracy of path planning are ensured, and real-time adjustment and optimization can be carried out in a dynamically changing traffic environment to improve the traffic efficiency and safety of intelligent connected vehicles.

[0018] Parse the multiple groups of initial planned paths and identify the first set of connected vehicles located at the first intersection at the first prediction moment.

[0019] Optionally, the initial planned paths are multiple alternative driving routes generated by the V2X communication system based on information such as the current position, speed, and destination of each intelligent connected vehicle. These paths take into account the current conditions of the vehicle and the surrounding traffic environment, and through the evaluation of different paths, multiple preliminary path planning schemes are generated. However, these paths are not necessarily optimal, but a set of feasible routes after preliminary screening. Next, the system needs to further analyze and optimize these paths to ensure that the paths selected by each vehicle can meet the requirements of traffic fluency and safety during actual driving.

[0020] On this basis, identify the first set of connected vehicles located at the first intersection at the first prediction moment. The purpose of this step is to provide key traffic dynamic information for subsequent path optimization. Specifically, the first prediction moment refers to the future moment calculated by the system based on real-time traffic flow prediction and vehicle position during the path planning process. For example, near intersections, connected vehicles and traditional vehicles may be affected by traffic signal control or other traffic conditions. Therefore, when determining the path, these factors must be considered. Through the V2X communication system, vehicles can share data such as their current positions and speeds. The system calculates based on this information which vehicles will reach the first intersection at the first prediction moment. The first set of connected vehicles refers to the set of all connected vehicles near the intersection at this moment, which includes all vehicles that may intersect with the intersection. The movement trajectories of these vehicles at the intersection are one of the key factors in path planning and must be given special attention. For example, assume there are three connected vehicles A, B, and C, which come from different road segments and are expected to approach the same intersection at a certain future moment. At the first prediction moment, the system collects data such as the position information, driving speed, and driving intention of these vehicles through the V2X communication system, and predicts the specific moments and positions when they reach the intersection. These vehicles are thus identified as the "first set of connected vehicles" and become the objects for subsequent diversion optimization.

[0021] By analyzing the initially planned paths of vehicles and predicting the traffic dynamics at intersections, it is possible to identify in real time which connected vehicles may reach the intersection at the same time, thereby providing the necessary information support for subsequent path optimization and diversion strategies. This process is carried out in the early stage of path planning, laying the foundation for the subsequent fine-tuning of the path.

[0022] Based on the first prediction moment, predict the diversion characteristics of traditional vehicles at the first intersection to generate the first diversion prediction information.

[0023] Exemplarily, the prediction of the diversion characteristics of traditional vehicles refers to the traffic flow analysis of traditional vehicles (i.e., non-connected vehicles) passing through the first intersection to predict their driving behaviors and make path optimization decisions based on this. First, based on the first prediction moment, the system predicts the possible traffic conditions at this moment through real-time monitoring of the intersection and its surrounding traffic environment. The first prediction moment refers to the future moment calculated by the system based on data such as vehicle position information and traffic signal status. For example, if the system predicts that there will be a large number of vehicles passing through the intersection in the next 5 minutes, then this 5 minutes is the first prediction moment. At this time, the system generates a prediction result based on historical data, traffic flow models, and real-time road information.

[0024] Next, analyze and identify the traffic flow characteristics at the first intersection, especially the distribution of traditional vehicles. Specifically, the diversion characteristics of traditional vehicles refer to the passing behaviors and route selections of these vehicles at the intersection. Generally, the driving routes of traditional vehicles are relatively fixed and are greatly affected by factors such as traffic lights and road conditions. The system infers the possible traffic flow distributions on different lanes based on factors such as historical traffic flow data, traffic signal cycles at the intersection, and traffic volumes in each direction. For example, if the traffic volume on a certain lane is large while that on another lane is small, traditional vehicles may tend to choose the lane with a smaller traffic volume. These behaviors and preferences will affect the path planning decisions of connected and autonomous vehicles, so detailed analysis is required through a diversion prediction model.

[0025] Furthermore, generating the first diversion prediction information means generating a set of data through predicting the traditional traffic flow characteristics at the first intersection, and this data reflects the traditional vehicle flow conditions on different lanes and in different directions. This prediction information usually includes the traffic flow density of each lane, the vehicle passing time, and the possible congestion situations. For example, assuming that the traffic volume on the northbound lane at the first intersection is large at the first prediction moment, the system will predict that this lane may be congested at future moments by analyzing historical data and real-time data, and generate high congestion probability information for this lane. In contrast, if the traffic volume on the eastbound lane is small, the system will predict that the traffic flow on this lane is relatively smooth and generate low congestion probability information. By predicting the diversion characteristics of these traditional vehicles, it can provide necessary decision-making basis for subsequent path optimization. These diversion prediction information will be input into the path planning system of connected and autonomous vehicles to help the vehicles determine which lanes are currently relatively unobstructed and which lanes may be congested, so as to make reasonable path selections and adjustments.

[0026] Through predicting and analyzing the traffic flow, potential traffic congestion areas are identified in advance, and this information is integrated into the path planning of connected and autonomous vehicles to ensure that the vehicles can avoid congestion areas, improve traffic efficiency, and at the same time enhance the overall safety and smoothness of traffic.

[0027] Execute the diversion optimization of connected vehicles at the intersection by combining the first diversion prediction information, the driving purpose certainty label, and the initial planned paths corresponding to the first set of connected vehicles, and generate the first set of path optimization results corresponding to the first set of connected vehicles. Return the first set of path optimization results to the control terminals of the corresponding connected and autonomous vehicles.

[0028] Specifically, when performing the optimization of the diversion of connected vehicles at intersections, the system needs to comprehensively consider the diversion prediction information and the driving purpose tags, analyze the initial path and target state of the vehicle, so as to generate an optimized driving path. The process of performing the optimization of the diversion of connected vehicles at intersections includes the following key steps. First, path selection and adjustment. According to the first diversion prediction information, analyze the traffic flow and traffic conditions of each lane, and select the most suitable lane for each intelligent connected vehicle. For example, in a lane with a large traffic flow, the system may recommend that the vehicle choose another lane to avoid congestion, while on a smooth lane, the vehicle can pass smoothly. Second, combine the destination with the smoothness. For vehicles with a determined destination, the system guides them to the shortest path or the lane with the highest traffic flow smoothness according to their destination tags. For vehicles with an undetermined destination, the system will adjust their driving paths according to the real-time traffic flow and prediction results to ensure that they can optimize the path selection during driving and avoid congestion. Third, dynamic path update. At each intersection, the system will dynamically adjust the path planning, based on the current traffic flow conditions and the real-time position of the vehicle, to optimize the path selection and ensure the smooth passage of the vehicle. For example, if an unexpected congestion occurs in a certain lane, the system will automatically re-plan the path, adjust the driving direction or choice of the vehicle. Through these steps, a first set of path optimization results corresponding to the first set of connected vehicles is generated, that is, an optimized driving path will be generated for each intelligent connected vehicle. This path fully considers the traffic flow conditions at intersections, the driving intentions of vehicles, and the behaviors of other traffic participants, aiming to maximize traffic efficiency and reduce congestion. Finally, returning the first set of path optimization results to the control terminal of the corresponding intelligent connected vehicle is a key link in the entire path optimization process. The system transmits the optimized path results to the control terminal of each intelligent connected vehicle, and the vehicle makes real-time adjustments according to this information, so as to successfully execute the new path plan and drive according to the optimized route. In this way, intelligent connected vehicles can cooperate with other vehicles and traffic facilities in real time to ensure that the most suitable driving path is always selected in a complex traffic environment, thereby improving the smoothness and safety of the overall traffic.

[0029] Furthermore, by combining the first diversion prediction information, the driving purpose certainty tag, and the initial planned path corresponding to the first set of connected vehicles to perform the optimization of the diversion of connected vehicles at intersections, and generating the first set of path optimization results corresponding to the first set of connected vehicles, it includes: Call the diversion fluency prediction model, and analyze the traditional vehicle flow smoothness indicators in each direction of the first intersection according to the first diversion prediction information, where the diversion fluency prediction model is constructed by collecting historical diversion information and the corresponding vehicle communication fluency samples in each direction; combine the driving purpose certainty label and the initial planned path corresponding to the first set of connected vehicles to analyze the lane-changing cost at the first intersection, and generate the first set of lane-changing costs; combine the traditional vehicle flow smoothness indicators in each direction and the first set of lane-changing costs to perform diversion optimization, and generate the first set of path optimization results.

[0030] In a specific embodiment, calling the diversion fluency prediction model and analyzing the traditional vehicle flow smoothness indicators in each direction of the first intersection according to the first diversion prediction information aims to optimize the path planning of intelligent connected vehicles by predicting the traffic flow and driving state of traditional vehicles. This process involves dynamically analyzing the traffic flow situation at the intersection and combining historical data and real-time information to predict the traffic fluency in each direction. The diversion fluency prediction model is constructed by training the historical diversion information at the intersection and the vehicle passing fluency samples in each direction. The core of this model is to predict the future traffic flow state through historical traffic data (such as past traffic volume, fluency, traffic signal cycle, etc.) and current traffic information (such as real-time traffic density, traffic signal state, etc.). Historical data is used to provide a basis for the model to help the model learn the laws of traffic flow, while real-time fluency samples provide immediate analysis basis based on the current traffic conditions. Using these data, the diversion fluency prediction model can calculate the traditional vehicle flow smoothness indicators for each lane or direction, that is, the smoothness of each lane at a specific moment. This indicator reflects the difficulty of vehicles passing through this lane and is usually measured by parameters such as traffic density, passing speed, and congestion level. For example, if the traffic density in a certain direction is high and the passing speed is slow, then the fluency indicator in this direction may be low, and vice versa. By calculating these fluency indicators, the system can identify which lanes in which directions are relatively unobstructed and which may be congested, providing data support for subsequent path optimization.

[0031] Next, combine the driving purpose certainty label with the initial planned path corresponding to the first set of connected vehicles to perform a lane-changing cost analysis at the first intersection. The purpose of this step is to calculate the cost of a connected vehicle changing lanes (changing lines) at the first intersection. The calculation of the lane-changing cost is based on the driving purpose label of the vehicle and its initial planned path. The driving purpose certainty label can be a "destination certainty label" or a "destination uncertainty label". The former means that the vehicle has a clear destination, and the latter means that the vehicle does not have a fixed destination in the short term. For example, if a connected vehicle has a clear destination and its initial planned path points to a potentially congested lane, the system will consider the congestion situation of that lane and the lane-changing cost of the vehicle to optimize its path selection. The lane-changing cost usually consists of the following factors: path complexity, increase in travel time, potential congestion risk, etc. In this process, the calculation of the lane-changing cost also needs to consider generating a first set of lane-changing costs, which includes the cost that each vehicle needs to pay when changing lanes. Specifically, it may include an increase in path complexity (such as switching from the left lane to the right lane may require more time or more complex operations), an increase in the total route distance (such as the route after lane-changing is farther than the initial route), and an increase in historical congestion data (such as a certain direction of the lane has often been congested historically). These cost values will help the system determine whether certain lane-changes are worth performing and avoid unnecessary lane-changing operations.

[0032] On this basis, combine the traditional vehicle traffic smoothness indicators in each direction with the first set of lane-changing costs to perform diversion optimization. Specifically, the system will comprehensively consider the traditional vehicle traffic smoothness indicators in different directions and the set of lane-changing costs to perform diversion optimization. The purpose of diversion optimization is to select the most suitable path for each connected vehicle according to traffic flow and lane-changing cost. For example, if the traffic flow in a certain direction is high and the lane-changing cost is low, the system may recommend that the vehicle change to another smoother lane; conversely, if the traffic flow in a certain direction is low and the lane-changing cost is high, the system may choose to maintain the original path. In this way, the system can generate a first set of path optimization results, which includes the optimal path selection for each connected vehicle when passing through the first intersection. The goal of path optimization is to minimize traffic congestion, improve the traffic efficiency of vehicles, and at the same time reduce the cost of lane-changing.

[0033] Finally, the first set of path optimization results can be returned to the control terminal of the corresponding intelligent connected vehicle, and this step ensures the execution of path optimization. The system transmits the optimized path information to the control system of each connected vehicle, and the vehicle adjusts its driving route according to these optimization results, so as to ensure that the vehicle can pass smoothly at the intersection and effectively avoid congestion. Through this path optimization mechanism, intelligent connected vehicles can not only improve traffic smoothness but also cooperate with other vehicles to achieve a more efficient overall traffic operation.

[0034] Furthermore, the lane-changing cost analysis of the first intersection is performed by combining the driving purpose certainty label with the initial planned path corresponding to the first set of connected vehicles, and a first set of lane-changing costs is generated, including: Analyze the initial planned paths of the connected vehicles in the first set of connected vehicles, perform path replanning for intersection commutation, calculate the route complexity growth index, the total route distance growth index, and the route historical congestion probability growth index of the replanned path compared to the initial planned path, and perform weighting to generate each first initial lane-changing cost; according to the driving purpose certainty labels of the connected vehicles in the first set of connected vehicles, perform a weight reduction assignment for the lane-changing cost to optimize each first initial lane-changing cost with the weight reduction assignment result, generate each first lane-changing cost, and add it to the first set of lane-changing costs.

[0035] Furthermore, analyze the initial planned paths of the connected vehicles in the first set of connected vehicles and perform path replanning for intersection commutation. First, deeply analyze the initial path of each connected vehicle to determine whether a commutation operation needs to be performed at the intersection. The goal of path replanning is to dynamically adjust the driving route of each vehicle according to the specific traffic conditions at the intersection and the driving needs of the vehicle to optimize traffic flow and avoid congestion as much as possible.

[0036] After analyzing the initial path, the system calculates the route complexity growth index of the replanned path compared to the initial planned path. This index is used to measure the increase in route complexity in the replanned path. Specifically, if the path needs to cross more intersections or involves more lane changes, the complexity growth will be higher; otherwise, it will be lower. This index reflects the operation complexity of the path replanning and the control resources that the vehicle needs to invest. For example, if a vehicle turns from a straight lane to a left-turn lane, its operation complexity is higher than staying in the original lane, so the complexity growth index of this path will increase. In addition, the system also calculates the total route distance growth index, which is used to measure the change in the total driving distance in the replanned path. If the replanned path is longer than the initial path, this index will increase. This usually occurs when choosing a detour to avoid congestion. The calculation of the total route distance growth index helps the system evaluate the efficiency of the replanned path and ensure that the optimized path will not cause unnecessary time waste due to too long a detour. At the same time, the route historical congestion probability growth index is also calculated to evaluate the increase in the probability of congestion occurring in the replanned path within a certain period in the past. The purpose of this index is to avoid choosing those paths that are often congested in history. Specifically, the system will predict the congestion risk of the current path based on historical traffic data, such as traffic flow and the frequency of congestion occurrence in the past period. If a certain path is often congested in the historical data, the replanned path may avoid choosing this path, thus reducing the probability of traffic congestion.

[0037] Based on the above steps, weighting is performed to generate each first initial lane-changing cost. The purpose of the weighting step is to comprehensively consider factors such as route complexity, total route distance growth, and historical congestion probability growth, and allocate weights according to the relative importance of these factors. For example, if avoiding congestion is more important than shortening the distance, the system may assign a higher weight to the congestion probability growth index. Through weighting, the finally generated first initial lane-changing cost can comprehensively reflect the optimization effect of each path, helping the decision-making system evaluate which path is most suitable for the current traffic situation.

[0038] Next, based on the driving purpose certainty tags of each connected vehicle in the first set of connected vehicles, perform a reduction weight assignment for the lane change cost. The driving purpose certainty tag reflects whether the driving purpose of the vehicle is clear. If the destination of the vehicle is clear and unchangeable, the system will consider that the vehicle has a lower need for route adjustment, so the lane change cost weight of the vehicle is reduced. In this way, the optimization process is more inclined to select a more flexible route plan for vehicles with unclear goals in order to better cope with traffic changes. After the reduction weight assignment, the system will optimize the respective first initial lane change costs. Specifically, the result of the reduction weight assignment will reduce the influence of certain cost factors on route selection, and then make the system pay more attention to the driving purpose of the vehicle and the current road conditions during the optimization process. For example, for vehicles with a clear destination, route optimization pays more attention to the shortest time and lowest risk routes; for vehicles without a fixed destination, route optimization can more flexibly select routes with greater changes.

[0039] Finally, generate each first lane change cost and add it to the first lane change cost set. In this step, the system generates a final lane change cost for each vehicle based on the lane change cost optimized by the above weighting and weight assignment. The lane change cost of each vehicle represents the cost required for the vehicle to perform a lane change operation under the current road conditions and purpose. The lane change costs of all vehicles will be collected and form a set, which serves as the basis for route optimization to help the system determine which vehicles need to change lanes and which vehicles should continue to maintain their original routes.

[0040] Through the above steps, it is possible to optimize the routes of connected vehicles at intersections, ensure that each vehicle can pass through the intersection in an optimal manner in a complex traffic environment, improve the overall traffic flow, and avoid traffic congestion.

[0041] Furthermore, based on the driving purpose certainty tags of each connected vehicle in the first set of connected vehicles, perform a reduction weight assignment for the lane change cost to optimize the respective first initial lane change costs with the result of the reduction weight assignment, and generate each first lane change cost, including: If the driving purpose certainty tag of any connected vehicle is a destination determined tag, set the corresponding reduction weight assignment to 0; if the driving purpose certainty tag of any connected vehicle is a destination undetermined tag, with the goal of minimizing the lane change cost among all connected vehicles, determine the corresponding reduction weight assignment and adjust the first lane change cost to generate the respective first lane change costs.

[0042] Exemplarily, if the driving purpose certainty label of any connected vehicle is a destination certainty label, the corresponding reduced weight assignment is set to 0. The purpose of this step is to ensure that for those connected vehicles with clear driving tasks and determined destinations, the path optimization process will not be overly affected by the lane-changing cost. Specifically, the destination certainty label means that the driving task of the connected vehicle has been determined and the destination is fixed. For example, a fixed destination has been reserved through the travel platform system, and the driving path of the vehicle does not require much flexibility adjustment. Therefore, the system assigns a reduced weight of zero to the lane-changing cost of these vehicles, that is, the originally calculated cost remains unchanged, and it is defaulted that the path optimization of these vehicles only focuses on the shortest path or the most direct route, without considering the complexity or potential cost of changing directions at intersections.

[0043] If the driving purpose certainty label of any connected vehicle is a destination uncertainty label, with the goal of minimizing the lane-changing cost among all connected vehicles, the corresponding reduced weight assignment is determined. At this time, the driving purpose of the vehicle is not clear and it may be in a flexible state. For example, driverless vehicles in a shared mobility platform, where the destination of the vehicle has not been determined or is diverse. For these vehicles, minimizing the lane-changing cost becomes an important goal for optimizing the path. This is because vehicles with destination uncertainty labels are more likely to make multiple path adjustments on the road to adapt to the changing traffic environment and goals. Therefore, the system will adjust the lane-changing cost of these vehicles based on the current road conditions and other factors, and promote more flexible path selection by adjusting the reduced weight assignment, reducing the complexity of path adjustments.

[0044] Specifically, the system will adjust according to the lane-changing cost of each connected vehicle to generate the respective first lane-changing costs. For vehicles with uncertain destinations, the system will dynamically adjust their lane-changing costs based on their location and the real-time conditions of the surrounding traffic. The adjustment process involves comparing the complexity of different paths, and paths with lower lane-changing costs will be preferentially selected. For example, if there is a high traffic flow on a certain path and the vehicle is in a congested area, the system will assign a higher lane-changing cost to this path, and lower costs to those paths that are smoother or in better condition. Therefore, the system will adjust the lane-changing cost in a timely manner during the vehicle's driving process to ensure that the goal is to minimize the lane-changing cost among all connected vehicles.

[0045] In this way, the respective first lane-changing costs are generated. The adjustment and optimization of these lane-changing costs ensure that vehicles with uncertain destinations can flexibly select the optimal path in a dynamic traffic environment. The lane-changing costs take into account not only the length of the road and traffic flow, but also factors such as the complexity of the road section and the status of traffic lights. Finally, by adjusting the lane-changing costs of each connected vehicle, the system can generate a more accurate set of first lane-changing costs that meet the actual traffic demand, enabling each vehicle to select the most suitable driving path at intersections and other complex road conditions.

[0046] Furthermore, the first path optimization result set is generated by combining the smoothness indicators of traditional vehicle flows in each direction with the first lane-changing cost set, including: Based on the first lane-changing cost set, a set of target vehicles with lane-changing costs less than a preset cost threshold is extracted, and the corresponding lane-changing directions are determined; with the smoothness indicators of traditional vehicle flows in each direction, as well as the set of target vehicles and the corresponding lane-changing directions, the balanced optimization of passing vehicles in each direction is performed to generate the diversion optimization results in each direction; the initial planned paths of each connected vehicle in the first connected vehicle set are updated with the diversion optimization results in each direction to generate the first path optimization result set.

[0047] Optionally, the core purpose of the step of generating the first path optimization result set by combining the smoothness indicators of traditional vehicle flows in each direction with the first lane-changing cost set is to optimize the driving paths of intelligent connected vehicles at intersections by combining traffic flow (smoothness indicators of traditional vehicle flows) with lane-changing cost information, so as to achieve a more efficient traffic flow distribution. By combining these two types of information, the system can make more accurate path optimization decisions. Specifically, based on the first lane-changing cost set, a set of target vehicles with lane-changing costs less than a preset cost threshold is extracted, and the corresponding lane-changing directions are determined. The purpose of this step is to screen out vehicles with relatively low lane-changing costs from all the connected vehicles to be optimized. These vehicles are more likely to adapt to the lane-changing adjustment and have less impact on the traffic flow. The preset cost threshold is a standard set according to traffic conditions, and usually factors such as traffic congestion degree and road section complexity are considered during system design. When the lane-changing cost of a connected vehicle is lower than this threshold, the system considers these vehicles as a set of target vehicles that can perform path optimization and further determines their lane-changing directions. This direction determines where these vehicles will change lanes at intersections, thus achieving diversion.

[0048] Execute the balanced optimization of passing vehicles in each direction based on the traditional vehicle passing indicators in each of these directions, as well as the target vehicle set and the corresponding lane-changing directions, to generate the diversion optimization results in each direction. In this step, the system not only considers the passing conditions of traditional vehicles in each direction but also combines the lane-changing directions of target vehicles to further optimize the distribution of traffic flow. The goal of the balanced optimization of passing vehicles is to ensure a reasonable distribution of traffic flow in each direction, avoiding excessive congestion in some directions while other directions are overly idle. This optimization process comprehensively evaluates the traffic flow in each direction and the lane-changing strategies of target vehicles to generate an optimal diversion plan suitable for the current road conditions. Furthermore, update the initial planned paths of each connected vehicle in the first connected vehicle set with the diversion optimization results in each direction to generate the first set of path optimization results. At this time, based on the diversion optimization results calculated previously in each direction, the system adjusts the initial planned paths of each vehicle in the first connected vehicle set. The optimized paths may include suggestions for lane-changing, shortening of path lengths, or adjustment of traffic signals to ensure smoother and more efficient driving of vehicles at intersections and their surroundings. The path of each connected vehicle is finely optimized to achieve the optimal traffic flow effect, and finally, the first set of path optimization results is generated, which contains all the optimized paths.

[0049] Through the above steps, it is possible to effectively optimize the diversion problem of intelligent connected vehicles at intersections, ensure smoother passing of vehicles at intersections, reduce traffic congestion, and improve the overall road utilization efficiency.

[0050] Furthermore, based on the first prediction moment, predict the diversion characteristics of traditional vehicles at the first intersection to generate the first diversion prediction information, including: Collect the historical diversion data of traditional vehicles at the first intersection, where the historical diversion data of traditional vehicles carries historical moment tags, and the historical diversion data of traditional vehicles includes historical data with periodic coherence; use the historical moment tags and the historical diversion data of traditional vehicles as sample data to train and learn the relationship between diversion characteristics and moments to generate a time-series diversion prediction model; use the time-series diversion prediction model to perform diversion prediction for the first prediction moment to generate the first diversion prediction information.

[0051] Specifically, historical traditional vehicle diversion data of the first intersection is collected. Among them, the historical traditional vehicle diversion data carries historical time tags, and the historical traditional vehicle diversion data includes historical data with periodic coherence. This step aims to obtain the historical traffic flow data of the first intersection to provide training samples for subsequent diversion prediction models. The historical traditional vehicle diversion data refers to the diversion situation and traffic conditions data of traditional vehicles at specific times, specific dates, and specific road sections. For example, the data may include traffic flow, traffic density and other indicators during specific periods (such as weekdays, holidays, etc.) or per hour. At the same time, these data also need to carry historical time tags to ensure that each piece of data can correspond to a specific time and date for time series processing during analysis.

[0052] Among them, the historical traditional vehicle diversion data includes historical data with periodic coherence. This means that the collected historical data should have continuity and periodicity to ensure that the prediction model can identify and learn the regular changes in traffic flow. For example, when collecting data, traffic data for each day and each moment within two years can be selected to ensure sufficient time intervals and regularity between the data. Such periodic data can help the model learn the flow change trends during different time periods (such as morning rush hour, evening rush hour), thereby improving the accuracy of prediction results.

[0053] Using the historical time tags and the historical traditional vehicle diversion data as sample data, the relationship between diversion features and time is trained and learned to generate a time-series diversion prediction model. In this step, through the analysis of historical data, the system uses machine learning methods to train the model to learn the time-series features of traffic flow. Specifically, the historical time tags represent the time information of each piece of data. By combining with the corresponding traditional vehicle diversion data, the model can learn the diversion rules at different time nodes (such as different hours, different dates, or different seasons). This time-series diversion prediction model can capture the rules of traffic flow changing over time and provide an accurate basis for future flow prediction.

[0054] Using the time-series diversion prediction model to conduct diversion prediction for the first prediction moment to generate the first diversion prediction information. At this time, the trained time-series diversion prediction model is applied to future traffic flow prediction. Specifically, the model predicts the diversion situation of traditional vehicles at the first prediction moment based on the current historical data and time-series features. This prediction result is the so-called first diversion prediction information, which contains the predicted data of vehicle flows in all directions at the intersection at the first prediction moment. This prediction information is crucial for subsequent route optimization and traffic scheduling because it can provide the future traffic flow change trend for the system and help intelligent connected vehicles make more reasonable route adjustments.

[0055] In summary, by collecting historical traditional vehicle diversion data with periodic coherence and using a time-series diversion prediction model, the traffic flow at future moments can be accurately predicted. This method effectively improves the prediction accuracy of traffic flow and provides data support for the path optimization of intelligent connected vehicles and the diversion optimization of intersections.

[0056] Through the technical solutions of the above embodiments, the collaborative path planning method for intelligent connected vehicles provided by this application solves the technical problems in the prior art that there is a lack of collaborative traffic flow management between intelligent connected vehicles and traditional vehicles, resulting in decision-making conflicts between intelligent connected vehicles and traditional vehicles, leading to path conflicts and low diversion efficiency, causing traffic hazards and congestion. It achieves the collaborative path planning of intelligent connected vehicles and traditional vehicles, thereby improving the traffic fluency at intersections and the overall traffic efficiency, reducing traffic congestion and conflicts, and enhancing traffic safety and efficiency.

[0057] Embodiment 2, based on the same inventive concept as the collaborative path planning method for intelligent connected vehicles in the foregoing embodiment, as Figure 2 shown, this application provides a collaborative path planning system for intelligent connected vehicles, and the system includes: An initial planned path acquisition module 11, configured to connect to a V2X communication system and determine multiple groups of initial planned paths for multiple intelligent connected vehicles within a preset area, where any intelligent connected vehicle carries a corresponding driving purpose certainty label.

[0058] A path analysis module 12, configured to analyze the multiple groups of initial planned paths and identify a first set of connected vehicles located at a first intersection at a first prediction moment.

[0059] A diversion feature prediction module 13, configured to predict the diversion features of traditional vehicles at the first intersection based on the first prediction moment, and generate first diversion prediction information.

[0060] A diversion optimization module 14, configured to perform intersection connected vehicle diversion optimization by combining the first diversion prediction information, the driving purpose certainty label, and the initial planned paths corresponding to the first set of connected vehicles, and generate a first path optimization result set corresponding to the first set of connected vehicles.

[0061] An optimization control module 15, configured to return the first path optimization result set to the control terminals of the corresponding intelligent connected vehicles.

[0062] Furthermore, the diversion optimization module 14 is further configured to perform the following steps: Call the diversion fluency prediction model, and analyze the traditional vehicle fluency indicators in each direction of the first intersection according to the first diversion prediction information, where the diversion fluency prediction model is trained and constructed by collecting historical diversion information and the corresponding vehicle communication fluency samples in each direction; combine the driving purpose certainty label and the initial planned path corresponding to the first set of connected vehicles to analyze the lane-changing cost at the first intersection, and generate the first set of lane-changing costs; combine the traditional vehicle fluency indicators in each direction and the first set of lane-changing costs to perform diversion optimization, and generate the first set of path optimization results.

[0063] Furthermore, the diversion optimization module 14 is further configured to perform the following steps: Analyze the initial planned paths of each connected vehicle in the first set of connected vehicles, perform path replanning for lane-changing at the intersection, calculate the route complexity growth index, the total route distance growth index, and the route historical congestion probability growth index of the replanned path compared to the initial planned path, and perform weighting to generate each first initial lane-changing cost; according to the driving purpose certainty label of each connected vehicle in the first set of connected vehicles, perform a reduction weight assignment for the lane-changing cost to optimize each first initial lane-changing cost with the reduction weight assignment result, generate each first lane-changing cost, and add it to the first set of lane-changing costs.

[0064] Furthermore, the diversion optimization module 14 is further configured to perform the following steps: If the driving purpose certainty label of any connected vehicle is the destination certainty label, set the corresponding reduction weight assignment to 0; if the driving purpose certainty label of any connected vehicle is the destination uncertainty label, with the goal of minimizing the lane-changing cost among all connected vehicles, determine the corresponding reduction weight assignment, and adjust the first lane-changing cost to generate each first lane-changing cost.

[0065] Furthermore, the diversion optimization module 14 is further configured to perform the following steps: Based on the first set of lane-changing costs, extract the target vehicle set with the lane-changing cost less than the preset cost threshold, and determine the corresponding lane-changing direction; with the traditional vehicle fluency indicators in each direction, as well as the target vehicle set and the corresponding lane-changing direction, perform the balanced optimization of the passing vehicles in each direction, generate the diversion optimization results in each direction; update the initial planned paths of each connected vehicle in the first set of connected vehicles with the diversion optimization results in each direction to generate the first set of path optimization results.

[0066] Furthermore, the initial planned path acquisition module 11 further includes: The driving purpose certainty label includes a destination certainty label and a destination uncertainty label. Among them, the destination certainty label is the label when the vehicle has a definite driving task and a task destination, and the destination uncertainty label is the label when the vehicle does not have a definite driving task and a task destination.

[0067] Furthermore, the shunt feature prediction module 13 is further configured to perform the following steps: Collect the historical shunt data of traditional vehicles at the first intersection. Among them, the historical shunt data of traditional vehicles carries a historical moment label, and the historical shunt data of traditional vehicles includes historical data with periodic coherence; use the historical moment label and the historical shunt data of traditional vehicles as sample data to train and learn the relationship between shunt features and moments, and generate a temporal shunt prediction model; use the temporal shunt prediction model to perform shunt prediction on the first prediction moment to generate the first shunt prediction information.

[0068] Through the foregoing detailed description of the collaborative path planning method for intelligent connected vehicles in this specification, those skilled in the art can clearly know the collaborative path planning system of intelligent connected vehicles in this embodiment. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple. For related parts, refer to the description in the method part.

[0069] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A collaborative path planning method for intelligent connected vehicles, characterized in that, Including: Connect to the V2X communication system to determine multiple groups of initial planned paths of multiple intelligent connected vehicles within a preset area, where any intelligent connected vehicle carries a corresponding driving purpose certainty label; Analyze the multiple groups of initial planned paths to identify the first set of connected vehicles located at the first intersection at the first prediction moment; Based on the first prediction moment, predict the diversion characteristics of traditional vehicles at the first intersection to generate first diversion prediction information; Combine the first diversion prediction information, the driving purpose certainty label, and the initial planned paths corresponding to the first set of connected vehicles to perform diversion optimization of connected vehicles at the intersection, and generate the first path optimization result set corresponding to the first set of connected vehicles; Return the first path optimization result set to the control terminals of the corresponding intelligent connected vehicles.

2. The collaborative path planning method for intelligent networked vehicles according to claim 1, wherein Combining the first diversion prediction information, the driving purpose certainty label, and the initial planned paths corresponding to the first set of connected vehicles to perform diversion optimization of connected vehicles at the intersection, and generating the first path optimization result set corresponding to the first set of connected vehicles, including: Call the diversion fluency prediction model, and analyze the smoothness indexes of traditional vehicle flows in each direction of the first intersection according to the first diversion prediction information, where the diversion fluency prediction model is trained and constructed by collecting historical diversion information and vehicle communication fluency samples in the corresponding directions; Combine the driving purpose certainty label and the initial planned paths corresponding to the first set of connected vehicles to perform lane-changing cost analysis at the first intersection to generate the first lane-changing cost set; Combine the smoothness indexes of traditional vehicle flows in each direction and the first lane-changing cost set to perform diversion optimization to generate the first path optimization result set.

3. The collaborative path planning method for intelligent connected vehicles according to claim 2, wherein, Combining the driving purpose certainty label and the initial planned paths corresponding to the first set of connected vehicles to perform lane-changing cost analysis at the first intersection to generate the first lane-changing cost set, including: Analyze the initial planned paths of each connected vehicle in the first set of connected vehicles, perform path replanning for intersection commutation, calculate the route complexity growth index, the total route distance growth index, and the route historical congestion probability growth index of the replanned path compared with the initial planned path, and perform weighting to generate each first initial lane-changing cost; According to the driving purpose certainty labels of each connected vehicle in the first set of connected vehicles, perform weight reduction assignment for lane-changing costs, and optimize each first initial lane-changing cost with the weight reduction assignment result to generate each first lane-changing cost, and add it to the first lane-changing cost set.

4. The collaborative path planning method for intelligent connected vehicles according to claim 3, wherein According to the driving purpose certainty labels of each connected vehicle in the first set of connected vehicles, perform weight reduction assignment for lane-changing costs, and optimize each first initial lane-changing cost with the weight reduction assignment result to generate each first lane-changing cost, including: If the driving purpose certainty label of any connected vehicle is the destination certainty label, assign the corresponding weight reduction value to 0; If the driving purpose certainty label of any connected vehicle is the destination uncertainty label, with the goal of minimizing the lane-changing cost in each connected vehicle, determine the corresponding reduced weight assignment, adjust the first lane-changing cost, and generate the first lane-changing costs for each vehicle.

5. The collaborative path planning method for intelligent connected vehicles according to claim 2, characterized in that, Combine the traditional vehicle traffic flow smoothness indicators in each direction with the first lane-changing cost set for shunt optimization to generate the first path optimization result set, including: Based on the first lane-changing cost set, extract the set of target vehicles with a lane-changing cost less than the preset cost threshold and determine the corresponding lane-changing directions. Using the traditional vehicle traffic flow smoothness indicators in each direction, as well as the set of target vehicles and the corresponding lane-changing directions, perform balanced optimization of the passing vehicles in each direction to generate the shunt optimization results in each direction. Update the initial planned paths of each connected vehicle in the first connected vehicle set with the shunt optimization results in each direction to generate the first path optimization result set.

6. The collaborative path planning method for intelligent connected vehicles according to claim 1, characterized in that, The driving purpose certainty label includes a destination certainty label and a destination uncertainty label. Among them, the destination certainty label is the label when the vehicle has a definite driving task and task destination, and the destination uncertainty label is the label when the vehicle does not have a definite driving task and task destination.

7. The collaborative path planning method for intelligent connected vehicles according to claim 1, characterized in that, Based on the first prediction moment, predict the shunt characteristics of traditional vehicles at the first intersection to generate the first shunt prediction information, including: Collect the historical traditional vehicle shunt data at the first intersection, where the historical traditional vehicle shunt data carries a historical moment label, and the historical traditional vehicle shunt data includes historical data with periodic coherence. Use the historical moment label and the historical traditional vehicle shunt data as sample data to train the relationship between shunt characteristics and moments, and generate a time-series shunt prediction model. Use the time-series shunt prediction model to predict the shunt at the first prediction moment to generate the first shunt prediction information.

8. The collaborative path planning system for intelligent connected vehicles, characterized in that, For implementing the collaborative path planning method of the intelligent connected vehicle according to any one of claims 1-7, the system includes An initial planned path acquisition module for connecting to the V2X communication system to determine multiple groups of initial planned paths of multiple intelligent connected vehicles in a preset area, where any intelligent connected vehicle carries a corresponding driving purpose certainty label. A path parsing module for parsing the multiple groups of initial planned paths to identify the first connected vehicle set located at the first intersection at the first prediction moment. A shunt characteristic prediction module for predicting the shunt characteristics of traditional vehicles at the first intersection based on the first prediction moment to generate the first shunt prediction information. A shunt optimization module for performing shunt optimization of connected vehicles at the intersection by combining the first shunt prediction information, the driving purpose certainty label, and the initial planned paths corresponding to the first connected vehicle set, and generating the first path optimization result set corresponding to the first connected vehicle set. An optimization control module for returning the first path optimization result set to the control terminal of the corresponding intelligent connected vehicle.

Images