Multi-vehicle cooperative lane changing control method, device and equipment and medium

Abstract

The invention provides a multi-vehicle cooperative lane changing control method, device and equipment and a medium, and relates to the technical field of vehicles, and the method comprises the steps: planning a plurality of lane changing actions for each cooperative vehicle in a plurality of cooperative vehicles in a future time period, and planning a plurality of lane changing actions for each cooperative vehicle in the plurality of cooperative vehicles in a future time period; a plurality of lane changing actions of each cooperative vehicle in the future time period are evaluated, a reward function value corresponding to each cooperative vehicle is obtained, and the reward function value corresponding to each cooperative vehicle is used for representing the influence of the cooperative vehicle driving in the future time period according to the planned lane changing actions on the traffic flow in the future time period; all combinations of the multiple lane changing actions of the multiple cooperative vehicles in the future time period are traversed, and multiple multi-vehicle cooperative lane changing schemes are generated; and based on the reward function values corresponding to the multiple cooperative vehicles, determining a total reward function value of each multi-vehicle cooperative lane changing scheme, and sending a control instruction to the multiple cooperative vehicles according to the multi-vehicle cooperative lane changing scheme with the maximum total reward function value.

Description

Technical Field

[0001] The present application relates to the field of vehicle technology, and in particular to a multi-vehicle cooperative lane change control method, device, equipment and medium. Background Art

[0002] With the rapid development of autonomous driving technology, autonomous vehicles are increasingly being used in road traffic. However, in real-world road environments, autonomous vehicles must operate alongside other vehicles (both autonomous and traditionally driven), creating complex mixed traffic flows. In this complex traffic environment, multi-vehicle coordinated lane changing has become a crucial technical tool for improving traffic efficiency and safety.

[0003] However, when planning lane changes, real-time assessment of the dynamic traffic environment is often lacking, making it difficult to effectively respond to changes in traffic flow. Furthermore, when multiple vehicles are coordinating lane changes, the decision-making process is relatively crude, lacking detailed consideration of the interactions between vehicles. This can lead to traffic congestion or collision risks, and it is difficult to balance safety and efficiency simultaneously, resulting in an unstable lane change process and affecting the efficiency of overall traffic flow. Summary of the Invention

[0004] In view of this, embodiments of the present application provide a multi-vehicle cooperative lane change control method, device, equipment and medium to overcome the above-mentioned problems or at least partially solve the above-mentioned problems.

[0005] A first aspect of an embodiment of the present application provides a multi-vehicle coordinated lane change control method, the method comprising: For each of the plurality of cooperating vehicles, planning a plurality of lane changing actions for each cooperating vehicle in a future time period, wherein the lane changing action is any one of changing lanes to the left, maintaining the lane, and changing lanes to the right; For each of the multiple cooperative vehicles, multiple lane change actions of each cooperative vehicle in a future time period are evaluated to obtain a reward function value corresponding to each cooperative vehicle, wherein the reward function value corresponding to the cooperative vehicle is used to represent: the impact of the cooperative vehicle's driving in the future time period using the planned lane change action on the traffic flow in the future time period; Traverse various combinations of multiple lane change actions of multiple cooperative vehicles in the future time period to generate multiple multi-vehicle cooperative lane change plans; Determining a total reward function value for each multi-vehicle coordinated lane change solution based on the reward function values corresponding to the multiple coordinated vehicles; According to the multi-vehicle cooperative lane changing plan with the largest total reward function value, control instructions are sent to the multiple cooperative vehicles.

[0006] Optionally, after planning multiple lane changing actions for each of the multiple cooperative vehicles in a future time period, the method further includes at least one of the following: For each of the multiple cooperative vehicles, predict the risk of collision with obstacles or surrounding vehicles after the cooperative vehicle performs the planned lane change maneuver in a future time period, and filter out lane change maneuvers with a collision risk greater than a risk threshold from the lane change maneuvers planned for the cooperative vehicle; For each of the multiple cooperating vehicles, when the lane change action for the cooperating vehicle includes at least two lane change actions, determining a time interval between each two adjacent lane change actions, and filtering out lane change actions with a time interval less than a threshold from the two adjacent lane change actions; For each of the multiple cooperating vehicles, based on lane congestion or navigation information of the cooperating vehicle, verifying the rationality of the planned lane change maneuvers for the cooperating vehicle in a future time period, and filtering out lane change maneuvers with rationality less than a rationality threshold from the lane change maneuvers planned for the cooperating vehicle; Traverse all combinations of multiple lane change actions of multiple coordinated vehicles in the future time period and generate multiple multi-vehicle coordinated lane change plans, including: Traverse each combination of filtered lane change actions of multiple cooperative vehicles in the future time period to generate multiple multi-vehicle cooperative lane change plans.

[0007] Optionally, for each of the multiple cooperative vehicles, multiple lane change actions of each cooperative vehicle in a future time period are evaluated to obtain a reward function value corresponding to each cooperative vehicle, including: For each of the multiple cooperative vehicles, predict the risk of collision with obstacles or surrounding vehicles after the cooperative vehicle uses the planned lane change maneuver in the future time period, and obtain a safety score for the cooperative vehicle; For each of the multiple cooperative vehicles, predict the speed of vehicles in each lane after the cooperative vehicle uses the planned lane change action in the future time period to obtain an efficiency score for the cooperative vehicle; For each of the multiple cooperative vehicles, predict the cooperative information between the cooperative vehicle and other vehicles after the cooperative vehicle drives with the planned lane change maneuver in the future time period, and obtain a cooperative degree score for the cooperative vehicle; For each of the multiple cooperative vehicles, a reward function value corresponding to the cooperative vehicle is obtained according to the safety score, efficiency score and cooperation score of the cooperative vehicle.

[0008] Optionally, for each of the multiple cooperative vehicles, multiple lane change actions are planned for each cooperative vehicle in a future time period, including: Prioritizing the multiple cooperative vehicles according to their vehicle types, speeds, and distances between their current locations and navigation lane change points; Plan multiple lane change actions for the cooperative vehicle in the future time period in descending order of priority; Among them, the priority of a cooperative vehicle whose vehicle type is a special vehicle is higher than the priority of a cooperative vehicle whose vehicle type is a non-special vehicle; the higher the speed of the cooperative vehicle, or the closer the distance between the current position point and the navigation lane change point, the higher the priority of the cooperative vehicle.

[0009] Optionally, traversing various combinations of multiple lane change actions of multiple cooperative vehicles in a future time period to generate multiple multi-vehicle cooperative lane change schemes, including: Determining the levels of the plurality of cooperative vehicles in a decision tree in descending order of priority, wherein a first cooperative vehicle with the highest priority is located at a first level of the decision tree; According to the reward function values of the first cooperative vehicle for changing lanes to the left, keeping lane, and changing lanes to the right in the future time period, the lane changing action with the largest reward function value is determined as the root node of the decision tree; For the second level of the decision tree, the second cooperative vehicle at the second level of the decision tree changes lanes to the left, maintains lanes, and changes lanes to the right in the future time period, respectively, as a child node of the root node. Based on the cumulative reward function value of each decision path, the child node on the decision path with the largest cumulative reward function value is selected as the second target child node. A decision path is a path from a root node to a child node of the root node. For the nth level in the decision tree, the cooperative vehicles at the nth level of the decision tree that change lanes to the left, keep lanes, and change lanes to the right in the future time period are respectively used as a child node of the n-1th target child node. Based on the cumulative reward function value of each decision path, the child node on the decision path with the largest cumulative reward function value is used as the nth target child node. A decision path is a path from a root node to a child node of the n-1th target child node, where n is an integer greater than 2. For the lowest level in the decision tree, the cooperative vehicles at the lowest level of the decision tree that change lanes to the left, keep lanes, and change lanes to the right in the future time period are respectively regarded as leaf nodes of the decision tree, and each decision path is regarded as a multi-vehicle cooperative lane change plan. A decision path is the path from the root node to a leaf node.

[0010] Optionally, before sending control instructions to the multiple cooperative vehicles according to the multi-vehicle cooperative lane change solution with the maximum total reward function value, the method further includes: With the goal of minimizing the variance of the speeds of the multiple cooperative vehicles during the lane change process and minimizing the total lane change time of the multiple cooperative vehicles, the multi-vehicle cooperative lane change scheme with the maximum total reward function value is optimized while satisfying at least one constraint condition to obtain an optimized multi-vehicle cooperative lane change scheme; the at least one constraint condition includes at least one of the following: The shortest lane change time is greater than the first time duration and the longest lane change time is less than the second time duration; The speed parameter values of the multiple cooperative vehicles during the lane change process are within a speed parameter value threshold range; During the lane change process, the relative driving parameter values between the multiple cooperative vehicles and the vehicles in front and behind are greater than the minimum safe driving parameter value; According to the multi-vehicle coordinated lane change solution with the maximum total reward function value, control instructions are sent to the multiple coordinated vehicles, including: According to the optimized multi-vehicle cooperative lane changing plan, control instructions are sent to the multiple cooperative vehicles.

[0011] Optionally, the method further includes: Determine multiple vehicles in the same lane as the multiple cooperative vehicles, or Determine a plurality of vehicles whose mutual distances are within a distance threshold range as the plurality of cooperative vehicles; A plurality of vehicles having target key features are determined as the plurality of cooperative vehicles.

[0012] A second aspect of an embodiment of the present application provides a multi-vehicle coordinated lane change control device, the device comprising: a lane change action planning module, configured to plan, for each of the plurality of cooperating vehicles, a plurality of lane change actions within a future time period, wherein the lane change action is any one of a left lane change, a lane hold, and a right lane change; a lane change action evaluation module for evaluating, for each of the plurality of coordinated vehicles, a plurality of lane change actions of each coordinated vehicle in a future time period, and obtaining a reward function value corresponding to each coordinated vehicle, wherein the reward function value corresponding to each coordinated vehicle is used to represent: the impact of the coordinated vehicle's driving in the future time period using the planned lane change action on the traffic flow in the future time period; A multi-vehicle coordinated lane change plan generation module is used to traverse various combinations of multiple lane change actions of multiple coordinated vehicles in the future time period and generate multiple multi-vehicle coordinated lane change plans; a total reward function value determination module, configured to determine a total reward function value for each multi-vehicle coordinated lane change solution based on the reward function values corresponding to the plurality of coordinated vehicles; A control instruction sending module is used to send control instructions to the multiple cooperative vehicles according to the multi-vehicle cooperative lane change solution with the maximum total reward function value.

[0013] According to a third aspect of an embodiment of the present application, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the method described in the first aspect.

[0014] In a fourth aspect of the embodiments of the present application, a computer program product is provided, including a computer program, which implements the method described in the first aspect when executed by a processor.

[0015] According to a fifth aspect of the embodiments of the present application, a computer-readable storage medium is provided, on which a computer program is stored, wherein when the computer program is executed by a processor, the method described in the first aspect is implemented.

[0016] Beneficial effects of this application: An embodiment of the present application provides a multi-vehicle cooperative lane change control method, apparatus, equipment and medium, including: for each of a plurality of cooperative vehicles, planning a plurality of lane change actions for each cooperative vehicle in a future time period, wherein the lane change action is any one of changing lanes to the left, keeping lanes and changing lanes to the right; for each of a plurality of cooperative vehicles, evaluating the plurality of lane change actions of each cooperative vehicle in the future time period to obtain a reward function value corresponding to each cooperative vehicle, wherein the reward function value corresponding to the cooperative vehicle is used to characterize: the impact of the cooperative vehicle driving with the planned lane change action in the future time period on the traffic flow in the future time period; traversing various combinations of the plurality of lane change actions of the plurality of cooperative vehicles in the future time period to generate a plurality of multi-vehicle cooperative lane change schemes; determining the total reward function value of each multi-vehicle cooperative lane change scheme based on the reward function values corresponding to the plurality of cooperative vehicles; and sending control instructions to the plurality of cooperative vehicles according to the multi-vehicle cooperative lane change scheme with the largest total reward function value.

[0017] Through the technical solution of the present application, multiple lane changing actions are planned for each cooperative vehicle, which increases the diversity of decision-making and provides a basis for subsequent optimization. The lane changing action of each vehicle is quantitatively evaluated through the reward function, and safety, efficiency and coordination are comprehensively considered to ensure the comprehensiveness and objectivity of the evaluation. Subsequently, multiple complete multi-vehicle cooperative lane changing schemes are generated. By calculating the total reward function value of each scheme, the pros and cons of each scheme can be evaluated as a whole, and the optimal lane changing scheme can be selected. Finally, control instructions are sent to the cooperative vehicle according to the scheme with the largest total reward function value, realizing the cooperative lane changing of multiple vehicles, significantly improving the traffic efficiency and safety in the autonomous driving scenario, not only improving the safety of lane changing of multiple cooperative vehicles, but also improving the overall traffic efficiency, ensuring the coordination and consistency of the lane changing actions of multiple cooperative vehicles, avoiding traffic chaos, and realizing the overall optimization of the entire transportation system. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The drawings that constitute a part of this application are used to provide further understanding of this application. The illustrative embodiments of this application and their descriptions are used to explain this application and do not constitute improper limitations on this application.

[0019] In order to more clearly illustrate the technical solution of the present application, the following is a brief introduction to the drawings required for the description of the present application. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0020] Figure 1 This is a flowchart of a multi-vehicle cooperative lane change control method provided by an embodiment of the present application; Figure 2 is a schematic diagram of a decision tree shown in an embodiment of the present application; Figure 3 This is a schematic diagram of a framework of a multi-vehicle cooperative lane change control device provided by an embodiment of the present application; Figure 4 Schematic diagram of an electronic device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0021] It should be noted that, unless there is any conflict, the embodiments and features in the embodiments of this application can be combined with each other.

[0022] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0023] Figure 1 This is a flow chart of a multi-vehicle cooperative lane change control method provided by an embodiment of the present application, with reference to Figure 1 An embodiment of the present application provides a multi-vehicle coordinated lane change control method, the method comprising steps S11 to S15: In step S11 , for each of the multiple cooperative vehicles, a plurality of lane changing actions are planned for each cooperative vehicle in a future time period, wherein the lane changing action is any one of changing lanes to the left, keeping the lane, and changing lanes to the right.

[0024] In this embodiment, when there are multiple cooperative vehicles on the current road, in order to reasonably plan multiple lane changing actions of each cooperative vehicle in the future time period, it is necessary to perform forward state deduction on each of the multiple cooperative vehicles.

[0025] Forward state deduction expands the state of each cooperative vehicle into the joint state of multiple cooperative vehicles. Based on the lane change problem, three possible lane change actions are set for each cooperative vehicle: left lane change, lane hold, and right lane change. The state of each cooperative vehicle is then modeled so that each cooperative vehicle's lane change action can take into account the lane changes of other vehicles. In this process, approximate deduction is performed using dynamic models and control. Vehicles that do not participate in the coordinated lane change decision-making control are divided into cooperative vehicles and non-cooperative vehicles. Since non-cooperative vehicles' lane changes are not under unified system control, their positions are deduced using predicted trajectories and pure lateral tracking control. Vehicle acceleration and deceleration are deduced using longitudinal PID control. Since the lane change actions of cooperative vehicles are under unified system control, simultaneous path planning is performed for the three states of left lane change, lane hold, and right lane change. Lateral pure tracking control is used for deduction, and vehicle acceleration and deceleration are deduced using longitudinal PID control.

[0026] (1) Longitudinal PID control is based on the following formula:

[0027] in: To control the acceleration and deceleration of the output, is the error term, , To achieve the desired safety distance, is the actual distance, are the proportional, integral, and differential coefficients.

[0028] (2) Lateral pure tracking control is as follows: Define a "look-ahead point" in front of the vehicle and calculate the front wheel steering angle to make the vehicle face this point.

[0029]

[0030] in: To control the output front wheel angle, is the vehicle wheelbase, The current heading angle of the vehicle Direction of preview point The angle of It is the preview distance (generally 1.5 to 2 times the vehicle speed).

[0031] (3) The vehicle dynamics model is as follows:

[0032] in: and is the position of the vehicle in the Cartesian coordinate system, is the current heading angle of the vehicle, is the vehicle wheelbase, is the vehicle speed.

[0033] (4) The lane change path planning of the cooperative vehicle is based on the following formula:

[0034] in: is the longitudinal distance along the lane centerline, is the lateral distance along the lane centerline, For time, and are the coefficients of the polynomial to be determined.

[0035] In combination with the above embodiments, the present application also provides another multi-vehicle coordinated lane change control method. In this method, in addition to the above steps, the method further includes: In step S21 , a plurality of vehicles located in the same lane are determined as the plurality of cooperative vehicles.

[0036] In this embodiment, in the multi-vehicle coordinated lane change control method, it is necessary to first determine which vehicles will participate in the coordinated lane change. This step ensures the pertinence and effectiveness of the lane change decision. Specifically, the cooperative vehicles can be determined in one of the following three ways: First, the lane-position-based approach identifies multiple vehicles in the same lane as cooperative vehicles. For example, if vehicles A, B, and C are all traveling in the right lane, these three vehicles are identified as cooperative vehicles so that their lane change behaviors can be planned in a unified manner.

[0037] Or, in step S22, a plurality of vehicles whose mutual distances are within a distance threshold range are determined as the plurality of cooperative vehicles.

[0038] Secondly, the vehicle-to-vehicle distance-based approach identifies multiple vehicles as cooperative vehicles if their distances are within a distance threshold. For example, if the distance between vehicle A and vehicle B is less than a preset distance threshold (e.g., 20 meters), the two vehicles will be identified as cooperative vehicles.

[0039] In step S23 , a plurality of vehicles having the target key features are determined as the plurality of cooperative vehicles.

[0040] Finally, the key feature-based approach identifies multiple vehicles with target key features as cooperative vehicles. Target key features can be determined based on vehicle type (e.g., special vehicles), the vehicle's navigation goal (e.g., needing to change lanes to a specific lane), and the vehicle's urgency.

[0041] For example, if vehicle C is an ambulance and needs to pass through the current road section quickly, vehicle C and its surrounding vehicles can be identified as cooperative vehicles to provide vehicle C with a priority lane change path.

[0042] In step S12, for each of the multiple cooperative vehicles, the lane changing action of each cooperative vehicle in the future time period is evaluated to obtain a reward function value corresponding to each cooperative vehicle. The reward function value corresponding to the cooperative vehicle is used to represent: the impact of the cooperative vehicle driving with the planned lane changing action in the future time period on the traffic flow in the future time period.

[0043] In this embodiment, each coordinated vehicle's planned lane change is evaluated. The goal of this evaluation is to calculate the impact of each coordinated vehicle's planned lane change on future traffic flow. This impact is quantified using a reward function that comprehensively considers multiple factors, including the safety, efficiency, and coordination of the coordinated vehicle's planned lane change.

[0044] For example, if a coordinated vehicle plans to change lanes to the left, it will be evaluated to see whether this action will bring it too close to other coordinated vehicles (a safety issue), whether it will cause traffic congestion (an efficiency issue), and whether it is consistent with the driving status of other coordinated vehicles (a coordination issue). The reward function value of each coordinated vehicle reflects the quality of its lane change action. A higher score indicates that the coordinated vehicle's planned lane change action is more beneficial to the overall traffic flow of multiple coordinated vehicles.

[0045] Step S13, traverse various combinations of multiple lane changing actions of multiple cooperative vehicles in the future time period to generate multiple multi-vehicle cooperative lane changing plans.

[0046] In this embodiment, after evaluating the lane changing action of each cooperative vehicle, it is necessary to further consider the lane changing action combination of all cooperative vehicles. In this process, by traversing all possible combinations of multiple lane changing actions of multiple cooperative vehicles in the future time period, multiple multi-vehicle cooperative lane changing schemes are generated. A multi-vehicle cooperative lane changing scheme corresponds to a combination of multiple lane changing actions of multiple cooperative vehicles in the future time period.

[0047] Optionally, in the process of traversing all possible combinations of multiple lane changing actions of multiple cooperative vehicles in a future time period, a corresponding multi-vehicle cooperative lane changing solution can be obtained by generating a decision tree.

[0048] For example, there are two cooperating vehicles (vehicle C and vehicle D), each of which has three possible lane change actions, and all three possible lane change actions are reasonable, so 9 combinations can be obtained, and then a corresponding multi-vehicle cooperative lane change plan is generated based on each of these 9 combinations.

[0049] Step S14: determining a total reward function value for each multi-vehicle cooperative lane change scheme based on the reward function values corresponding to the multiple cooperative vehicles.

[0050] In this embodiment, for each generated multi-vehicle coordinated lane change plan, a total reward function value is calculated for that plan. This total reward function value is calculated based on the reward function value corresponding to each cooperating vehicle, reflecting the impact of the multi-vehicle coordinated lane change plan on overall traffic flow. A higher total reward function value indicates that the multi-vehicle coordinated lane change plan is more beneficial to overall traffic flow.

[0051] Step S15: Send control instructions to the multiple cooperative vehicles according to the multi-vehicle cooperative lane change plan with the largest total reward function value.

[0052] In this embodiment, after obtaining the total reward function values corresponding to each multi-vehicle coordinated lane change scheme, the scheme with the highest total reward function value is selected, and control instructions are sent to each coordinated vehicle based on this scheme. The control instructions are used to guide the coordinated vehicles to change lanes according to the lane change action corresponding to the multi-vehicle coordinated lane change scheme with the highest total reward function value in the future time period to achieve optimal traffic flow.

[0053] Through the above embodiment, multiple lane changing actions are planned for each cooperative vehicle, which increases the diversity of decision-making and provides a basis for subsequent optimization. The lane changing action of each vehicle is quantitatively evaluated through the reward function, and safety, efficiency and coordination are comprehensively considered to ensure the comprehensiveness and objectivity of the evaluation. Subsequently, multiple complete multi-vehicle cooperative lane changing schemes are generated. By calculating the total reward function value of each scheme, the pros and cons of each scheme can be evaluated as a whole, and the optimal lane changing scheme can be selected. Finally, control instructions are sent to the cooperative vehicle according to the scheme with the largest total reward function value, realizing the cooperative lane changing of multiple vehicles, significantly improving the traffic efficiency and safety in the autonomous driving scenario, not only improving the safety of lane changing of multiple cooperative vehicles, but also improving the overall traffic efficiency, ensuring the coordination and consistency of the lane changing actions of multiple cooperative vehicles, avoiding traffic chaos, and realizing the overall optimization of the entire transportation system.

[0054] In combination with the above embodiments, the present application further provides another multi-vehicle coordinated lane change control method. In this method, in addition to the above steps, after executing step S11 of "planning multiple lane change actions for each of the multiple coordinated vehicles in a future time period," the method further includes at least one of the following: In step S31, for each of the multiple cooperative vehicles, the risk of collision with obstacles or surrounding vehicles after the cooperative vehicle drives with the planned lane change action in the future time period is predicted, and the lane change actions with a collision risk greater than the risk threshold are filtered out from the lane change actions planned for the cooperative vehicle.

[0055] In this embodiment, after planning each coordinated vehicle's lane change, the system predicts the risk of collision with obstacles or surrounding vehicles during the planned lane change. If the collision risk of a lane change exceeds a preset risk threshold, the lane change will be filtered out. For example, if the distance between the vehicle in front and behind the lane change is too close or the TTC (Time-To-Collision) is too low (TTC, typically calculated as distance divided by relative speed; a TTC < 2 is considered dangerous).

[0056] For example, if the cooperative vehicle plans to change lanes to the left, but the prediction result indicates that the distance to the vehicle D on the left will be too close, making the collision risk high, then the lane change to the left will be filtered out, and the unfiltered lane change will be retained to ensure that all lane change actions entering the subsequent evaluation are reasonable.

[0057] In step S32, for each of the multiple cooperative vehicles, its lane changing action cannot be repeated, that is, after completing one lane change, a certain time interval must be allowed before the next lane change can be performed. Therefore, when the lane changing action of the cooperative vehicle includes at least two lane changing actions, the time interval between each two adjacent lane changing actions is determined, and the lane changing actions with a time interval less than the interval threshold are filtered out from the two adjacent lane changing actions.

[0058] In this embodiment, when the lane change action of a cooperative vehicle includes multiple lane change actions, it is necessary to check the time interval between each two adjacent lane change actions in the lane change action. When the time interval is less than the preset interval threshold, it will be considered that such lane change actions are too frequent and may cause unstable driving or increase the risk of collision, so these lane change actions will be filtered out.

[0059] For example, assuming the preset interval threshold is 10 seconds, the cooperative vehicle plans to change lanes to the left first and then to the right, and the time interval between the two lane changes is 7 seconds, which is less than the interval threshold of 10 seconds. The second lane change action is considered unreasonable, so the second lane change action is filtered out to ensure the rationality and safety of each lane change action.

[0060] In step S33, for each of the multiple cooperative vehicles, based on the lane congestion situation or the navigation information of the cooperative vehicle, the rationality of the cooperative vehicle driving with the planned multiple lane changing actions in the future time period is verified, and the lane changing actions with rationality less than the rationality threshold are filtered out from the multiple lane changing actions planned for the cooperative vehicle.

[0061] In this embodiment, the rationality of each lane changing action among multiple lane changing actions planned for the cooperative vehicle can be verified based on the congestion situation of the lane where each cooperative vehicle is located and the vehicle navigation information of the cooperative vehicle.

[0062] For example, if the cooperative vehicle plans to change lanes to the left but the left lane is already congested, or if navigation information indicates that the cooperative vehicle does not need to change lanes, the rationality of the lane change to the left will be low. If the rationality falls below a preset rationality threshold, the lane change will be filtered out, ensuring that the lane change is not only safe but also meets the actual traffic conditions and driving needs.

[0063] Optionally, in the process of obtaining the corresponding multi-vehicle cooperative lane changing plan by generating a decision tree, unreasonable lane changing actions in the decision tree can be pruned according to the various considerations in the above steps S31 to S33, that is, the corresponding lane changing actions can be filtered out.

[0064] Figure 2 This is a schematic diagram of a decision tree shown in an embodiment of the present application, with reference to Figure 2 , shows each lane change action among multiple lane change actions planned for vehicles A to D (wherein centering means maintaining the lane), wherein individual lane change actions are filtered out according to the above-mentioned considerations.

[0065] Specifically, in Figure 2 In the figure, the dotted arrows represent multiple optional lane change actions, the crossed-out lane change actions indicate that unsafe, unreasonable or unstable lane change actions are filtered out, and the solid arrows correspond to the lane change actions finally selected by the cooperative vehicle.

[0066] For example: the multiple lane change actions available to vehicle A include changing left, driving in the center, and changing right. Among them, the lane change action of vehicle A to change left is filtered out, and changing right is the optimal lane change action finally selected by vehicle A from the two lane change actions of driving in the center and changing right. After determining the lane change action finally selected by vehicle A, the lane change action of vehicle B is selected. Similarly, the lane change action of vehicle B to change right is filtered out, and driving in the center is the lane change action finally selected by vehicle B. After determining the lane change action finally selected by vehicle B, the lane change action of vehicle C is determined. The three lane change actions of vehicle C are all safe and reasonable and are not filtered out. Changing left is the lane change action finally selected by vehicle C. After determining the lane change action finally selected by vehicle C, the lane change action of vehicle D is determined. The three lane change actions of vehicle D are all safe and reasonable and are not filtered out. Changing left is the lane change action finally selected by vehicle D.

[0067] The step S13 of "traversing various combinations of multiple lane change actions of multiple coordinated vehicles in the future time period to generate multiple multi-vehicle coordinated lane change plans" includes: Step S13-1, traverses various combinations of filtered lane change actions of multiple cooperative vehicles in the future time period to generate multiple multi-vehicle cooperative lane change plans.

[0068] In this embodiment, after the filtering step, multiple multi-vehicle coordinated lane change plans are generated by traversing all filtered lane change combinations of all cooperating vehicles. These plans serve as candidate plans for subsequent optimization. By filtering out unsafe, unreasonable, or unstable lane change maneuvers, the optimal multi-vehicle coordinated lane change plan can be more efficiently evaluated and selected.

[0069] Through the above-mentioned embodiments, the safety and rationality of multi-vehicle cooperative lane changes can be further improved. Specifically, by predicting collision risks and filtering out high-risk lane change actions, it is ensured that all lane change actions entering subsequent evaluations have acceptable safety, thereby effectively avoiding collision accidents. By determining the time interval of lane change actions and filtering out unreasonable lane change actions, the stability and rationality of lane change behavior are ensured, and driving instability or traffic chaos caused by frequent lane changes are avoided. By verifying the rationality of lane change actions based on lane congestion and navigation information, it is ensured that lane change behavior is not only safe, but also meets actual traffic conditions and driving needs. Finally, by generating filtered multi-vehicle cooperative lane change plans, it is possible to more efficiently evaluate and select the optimal multi-vehicle cooperative lane change plan, thereby improving overall traffic efficiency.

[0070] In combination with the above embodiments, the present application further provides another multi-vehicle coordinated lane change control method. In this method, in addition to the above steps, the step S13 of "traversing various combinations of multiple lane change actions of multiple coordinated vehicles in the future time period to generate multiple multi-vehicle coordinated lane change plans" includes steps S13-2 to S13-6: Step S13-2: Determine the levels of the multiple cooperative vehicles in the decision tree in descending order of priority, wherein the first cooperative vehicle with the highest priority is located at the first level of the decision tree.

[0071] In this embodiment, as a preferred approach, a Monte Carlo tree search algorithm may be used to generate multiple multi-vehicle cooperative lane change schemes.

[0072] First, the decision-making priority of each of the multiple cooperative vehicles needs to be determined. A priority value is calculated for each cooperative vehicle based on factors such as vehicle type, speed, and distance from the lane change point.

[0073] Among vehicle types, ambulances have the highest priority, followed by fire trucks, buses, and finally ordinary passenger cars. Faster vehicles have higher priority, as do vehicles closer to the lane change point. By assigning different weights to different factors, the priority of each coordinated vehicle can be calculated. Multiple coordinated vehicles are then ranked from highest to lowest priority to determine the order of decision-making priorities. The highest-priority coordinated vehicle (i.e., the first coordinated vehicle) is placed at the first level of the decision tree.

[0074] Step S13-3: Based on the respective reward function values of the first cooperative vehicle for changing lanes to the left, keeping lane, and changing lanes to the right in the future time period, the lane changing action with the largest reward function value is determined as the root node of the decision tree.

[0075] In this embodiment, the next step is the selection phase. Starting with the highest-priority vehicle (the first cooperating vehicle), three possible lane change maneuvers are considered: left lane change, lane hold, and right lane change. For each lane change, a corresponding reward function value is calculated. The reward function value represents the impact of the planned lane change on future traffic flow. The lane change with the highest reward function value is then determined as the root node of the decision tree.

[0076] Step S13-4, for the second level of the decision tree, the second cooperative vehicle located at the second level of the decision tree changes lanes to the left, maintains lanes, and changes lanes to the right in the future time period, respectively, as a child node of the root node, and according to the cumulative reward function value of each decision path, the child node on the decision path with the largest cumulative reward function value is used as the second target child node. A decision path is a path from the root node to a child node of the root node.

[0077] In this example, the optimal action for the first cooperating vehicle has been selected during the above process. The node corresponding to this optimal action serves as the root node. Now, the focus is on the decision of the second cooperating vehicle (the vehicle with the second highest priority). For this second cooperating vehicle, the expansion phase identifies three possible lane change actions: left lane change, lane hold, and right lane change. These three lane change actions correspond to the three child nodes connected to the root node. In other words, the second level of the decision tree has three child nodes, each representing a possible lane change action for the second cooperating vehicle.

[0078] For each child node of the second level, it is necessary to calculate the cumulative reward function value of the decision path from the root node to the child node. The cumulative reward function value is obtained through the simulation stage. The cumulative reward function value reflects the impact of these cooperative vehicles on future traffic flow when they drive according to a series of lane changing actions along the planned decision path from the root node to the child node. By comparing the cumulative reward function values of the three decision paths corresponding to the three child nodes of the second level, the decision path with the largest cumulative reward function value is selected. The child node on the decision path with the largest cumulative reward function value is the second target child node. The second target child node corresponds to the optimal lane changing action of the second cooperative vehicle. The second target child node is used as the current node to continue to select the optimal lane changing action for the next cooperative vehicle.

[0079] Step S13-5, for the nth level in the decision tree, the cooperative vehicle located at the nth level of the decision tree changes lanes to the left, keeps lanes, and changes lanes to the right in the future time period, respectively, as a child node of the n-1th target child node, and according to the cumulative reward function value of each decision path, the child node on the decision path with the largest cumulative reward function value is used as the nth target child node. A decision path is a path from the root node to a child node of the n-1th target child node, and n is an integer greater than 2.

[0080] In this embodiment, in the above process, the optimal actions have been selected for the first cooperative vehicle and the second cooperative vehicle. At this time, the decision of the cooperative vehicle at the nth level (the vehicle with the nth highest priority) needs to be paid attention to. Through the expansion stage, three possible lane changing actions are determined: changing lanes to the left, keeping the lane, and changing lanes to the right. These three lane changing actions correspond to the three sub-nodes connected to the previous level (that is, each of which is a sub-node of the n-1th target sub-node). Each sub-node represents a possible lane changing action of the cooperative vehicle at the nth level.

[0081] For each child node of the nth level, it is necessary to calculate the cumulative reward function value of the decision path from the root node to the child node. The cumulative reward function value is obtained through the simulation stage. By comparing the cumulative reward function values of the three decision paths corresponding to the three child nodes of the nth level, the decision path with the largest cumulative reward function value is selected. The child node on the decision path with the largest cumulative reward function value is the nth target child node. The nth target child node corresponds to the optimal lane changing action of the cooperative vehicle at the nth level. The nth target child node is used as the current node to continue to select the optimal lane changing action for the next cooperative vehicle.

[0082] In step S13-6, for the lowest level in the decision tree, the cooperative vehicles at the lowest level of the decision tree that change lanes to the left, keep lanes, and change lanes to the right in the future time period are respectively regarded as leaf nodes of the decision tree, and each decision path is regarded as a multi-vehicle cooperative lane change plan. A decision path is the path from the root node to a leaf node.

[0083] In this embodiment, in the above process, the optimal action has been selected for all cooperative vehicles except the lowest-level cooperative vehicle. The method for determining the optimal action of the lowest-level cooperative vehicle also refers to the determination process of the above level. In the decision tree, each decision path is regarded as a multi-vehicle cooperative lane changing scheme. Therefore, the lane changing action corresponding to each node on the decision path from the root node to a leaf node is the lane changing action of each cooperative vehicle in a multi-vehicle cooperative lane changing scheme, and the lane changing action of each node corresponding to each decision path.

[0084] Combined with the contents of step S14 and step S15, after obtaining all the decision paths from the root node to each leaf node, the cumulative reward function value of each decision path can be obtained. The cumulative reward function value of each decision path from the root node to the leaf node corresponds to the total reward function value of a multi-vehicle cooperative lane changing scheme. The combination of lane changing actions corresponding to each node in the decision path with the largest total reward function value is the optimal lane changing action of each of the multiple cooperative vehicles, and then control instructions are issued to each cooperative vehicle according to the optimal lane changing action.

[0085] Alternatively, the target child node can be determined by calculating the UCT value of each child node. For each child node, the UCT value of the child node is calculated, and the child node with the largest UCT value is selected as the target child node. The UCT value is calculated by combining the average reward value (Q value) of the node and an exploration term. The exploration term encourages the algorithm to explore child nodes that have been visited less frequently.

[0086] During the expansion process, when an incompletely expanded child node is explored, that is, a node corresponding to a new lane-changing action of a cooperative vehicle, a new child node is generated. Starting from this newly expanded child node, the lane-changing actions of subsequent cooperative vehicles are simulated randomly until the simulation of the lane-changing action of the last cooperative vehicle is completed.

[0087] Finally, through the back-propagation stage, the cumulative reward function value under the simulated path is back-propagated to the root node, and the statistical information of each node on the path is updated, that is, the number of visits and the average reward value of each node are updated (for example, the number of visits increases by 1, and the average reward value is updated according to the new reward value), so that when the next selection is made, the UCT values of these nodes can more accurately reflect the potential value of each node in the decision tree.

[0088] For example, suppose there are three coordinated vehicles (car 1, car 2, car 3), and the initial state of each coordinated vehicle is as follows: Vehicle 1: Position 100, speed 120, lane 1, type: ambulance, cooperative vehicle.

[0089] Car 2: Position 150, speed 100, lane 2, type is passenger car, cooperative vehicle.

[0090] Vehicle 3: Position 50, speed 80, lane 1, type is bus, cooperative vehicle.

[0091] Calculate the priority value: Vehicle type T: ambulance (4), fire truck (3), bus (2), passenger car (1) Speed V: The higher the speed, the higher the priority. Considering that compared to slower cooperating vehicles, faster cooperating vehicles require less time to execute lane changes in the future time period and are less likely to cause traffic jams in the future time period, lane changes in the future time period are prioritized for faster cooperating vehicles. Among them, in order to ensure the introduction of speed factors, various priority value factors can be fairly compared and optimized at the same scale. For speed, the Z-score standardization method can be used. First, the average speed of each cooperative vehicle is calculated. and standard deviation , for the speed of each cooperative vehicle ,according to Perform standardization transformation to obtain Then, Speed is factored into the priority calculation. This process maps the speeds of all coordinated vehicles to a relatively uniform scale space. This ensures that all factors in the priority calculation have a fair influence and avoids biased targeting due to scale issues.

[0092] Among them, the average speed of each cooperative vehicle is 100 and the standard deviation is 16.33; Therefore, the normalized speed of car 1 is 1.22, the normalized speed of car 2 is 0, and the normalized speed of car 3 is -1.22.

[0093] Distance to lane change point D: The closer the distance, the higher the priority (assuming the lane change point is at position 200). When the distance is not less than 150, the priority value P is calculated as 1; when the distance is not less than 100, the priority value P is calculated as 2; when the distance is not less than 50, the priority value P is calculated as 3, and so on; Calculate the priority value P of each cooperative vehicle: P=α*T+β*V+γ*D; Where α, β, and γ ∈ [0, 1], and α + β + γ = 1. The weight coefficients α, β, and γ can be initialized based on expert experience. For example, the following weight coefficients can be preset: α = 0.5, β = 0.3, and γ = 0.2. This embodiment does not limit the specific method for setting α, β, and γ. Any setting method that satisfies α + β + γ = 1 can implement the content of this embodiment.

[0094] Assume the weights are: α=0.5, β=0.3, γ=0.2; For each cooperative vehicle’s priority value P: Car 1: P = 0.5*4 + 0.3*1.22 + 0.2*2 = 6.06; Car 2: P=0.5*1+0.3*0+0.2*3=1.1; Car 3: P = 0.5*1 + 0.3*(-1.22) + 0.2*1 = -2.96; Therefore, the decision priority of each cooperative vehicle is ranked as follows (priority vehicle 1 > vehicle 2 > vehicle 3): vehicle 1, vehicle 2, vehicle 3; Car 1 is at the first level of the decision tree, car 2 is at the second level, and car 3 is at the third level.

[0095] Lane change actions of vehicle 1: change lane left, keep lane, change lane right; Calculate the reward function value for each lane change action: Assume that the reward function values of car 1's lane change actions are: changing lane left: 10, keeping lane: 12, changing lane right: 8. Select the lane change action with the largest reward function value (keeping lane, reward value 12) as the root node.

[0096] At this point, the lane change action of car 2 is expanded to obtain three lane change actions: change lane to the left, keep lane, and change lane to the right. These three actions correspond to the three child nodes of the root node respectively: Subnode 1.1: Car 2 changes lanes to the left; Child node 1.2: Car 2 maintains lane; Subnode 1.3: Car 2 changes lanes to the right; Assume that the cumulative reward function value of each path is calculated through the simulation stage: Path 1.1 (Car 1 stays in lane - Car 2 changes lanes left): Cumulative reward function value 18; Path 1.2 (Car 1 keeps lane - Car 2 keeps lane): Cumulative reward function value is 20; Path 1.3 (Car 1 stays in lane - Car 2 changes lanes right): Cumulative reward function value is 17; Select the child node on the path with the largest cumulative reward function value (path 1.2, car 1 keeps lane - car 2 keeps lane, cumulative reward function value 20) as the second target child node.

[0097] At this point, the lane change action of car 3 is expanded to obtain three lane change actions: change lane to the left, keep lane, and change lane to the right. These three lane change actions correspond to the three child nodes connected to the second target child node: Subnode 1.2.1: Car 3 changes lanes to the left; Subnode 1.2.2: Car 3 maintains lane; Subnode 1.2.3: Car 3 changes lanes to the right; Calculate the cumulative reward function value for each path: Assume that the cumulative reward function values obtained through the simulation phase are: Path 1.2.1 (Car 1 maintains lane - Car 2 changes lane to the left - Car 3 changes lane to the left): Cumulative reward function value is 22; Path 1.2.2 (Car 1 maintains lane - Car 2 changes lane left - Car 3 maintains lane): Cumulative reward function value 24; Path 1.2.3 (Car 1 maintains lane - Car 2 changes lanes to the left - Car 3 changes lanes to the right): Cumulative reward function value 21; Select the child node on the path with the largest cumulative reward function value (path 1.2.2, car 1 stays in the lane - car 2 changes lanes left - car 3 stays in the lane, cumulative reward function value 24) as the nth target child node.

[0098] In this example, car 3 is already the lowest-level cooperative car. Therefore, the path with the largest cumulative reward function value (path 1.2.2, car 2 keeps lane, cumulative reward function value 24) is selected as the final multi-vehicle cooperative lane change solution.

[0099] That is, the final multi-vehicle cooperative lane change solution is: Car 1: Keep lane; Car 2: Change lanes to the left; Car 3: Change lanes to the right; Through the above steps, the Monte Carlo Tree Search (MCTS) algorithm was utilized, and combined with the vehicle priority and reward function value, the optimal lane change action was selected for each vehicle, ensuring that vehicles can change lanes collaboratively in complex traffic environments, improving traffic efficiency and safety.

[0100] In combination with the above embodiments, the present application also provides another multi-vehicle coordinated lane change control method. In this method, in addition to the above steps, step S12 of "evaluating multiple lane change actions of each of the multiple coordinated vehicles in the future time period to obtain a reward function value corresponding to each coordinated vehicle" further specifically includes steps S12-1 to S12-4: In step S12-1, for each of the multiple cooperative vehicles, the risk of collision with obstacles or surrounding vehicles after the cooperative vehicle drives with the planned lane change action in a future time period is predicted to obtain a safety score of the cooperative vehicle.

[0101] In this embodiment, after planning the lane change action of each cooperative vehicle, it is necessary to predict the risk of collision between the cooperative vehicle and obstacles or surrounding vehicles after the cooperative vehicle drives in the planned lane change action in the future time period. The collision risk can be calculated by calculating the minimum distance and relative speed between the cooperative vehicle and other vehicles or obstacles after changing lanes, and then evaluating the possibility of collision (i.e., collision risk) based on these minimum distances and relative speeds, and finally obtaining the safety score of the cooperative vehicle according to the collision risk. The relationship between the safety score and the collision risk satisfies: the lower the collision risk, the higher the safety score. In practical applications, the calculation formula for determining the safety score y based on the collision risk x can be: y=-ax+b, where the coefficients a and b can be obtained based on the analysis of multiple real vehicle data. For example: manually annotate the collision risks and corresponding safety scores of multiple vehicles, take the collision risk and corresponding safety score of each vehicle as a data point, and use multiple data points for fitting to obtain the values of coefficients a and b.

[0102] For example, if the cooperative vehicle plans to change lanes to the left, the minimum distance and relative speed between the cooperative vehicle and the vehicle D on the left after the lane change are predicted. If the prediction result shows that the minimum distance is too close or the relative speed is too high, the collision risk will increase. Based on this collision risk assessment, the corresponding safety score of the cooperative vehicle after driving with the planned lane change action (changing lanes to the left) is obtained. The lower the collision risk, the higher the safety score, which means that the lane change action is safer.

[0103] For example, if the minimum distance between the cooperative vehicle and vehicle D after changing lanes remains within a safe range, the safety score may be 0.9. If the minimum distance is too close, the safety score may be 0.3.

[0104] Step S12-2: For each of the multiple cooperative vehicles, predict the speed of the vehicles in each lane after the cooperative vehicle drives with the planned lane change action in the future time period to obtain the efficiency score of the cooperative vehicle.

[0105] In this embodiment, it is also necessary to predict the speed of vehicles in each lane after each cooperative vehicle drives with the planned lane change action in the future time period, so as to evaluate the impact of the cooperative vehicle on the overall traffic flow efficiency after the lane change.

[0106] Specifically, the efficiency score is calculated by calculating the change in the cooperating vehicle's speed after a lane change and its impact on the speeds of surrounding vehicles. For example, if the cooperating vehicle maintains a stable speed after a lane change and has minimal impact on the speeds of other vehicles, the efficiency score will be high. Conversely, if the lane change forces the cooperating vehicle to slow down or causes vehicles behind the cooperating vehicle to slow down as well, the efficiency score will be low.

[0107] For example, if the speed of the cooperative vehicle remains at 60 km / h after changing lanes and has little impact on surrounding vehicles, the efficiency score may be 0.8; if the speed drops to 50 km / h after changing lanes and causes the vehicle behind to slow down, the efficiency score may be 0.4.

[0108] In step S12-3, for each of the multiple cooperative vehicles, the cooperative information between the cooperative vehicle and other vehicles after the cooperative vehicle drives with the planned lane change action in the future time period is predicted to obtain a cooperation score of the cooperative vehicle.

[0109] In this embodiment, the coordination information between each coordinated vehicle and other vehicles after the coordinated vehicle performs the planned lane change in the future time period is also predicted. This includes information such as the relative position, speed, and acceleration between the vehicles.

[0110] Specifically, the smoothness of traffic flow is ensured by evaluating whether the cooperative vehicle can maintain the same relative position (i.e., consistent spacing), consistent speed and acceleration with other vehicles after changing lanes.

[0111] For example, if the cooperative vehicle can maintain the same speed, acceleration, and spacing with other vehicles after changing lanes, the cooperative score will be high. Conversely, if the speed and spacing between the cooperative vehicle and surrounding vehicles differ significantly after changing lanes, the cooperative score will be low.

[0112] Step S12-4: For each of the multiple cooperative vehicles, a reward function value corresponding to the cooperative vehicle is obtained according to the safety score, efficiency score, and cooperation score of the cooperative vehicle.

[0113] In this embodiment, a reward function value corresponding to each cooperating vehicle is calculated based on its safety score, efficiency score, and coordination score. This reward function value is a comprehensive indicator that reflects the impact of the lane change on the overall traffic flow.

[0114] Specifically, the reward function value is calculated based on preset weights (for example, a safety weight of 0.5, an efficiency weight of 0.3, and a cooperation weight of 0.2). For example, if a cooperative vehicle has a safety score of 0.8, an efficiency score of 0.7, and a cooperation score of 0.9, the reward function value calculated for the cooperative vehicle is: 0.5×0.8 + 0.3×0.7 + 0.2×0.9 = 0.8.

[0115] Through the above embodiments, by predicting collision risks and giving safety scores, the safety of each lane change action can be evaluated more accurately, ensuring that all lane change actions entering subsequent evaluations have acceptable safety, thereby effectively avoiding collision accidents. By predicting the vehicle's driving speed and giving an efficiency score, the impact of the lane change on the overall traffic flow efficiency can be evaluated, and the lane change action with the least impact on the traffic flow can be selected to improve the overall traffic efficiency. By predicting collaborative information and giving a collaborative score, it can ensure that the lane change behavior is coordinated with the driving status of other vehicles, avoid traffic chaos caused by the lane change behavior of individual vehicles, and enhance the coordination of multi-vehicle collaborative lane changes. By calculating the reward function value based on the comprehensive safety, efficiency and collaborative scores, the pros and cons of each lane change action can be evaluated more comprehensively, achieving overall optimization of the entire traffic system.

[0116] In combination with the above embodiments, the present application further provides another multi-vehicle coordinated lane change control method. In this method, in addition to the above steps, step S11 of "planning multiple lane change actions for each of the multiple coordinated vehicles in a future time period" specifically includes steps S11-1 to S11-2: Step S11-1: Prioritize the multiple cooperative vehicles according to their vehicle types, speeds, and distances between their current locations and navigation lane change points.

[0117] In this embodiment, before planning the lane change action of each cooperative vehicle, the priority of each cooperative vehicle is first determined based on the type of the cooperative vehicle, the speed of the cooperative vehicle, and the distance between the current position of the cooperative vehicle and the navigation lane change point.

[0118] Specifically, special vehicles (such as ambulances and fire trucks) are given higher priority than ordinary passenger cars. In addition, vehicles with higher speeds or vehicles closer to the navigation lane change point will also receive higher priority.

[0119] For example, the collaborative vehicles include vehicle A and vehicle B. Vehicle A is an ambulance with a speed of 80 km / h and is 10 meters away from the navigation lane change point. Vehicle B is an ordinary passenger car with a speed of 60 km / h and is 50 meters away from the navigation lane change point. In this way, the priority of vehicle A is higher than that of vehicle B. In this way, it can be ensured that when planning lane changes, priority is given to vehicles that are more urgent or closer to the lane change point, thereby improving the efficiency and safety of overall traffic.

[0120] Step S11-2: plan multiple lane changing actions for the cooperative vehicle in the future time period in order from high to low priority.

[0121] Among them, the priority of a cooperative vehicle whose vehicle type is a special vehicle is higher than the priority of a cooperative vehicle whose vehicle type is a non-special vehicle; the higher the speed of the cooperative vehicle, or the closer the distance between the current position point and the navigation lane change point, the higher the priority of the cooperative vehicle.

[0122] In this embodiment, after determining the priority of each cooperating vehicle, multiple lane change maneuvers are planned for each vehicle in the future time period, in descending order of priority. In other words, the highest-priority cooperating vehicle receives the lane change plan first, while lower-priority vehicles make plans based on remaining traffic conditions and resources.

[0123] For example, if vehicle E has the highest priority and vehicle F has the second highest priority, the lane change path for vehicle E will be planned first to ensure it can complete the lane change safely and efficiently. Subsequently, based on vehicle C's lane change path and other traffic conditions, a lane change path is planned for vehicle F, the second-highest priority vehicle, and so on. This ensures orderly traffic flow while maximizing the lane change needs of high-priority vehicles.

[0124] In combination with the above embodiments, the present application also provides another multi-vehicle cooperative lane change control method. In this method, in addition to the above steps, before executing step S15 of "sending control instructions to the multiple cooperative vehicles according to the multi-vehicle cooperative lane change plan with the largest total reward function value", it also includes step S41, and the step S15 specifically includes step S15-1.

[0125] Step S41, with the goal of minimizing the variance of the speeds of the multiple cooperative vehicles during the lane change process and minimizing the total lane change time of the multiple cooperative vehicles, optimize the multi-vehicle cooperative lane change scheme with the maximum total reward function value while satisfying at least one constraint condition to obtain an optimized multi-vehicle cooperative lane change scheme; the at least one constraint condition includes at least one of the following: The shortest lane change time is greater than the first time duration and the longest lane change time is less than the second time duration; The speed parameter values of the multiple cooperative vehicles during the lane change process are within a speed parameter value threshold range; The relative driving parameter values between the multiple cooperative vehicles and the front and rear vehicles during the lane change process are greater than the minimum safe driving parameter value.

[0126] In this embodiment, before executing step S15, it is necessary to further optimize the multi-vehicle cooperative lane changing scheme with the largest total reward function value. The optimization goal is to minimize the speed variance of the cooperative vehicles during the lane changing process and minimize the total lane changing time of the cooperative vehicles to ensure the smoothness and efficiency of the lane changing process.

[0127] At the same time, the optimization process needs to satisfy at least one constraint, which includes: Lane change time constraint: The minimum lane change time is greater than the first duration (for example, 4 seconds), and the maximum lane change time is less than the second duration (for example, 7 seconds). The second duration is greater than the first duration to ensure that the lane change process is neither too fast to cause danger nor too slow to affect traffic efficiency.

[0128] Speed parameter constraint: The speed parameter value of the cooperative vehicle during the lane change process is within the speed parameter value threshold range to ensure that the cooperative vehicle maintains a reasonable speed during the lane change process and avoid traffic problems caused by excessive speed or too slow speed.

[0129] Safe driving parameter value constraint: The relative driving parameter value between the cooperative vehicle and the vehicles in front and behind it during the lane change process is greater than the minimum safe driving parameter value. The relative driving parameter value is the relative distance between the cooperative vehicle and the vehicles in front and behind it, to ensure a safe distance between vehicles during the lane change process and avoid collision risks.

[0130] Based on these constraints, the multi-vehicle cooperative lane-changing scheme with the largest total reward function value is optimized to obtain the optimized multi-vehicle cooperative lane-changing scheme.

[0131] Step S15 of "sending control instructions to the multiple coordinated vehicles according to the multi-vehicle coordinated lane change plan with the maximum total reward function value" specifically includes: Step S15-1: Send control instructions to the multiple cooperative vehicles according to the optimized multi-vehicle cooperative lane change plan.

[0132] In this embodiment, after the optimization is completed, control instructions will be sent to the cooperative vehicles according to the optimized multi-vehicle cooperative lane changing plan. The control instructions will guide the cooperative vehicles on how to change lanes in the future time period to achieve optimal traffic flow.

[0133] Through the above embodiments, by minimizing the speed variance and total lane change time during the lane change process, the efficiency of the lane change process can be ensured, unnecessary deceleration and acceleration of cooperative vehicles during the lane change process can be reduced, and the efficiency of the overall traffic flow can be improved. At the same time, by introducing multiple constraints, the safety of the lane change process can be ensured, thereby achieving comprehensive optimization of the entire traffic system and improving traffic efficiency and safety in autonomous driving scenarios.

[0134] Figure 3 This is a schematic diagram of a multi-vehicle cooperative lane change control device provided by an embodiment of the present application, with reference to Figure 3 Based on the same inventive concept, another embodiment of the present application further provides a multi-vehicle coordinated lane change control device, the device comprising: a lane change action planning module 11, configured to plan, for each of the plurality of cooperating vehicles, a plurality of lane change actions within a future time period, wherein the lane change action is any one of a left lane change, a lane hold, and a right lane change; a lane change action evaluation module 12 for evaluating, for each of the plurality of coordinated vehicles, a plurality of lane change actions of each coordinated vehicle in a future time period, and obtaining a reward function value corresponding to each coordinated vehicle, wherein the reward function value corresponding to each coordinated vehicle is used to represent the impact of the coordinated vehicle's planned lane change action on traffic flow in the future time period; A multi-vehicle coordinated lane change plan generation module 13 is used to traverse various combinations of multiple lane change actions of multiple coordinated vehicles in a future time period and generate multiple multi-vehicle coordinated lane change plans; a total reward function value determination module 14, configured to determine a total reward function value for each multi-vehicle coordinated lane change solution based on the reward function values corresponding to the plurality of coordinated vehicles; The control instruction sending module 15 is used to send control instructions to the multiple cooperative vehicles according to the multi-vehicle cooperative lane change solution with the maximum total reward function value.

[0135] Optionally, the device further comprises at least one of the following: a first filtering module for, after planning a plurality of lane change actions for each of the plurality of cooperative vehicles in a future time period, predicting, for each of the plurality of cooperative vehicles, a risk of collision with an obstacle or a surrounding vehicle after the cooperative vehicle drives in the future time period using the planned lane change action, and filtering out lane change actions with a collision risk greater than a risk threshold from the lane change actions planned for the cooperative vehicle; a second filtering module configured to, after planning for each of the plurality of cooperative vehicles a plurality of lane change actions within a future time period for each of the plurality of cooperative vehicles, determine, for each of the plurality of cooperative vehicles, a time interval between each two adjacent lane change actions when the lane change actions for the cooperative vehicle include at least two lane change actions, and filter out, from the two adjacent lane change actions, lane change actions whose time interval is less than a threshold; a third filtering module for, after planning a plurality of lane change actions for each of the plurality of cooperative vehicles in a future time period, verifying, for each of the plurality of cooperative vehicles, the rationality of the planned lane change action for the cooperative vehicle in the future time period based on lane congestion or navigation information of the cooperative vehicle, and filtering out lane change actions having a rationality less than a rationality threshold from the lane change actions planned for the cooperative vehicle; The multi-vehicle cooperative lane change plan generation module 13 includes: a multi-vehicle cooperative lane change plan generation unit, which is used to traverse various combinations of filtered lane change actions of multiple cooperative vehicles in a future time period to generate multiple multi-vehicle cooperative lane change plans.

[0136] Optionally, the lane change maneuver evaluation module 12 includes: a safety score obtaining unit for predicting, for each of the plurality of cooperative vehicles, the risk of collision with obstacles or surrounding vehicles after the cooperative vehicle performs the planned lane change maneuver in a future time period, and obtaining a safety score for the cooperative vehicle; an efficiency score obtaining unit, configured to predict, for each of the plurality of cooperative vehicles, a speed of vehicles in each lane after the cooperative vehicle performs the planned lane change action in a future time period, and obtain an efficiency score for the cooperative vehicle; a coordination score obtaining unit for predicting, for each of the plurality of coordinated vehicles, coordination information between the coordinated vehicle and other vehicles after the coordinated vehicle performs the planned lane change maneuver in a future time period, and obtaining a coordination score for the coordinated vehicle; The lane change action evaluation unit is used to obtain a reward function value corresponding to each cooperative vehicle among the multiple cooperative vehicles according to the safety score, efficiency score and cooperation score of the cooperative vehicle.

[0137] Optionally, the lane change maneuver planning module 11 includes: a priority sorting unit, configured to prioritize the plurality of cooperative vehicles according to their vehicle types, speeds, and distances between their current locations and navigation lane change points; An execution unit is used to plan multiple lane change actions for the cooperative vehicle in a future time period in descending order of priority; Among them, the priority of a cooperative vehicle whose vehicle type is a special vehicle is higher than the priority of a cooperative vehicle whose vehicle type is a non-special vehicle; the higher the speed of the cooperative vehicle, or the closer the distance between the current position point and the navigation lane change point, the higher the priority of the cooperative vehicle.

[0138] Optionally, the device further comprises: The optimization module is configured to optimize the multi-vehicle coordinated lane change plan with the maximum total reward function value before sending control instructions to the multiple coordinated vehicles according to the multi-vehicle coordinated lane change plan with the maximum total reward function value, with the goal of minimizing the variance of the speeds of the multiple coordinated vehicles during the lane change process and minimizing the total lane change time of the multiple coordinated vehicles, while satisfying at least one constraint condition, to obtain an optimized multi-vehicle coordinated lane change plan; the at least one constraint condition includes at least one of the following: The shortest lane change time is greater than the first time duration and the longest lane change time is less than the second time duration; The speed parameter values of the multiple cooperative vehicles during the lane change process are within a speed parameter value threshold range; During the lane change process, the relative driving parameter values between the multiple cooperative vehicles and the vehicles in front and behind are greater than the minimum safe driving parameter value; The control instruction sending module 15 includes: a control instruction sending unit, which is used to send control instructions to the multiple cooperative vehicles according to the optimized multi-vehicle cooperative lane changing plan.

[0139] Optionally, the device further comprises: A first determining module is used to determine multiple vehicles located in the same lane as the multiple cooperative vehicles, or A second determining module is configured to determine a plurality of vehicles whose mutual distances are within a distance threshold range as the plurality of cooperative vehicles; The third determination module is configured to determine a plurality of vehicles having target key features as the plurality of cooperative vehicles.

[0140] Optionally, the multi-vehicle coordinated lane change plan generating module 13 includes: a level determination unit, configured to determine the level of each of the plurality of cooperative vehicles in the decision tree in descending order of priority, wherein a first cooperative vehicle having the highest priority is located at the first level of the decision tree; a root node determining unit, configured to determine, based on the respective reward function values of the first cooperative vehicle for changing lanes to the left, maintaining the lane, and changing lanes to the right in a future time period, the lane changing action with the largest reward function value as the root node of the decision tree; a second target subnode determining unit for determining, for the second level of the decision tree, a left lane change, a lane maintenance, and a right lane change by a second cooperative vehicle at the second level of the decision tree in a future time period as a subnode of the root node, and, based on a cumulative reward function value of each decision path, determining a subnode on a decision path with a maximum cumulative reward function value as a second target subnode, wherein a decision path is a path from a root node to a subnode of the root node; an nth target subnode determining unit for determining, for the nth level in the decision tree, a cooperative vehicle at the nth level of the decision tree that changes lanes to the left, maintains lanes, and changes lanes to the right in a future time period as a subnode of the n-1th target subnode, respectively; and, based on a cumulative reward function value of each decision path, determining a subnode on a decision path with a maximum cumulative reward function value as the nth target subnode, wherein a decision path is a path from a root node to a subnode of the n-1th target subnode, and n is an integer greater than 2; The multi-vehicle cooperative lane-changing plan determination unit is used to, for the lowest level in the decision tree, regard the cooperative vehicles at the lowest level of the decision tree changing lanes to the left, keeping lanes, and changing lanes to the right in the future time period as a leaf node of the decision tree, respectively, and regard each decision path as a multi-vehicle cooperative lane-changing plan. A decision path is the path from the root node to a leaf node.

[0141] Based on the same inventive concept, another embodiment of the present application also provides an electronic device, including a memory, a processor, and a computer program stored on the memory, wherein the processor executes the computer program to implement the multi-vehicle cooperative lane change control method as described in any of the above embodiments.

[0142] Among them, electronic equipment refers to Figure 4 , Figure 4 Schematic diagram of an electronic device provided by an embodiment of the present application. Figure 4 As shown, the electronic device 400 includes: a memory 410 and a processor 420. The memory 410 and the processor 420 are connected via a bus communication. A computer program is stored in the memory 410. The computer program can be run on the processor 420 to implement the steps in the multi-vehicle cooperative lane change control method disclosed in the above embodiment of the present application.

[0143] Based on the same inventive concept, another embodiment of the present application further provides a computer program product, including a computer program, which is executed by a processor to implement the multi-vehicle cooperative lane change control method as described in any of the above embodiments.

[0144] Based on the same inventive concept, another embodiment of the present application further provides a computer-readable storage medium on which a computer program is stored, wherein when the program is executed by a processor, the multi-vehicle cooperative lane change control method as described in any of the above embodiments is implemented.

[0145] As for the device, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0146] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0147] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, devices, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0148] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0149] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0150] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0151] Although preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they become aware of the basic inventive concepts. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present invention.

[0152] Finally, it should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or terminal device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or terminal device that includes the element.

[0153] The above is a detailed introduction to the multi-vehicle cooperative lane change control method, device, equipment and medium provided by the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for general technical personnel in this field, based on the ideas of the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A multi-vehicle coordinated lane change control method, characterized in that: The method comprises: For each of the plurality of cooperating vehicles, planning a plurality of lane changing actions for each cooperating vehicle in a future time period, wherein the lane changing action is any one of changing lanes to the left, maintaining the lane, and changing lanes to the right; For each of the multiple cooperative vehicles, multiple lane change actions of each cooperative vehicle in a future time period are evaluated to obtain a reward function value corresponding to each cooperative vehicle, wherein the reward function value corresponding to the cooperative vehicle is used to represent: the impact of the cooperative vehicle's driving in the future time period using the planned lane change action on the traffic flow in the future time period; Traverse various combinations of multiple lane change actions of multiple cooperative vehicles in the future time period to generate multiple multi-vehicle cooperative lane change plans; Determining a total reward function value for each multi-vehicle coordinated lane change solution based on the reward function values corresponding to the multiple coordinated vehicles; According to the multi-vehicle cooperative lane changing plan with the largest total reward function value, control instructions are sent to the multiple cooperative vehicles.

2. The multi-vehicle coordinated lane change control method according to claim 1, characterized in that: After planning multiple lane changing actions for each of the multiple cooperative vehicles in a future time period, the method further includes at least one of the following: For each of the multiple cooperative vehicles, predict the risk of collision with obstacles or surrounding vehicles after the cooperative vehicle performs the planned lane change maneuver in a future time period, and filter out lane change maneuvers with a collision risk greater than a risk threshold from the lane change maneuvers planned for the cooperative vehicle; For each of the multiple cooperating vehicles, when the lane change action for the cooperating vehicle includes at least two lane change actions, determining a time interval between each two adjacent lane change actions, and filtering out lane change actions with a time interval less than a threshold from the two adjacent lane change actions; For each of the multiple cooperating vehicles, based on lane congestion or navigation information of the cooperating vehicle, verifying the rationality of the planned lane change maneuvers for the cooperating vehicle in a future time period, and filtering out lane change maneuvers with rationality less than a rationality threshold from the lane change maneuvers planned for the cooperating vehicle; Traverse all combinations of multiple lane change actions of multiple coordinated vehicles in the future time period and generate multiple multi-vehicle coordinated lane change plans, including: Traverse each combination of filtered lane change actions of multiple cooperative vehicles in the future time period to generate multiple multi-vehicle cooperative lane change plans.

3. The multi-vehicle coordinated lane change control method according to claim 1, characterized in that: For each of the multiple cooperative vehicles, multiple lane change actions of each cooperative vehicle in the future time period are evaluated to obtain the reward function value corresponding to each cooperative vehicle, including: For each of the multiple cooperative vehicles, predict the risk of collision with obstacles or surrounding vehicles after the cooperative vehicle uses the planned lane change maneuver in the future time period, and obtain a safety score for the cooperative vehicle; For each of the multiple cooperative vehicles, predict the speed of vehicles in each lane after the cooperative vehicle uses the planned lane change action in the future time period to obtain an efficiency score for the cooperative vehicle; For each of the multiple cooperative vehicles, predict the cooperative information between the cooperative vehicle and other vehicles after the cooperative vehicle drives with the planned lane change maneuver in the future time period, and obtain a cooperative degree score for the cooperative vehicle; For each of the multiple cooperative vehicles, a reward function value corresponding to the cooperative vehicle is obtained according to the safety score, efficiency score and cooperation score of the cooperative vehicle.

4. The multi-vehicle coordinated lane change control method according to claim 1, characterized in that: For each of the multiple cooperative vehicles, multiple lane change actions are planned for each cooperative vehicle in the future time period, including: Prioritizing the multiple cooperative vehicles according to their vehicle types, speeds, and distances between their current locations and navigation lane change points; Plan multiple lane change actions for the cooperative vehicle in the future time period in descending order of priority; Among them, the priority of a cooperative vehicle whose vehicle type is a special vehicle is higher than the priority of a cooperative vehicle whose vehicle type is a non-special vehicle; the higher the speed of the cooperative vehicle, or the closer the distance between the current position point and the navigation lane change point, the higher the priority of the cooperative vehicle.

5. The multi-vehicle coordinated lane change control method according to claim 4, characterized in that: Traverse all combinations of multiple lane change actions of multiple coordinated vehicles in the future time period and generate multiple multi-vehicle coordinated lane change plans, including: Determining the levels of the plurality of cooperative vehicles in a decision tree in descending order of priority, wherein a first cooperative vehicle with the highest priority is located at a first level of the decision tree; According to the reward function values of the first cooperative vehicle for changing lanes to the left, keeping lane, and changing lanes to the right in the future time period, the lane changing action with the largest reward function value is determined as the root node of the decision tree; For the second level of the decision tree, the second cooperative vehicle at the second level of the decision tree changes lanes to the left, maintains lanes, and changes lanes to the right in the future time period, respectively, as a child node of the root node. Based on the cumulative reward function value of each decision path, the child node on the decision path with the largest cumulative reward function value is selected as the second target child node. A decision path is a path from a root node to a child node of the root node. For the nth level in the decision tree, the cooperative vehicles at the nth level of the decision tree that change lanes to the left, keep lanes, and change lanes to the right in the future time period are respectively used as a child node of the n-1th target child node. Based on the cumulative reward function value of each decision path, the child node on the decision path with the largest cumulative reward function value is used as the nth target child node. A decision path is a path from a root node to a child node of the n-1th target child node, where n is an integer greater than 2. For the lowest level in the decision tree, the cooperative vehicles at the lowest level of the decision tree that change lanes to the left, keep lanes, and change lanes to the right in the future time period are respectively regarded as leaf nodes of the decision tree, and each decision path is regarded as a multi-vehicle cooperative lane change plan. A decision path is the path from the root node to a leaf node.

6. The multi-vehicle coordinated lane change control method according to claim 1, characterized in that: Before sending control instructions to the multiple cooperative vehicles according to the multi-vehicle cooperative lane change solution with the maximum total reward function value, the method further includes: With the goal of minimizing the variance of the speeds of the multiple cooperative vehicles during the lane change process and minimizing the total lane change time of the multiple cooperative vehicles, the multi-vehicle cooperative lane change scheme with the maximum total reward function value is optimized while satisfying at least one constraint condition to obtain an optimized multi-vehicle cooperative lane change scheme; the at least one constraint condition includes at least one of the following: The shortest lane change time is greater than the first time duration and the longest lane change time is less than the second time duration; The speed parameter values of the multiple cooperative vehicles during the lane change process are within a speed parameter value threshold range; During the lane change process, the relative driving parameter values between the multiple cooperative vehicles and the vehicles in front and behind are greater than the minimum safe driving parameter value; According to the multi-vehicle coordinated lane change solution with the maximum total reward function value, control instructions are sent to the multiple coordinated vehicles, including: According to the optimized multi-vehicle cooperative lane changing plan, control instructions are sent to the multiple cooperative vehicles.

7. The multi-vehicle coordinated lane change control method according to any one of claims 1 to 6, characterized in that: The method further comprises: Determine multiple vehicles in the same lane as the multiple cooperative vehicles, or Determine a plurality of vehicles whose mutual distances are within a distance threshold range as the plurality of cooperative vehicles; A plurality of vehicles having target key features are determined as the plurality of cooperative vehicles.

8. A multi-vehicle coordinated lane change control device, characterized in that: The device comprises: a lane change action planning module, configured to plan, for each of the plurality of cooperating vehicles, a plurality of lane change actions within a future time period, wherein the lane change action is any one of a left lane change, a lane hold, and a right lane change; a lane change action evaluation module for evaluating, for each of the plurality of coordinated vehicles, a plurality of lane change actions of each coordinated vehicle in a future time period, and obtaining a reward function value corresponding to each coordinated vehicle, wherein the reward function value corresponding to each coordinated vehicle is used to represent: the impact of the coordinated vehicle's driving in the future time period using the planned lane change action on the traffic flow in the future time period; A multi-vehicle coordinated lane change plan generation module is used to traverse various combinations of multiple lane change actions of multiple coordinated vehicles in the future time period and generate multiple multi-vehicle coordinated lane change plans; a total reward function value determination module, configured to determine a total reward function value for each multi-vehicle coordinated lane change solution based on the reward function values corresponding to the plurality of coordinated vehicles; A control instruction sending module is used to send control instructions to the multiple cooperative vehicles according to the multi-vehicle cooperative lane change solution with the maximum total reward function value.

9. An electronic device, characterized in that: It includes a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the multi-vehicle cooperative lane change control method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that A computer program is stored thereon, wherein when the computer program is executed by a processor, the multi-vehicle cooperative lane change control method as described in any one of claims 1-7 is implemented.

Images