A cloud-edge-end architecture-based centralized cooperative lane-changing control method and system for mixed traffic under an accident bottleneck and a medium

Abstract

The application discloses a centralized cooperative lane-changing control method and system for mixed traffic groups in an accident bottleneck scene based on a cloud-edge-end architecture and a medium, and belongs to the technical field of cooperative control of Internet of Vehicles. The application divides the mixed traffic groups into sub-traffic groups by taking the networked automatic vehicle as a head vehicle, and pre-matches a target lane-changing gap for each sub-traffic group. A multi-gap search mechanism of geometric gap constraint, collision time constraint and comprehensive evaluation function is established to determine the target gap of each sub-traffic group. The longitudinal sliding mode control, the transverse cooperative control and the cooperative gap expansion control are implemented in stages to enable the sub-traffic groups to change lanes safely and orderly. After lane changing, the vehicle speed and the inter-vehicle distance are coordinated through the traffic group consistency control to enable the traffic group to quickly restore stable operation. Relying on the cooperative processing of the cloud control platform and the edge server, a perception, decision and control closed loop is formed to realize efficient and safe lane changing of the mixed traffic groups in the accident bottleneck area and stable traffic of the traffic groups, improve the traffic efficiency, reduce traffic disturbance and ensure driving safety.

Description

Technical Field

[0001] This invention belongs to the field of intelligent transportation system technology, and relates to a collaborative control method for mixed traffic groups, and particularly to a centralized collaborative lane-changing control method, system and medium for mixed traffic groups based on a cloud-edge-device architecture under accident bottleneck scenarios. Background Technology

[0002] Accident-induced traffic bottlenecks refer to road areas where traffic flow is restricted, vehicle speeds are reduced, and traffic congestion is dense due to traffic accidents, road closures, or other unforeseen events that reduce the number of lanes or road capacity. Within these areas, mixed traffic groups (connected autonomous vehicles and connected human-driven vehicles) must perform lane-changing and platoon reorganization operations under limited space and time constraints. Accident-induced traffic bottleneck areas typically include: Bottleneck entrance: the upstream area before lane reduction, where vehicles begin queuing or adjusting their driving status; Bottleneck section: the core area where the number of lanes is reduced or traffic capacity is limited; Bottleneck exit: the downstream area where vehicles return to normal lane numbers. Characteristics include: significant spatial constraints (insufficient number of lanes or vehicle width); time sensitivity (vehicles must complete safe lane changes within a limited time); and applicability to mixed traffic groups (including mixed flows of autonomous vehicles and human-driven vehicles).

[0003] With the rapid development of intelligent transportation systems, the deep integration of vehicle networking and wireless communication technologies is driving the research and demonstration applications of Connected Autonomous Vehicles (CAVs), providing new technological support for addressing complex traffic scenarios. Meanwhile, Connected Human-driven Vehicles (CHVs), which possess communication capabilities but still rely on human drivers, will continue to exist during the transition period. It is anticipated that in the coming decades, CAVs and CHVs will mix to form fleets, creating mixed-traffic groups. In such mixed-traffic groups, vehicles generally face challenges of information asymmetry and varying levels of intelligence, increasing the complexity of inter-vehicle interactions and raising the uncertainty of traffic operations, thus severely restricting road efficiency and increasing safety risks. These challenges are particularly pronounced in bottleneck sections caused by traffic accidents, where lane closures force mixed-traffic groups to perform mandatory lane-changing operations under strict time and safety constraints.

[0004] However, existing technologies still have significant shortcomings: on the one hand, they lack a phased longitudinal control mechanism that integrates target search and nonlinear following dynamics; on the other hand, a centralized optimal spacing search scheme for orderly lane changing of multiple sub-vehicle groups has not yet been proposed. Existing research on accident bottleneck control mostly adopts passive strategies, lacking active spacing selection and cluster-based queue reorganization techniques. Therefore, developing centralized collaborative control strategies for accident bottleneck scenarios has become a key research focus.

[0005] To address the aforementioned shortcomings, the deep integration of edge computing and cloud computing technologies has provided a new technological paradigm for intelligent transportation systems in recent years. In complex traffic scenarios, a closed loop of perception, decision-making, and control can be constructed through intelligent roadside units, in-vehicle intelligent terminals, and edge computing nodes. Although Vehicle-to-Everything (V2X) communication technology provides connected autonomous vehicles with low-latency information sharing capabilities, and cloud-edge-device collaborative mechanisms have been applied to traffic prediction and collaborative decision-making research, there is still a lack of comprehensive solutions integrating cloud-edge-device-based vehicle subgroup segmentation, edge-based spacing search, and vehicle-side execution mechanisms. This is crucial for improving the safety and efficiency of cooperative lane changing in mixed vehicle groups. Furthermore, connected autonomous vehicles differ substantially from traditional human-driven vehicles in terms of control and decision-making. These heterogeneous characteristics not only exacerbate the nonlinear coupling complexity of coordinated control of heterogeneous vehicle groups in single-lane scenarios but also pose core challenges to the consistency, stability, and driving efficiency of vehicle group operation. Therefore, research on comprehensive solutions for cloud-edge-device-based vehicle subgroup segmentation, edge-based spacing search, and vehicle-side execution mechanisms has significant practical implications. Summary of the Invention

[0006] In view of this, the purpose of this invention is to provide a centralized collaborative lane-changing control method for mixed traffic groups based on a cloud-edge-device architecture in accident bottleneck scenarios. By constructing a sub-group division mechanism for mixed traffic groups, a multi-gap search and matching mechanism, and a phased collaborative lane-changing control strategy, the method achieves safe and orderly lane changing of mixed traffic groups within the accident bottleneck area and restores the consistency of the traffic group after lane changing, thereby improving vehicle traffic efficiency and operational safety.

[0007] To achieve the above objectives, the present invention provides the following technical solution: A centralized cooperative lane-changing control method for hybrid vehicle clusters based on a cloud-edge-device architecture under accident bottleneck conditions, executed by a cloud control platform including a processor and non-transitory memory, the method includes the following steps: S0. Collect vehicle operation status information of mixed traffic groups in the accident bottleneck area through roadside sensing devices and upload it to the cloud control platform. The vehicle operation status information includes at least vehicle position, speed, acceleration, heading angle, vehicle length, vehicle type and lane information. S1. The cloud control platform constructs a collaborative lane-changing operation mechanism for mixed traffic groups based on a cloud-edge-device architecture under accident bottleneck scenarios, and establishes rules for dividing mixed traffic sub-groups to achieve collaborative organization and management of mixed traffic groups. S2. The cloud control platform acquires the vehicle operation status information of the mixed traffic group, constructs a candidate lane-changing gap set, and establishes a target lane-changing gap search and matching mechanism to determine the target lane-changing gap for each sub-group of vehicles. S3. The cloud control platform establishes a vehicle kinematic model based on the vehicle's kinematic characteristics and constructs a phased collaborative lane-changing control strategy to perform lane-changing trigger control, lane-changing execution control, and collaborative gap-expanding control on the sub-vehicle group to complete orderly lane changing. S4. The cloud control platform obtains the vehicle group status information after the lane change is completed, establishes a vehicle group consistency control mechanism, and coordinates and controls the speed and spacing of the mixed vehicle group so that the vehicle group can quickly return to a stable operating state after the lane change is completed. S5. The cloud control platform sends control commands to the lead vehicle of the target sub-vehicle group via the vehicle network communication channel, so that the vehicle power system of the lead vehicle performs lateral and longitudinal movement adjustments.

[0008] Furthermore, the rules for dividing mixed-traffic sub-groups in S1 include: The connected autonomous vehicle is designated as the lead vehicle of the sub-vehicle group, and the connected human-driven vehicles that follow it are classified as members of the sub-vehicle group until the next connected autonomous vehicle appears. When there are no connected human drivers following behind a connected autonomous vehicle, the connected autonomous vehicle constitutes a separate sub-group of vehicles.

[0009] Furthermore, the construction of the candidate lane change gap set in S2 includes: The longitudinal clearance between any two adjacent vehicles in the target lane is used as the candidate lane-changing clearance, and all candidate clearances are combined to form a candidate clearance set. Represented as:

[0010] in, The subscript indicates the corresponding candidate lane change gap number. This represents the candidate lane change gap between the first and second vehicles. The total number of vehicles in the target lane; The target lane change clearance matching in S2 is achieved through geometric clearance constraints, which satisfy the following:

[0011] in, For the group of vehicles Medium target candidate gap The position of the following vehicle at the moment of lane change; t The current moment; For the group of vehicles Interval between arrivals of the middle car Time required to change lanes; To ensure a safe distance or reserve a buffer distance; For the group of vehicles The length of the lead car; Sub-cart group The position of the lead car at the moment of lane change; Sub-cart group Medium target candidate gap The position of the vehicle in front at the moment of lane change; For the group of vehicles Intermediate gap The length of the vehicle in front; The geometric clearance constraint ensures that the lead car in the sub-car group has sufficient longitudinal space to complete the lane change operation when it reaches the candidate lane change gap position, preventing the lead car in the sub-car group from colliding with the vehicles in front and behind.

[0012] Furthermore, the target lane change gap matching in S2 also includes a collision time constraint, which satisfies: Forward collision time constraint:

[0013] in, For the group of vehicles Middle car and gap The time of the collision with the vehicle in front; , Position and speed of the lead car in the sub-car group; , The candidate gap is determined by the position and speed of the vehicle in front. Rear-end collision time constraint:

[0014] in, For the group of vehicles Middle car and gap The time of the collision with the following vehicle; , The position and speed of the vehicle following the candidate gap; Rear-end collision time constraint: To ensure lane-changing safety, the collision time between the front and rear vehicles must be greater than the safe headway. :

[0015] The collision time constraint ensures that the lead car in the sub-vehicle group and the vehicles in front and behind maintain time safety during lane changing. When both the collision time of the lead car and the collision time of the rear car are greater than the safe headway, the time safety condition for lane changing operation is confirmed to be met.

[0016] Furthermore, the target lane change gap matching in S2 also includes a comprehensive evaluation function for candidate gaps:

[0017] in, , , Let be the weight coefficient, and satisfy... ; To determine the execution cost, select The smallest candidate gap is taken as the target lane change gap; The comprehensive evaluation function quantifies and evaluates candidate gaps that meet geometric gap constraints and collision time constraints, and selects the candidate gap with the lowest execution cost as the target lane-changing gap, thereby reducing vehicle lane-changing conflicts and traffic disturbances.

[0018] Furthermore, the lane change trigger control in S3 includes: The cloud control platform generates longitudinal control commands based on the position of the target lane change gap and the position, speed, and acceleration differences between the lead car of the sub-car group and the cars following the target lane change gap, and sends them to the lead car of the sub-car group. The lead vehicle of the sub-vehicle group adjusts its longitudinal movement state according to the longitudinal control command, gradually approaching the lane change trigger position corresponding to the target lane change gap; Once the lead car in the sub-car group completes the lane change, the vehicle immediately following in the original sub-car group automatically becomes the new lead car and continues to receive longitudinal control commands from the cloud control platform until all vehicles in the sub-car group have completed the lane change.

[0019] Furthermore, the lane change execution control in S3 includes: When the lead car of the sub-vehicle group meets the preset lane-changing trigger condition, the lead car of the sub-vehicle group is controlled to move laterally towards the target lane; The lateral control generates steering control commands based on the vehicle's lateral position deviation and heading angle deviation to control the vehicle to gradually converge toward the center line of the target lane. Once the lead vehicle of the sub-vehicle group completes its lane change, subsequent vehicles will perform lane changes in a preset order to allow the entire sub-vehicle group to merge into the target lane.

[0020] Furthermore, the cooperative clearance control in S3 and the group consistency control in S4 include: Once the lead vehicle in the group of vehicles enters the target lane, it reduces its speed to increase the distance between itself and the vehicle in front, providing space for subsequent vehicles to change lanes. When multiple sub-car groups perform lane changes simultaneously, the gap widening control intensity corresponding to each sub-car group is determined based on the sorting result of the target lane change gap. After all vehicles have completed lane changing, the position and speed information of each vehicle in the mixed traffic group is obtained, and the spacing and speed difference between adjacent vehicles are coordinated and controlled so that the actual spacing between adjacent vehicles approaches the preset spacing and the speed difference between adjacent vehicles approaches zero, thereby restoring the mixed traffic group to a consistent operating state.

[0021] A centralized cooperative lane-changing control system for hybrid vehicle clusters based on a cloud-edge-device architecture under accident bottleneck conditions includes: The cloud control platform includes a processor and a non-transitory memory, wherein the non-transitory memory stores a computer program, and the processor executes the computer program for: Receive vehicle operating status information uploaded by roadside sensing devices; Construct a mixed-traffic sub-vehicle group division rule, based on the vehicle operation status information, and determine the target lane change gap for each sub-vehicle group according to the division rule; Establish a phased collaborative lane-changing control strategy to generate longitudinal and lateral control commands; Establish a vehicle group consistency control mechanism to coordinate vehicle group speed and spacing; Roadside sensing equipment, including at least one of roadside unit, millimeter-wave radar, lidar or video camera, is used to collect vehicle operating status information of mixed traffic vehicles; The vehicle-to-everything (V2X) communication module is used to establish a communication connection between the cloud control platform and the mixed-traffic vehicles, and to send the control commands to each vehicle so that the vehicle can perform lateral and longitudinal movement adjustments.

[0022] A computer-readable storage medium storing a computer program thereon, characterized in that, when the computer program is executed by a processor, it implements the above-described method, wherein the cloud control platform acquires vehicle operating status information through roadside sensing devices and sends control commands to the vehicle through a vehicle-to-everything (V2X) communication channel, causing the vehicle's power system, braking system, and / or steering system to physically execute corresponding control commands to complete lateral and longitudinal motion adjustments.

[0023] The beneficial effects of this invention are as follows: A mixed-traffic sub-vehicle grouping mechanism with connected autonomous vehicles as the lead vehicle is constructed. This mechanism divides large-scale mixed-traffic vehicle groups into multiple sub-vehicle groups that can be independently and collaboratively controlled, thereby achieving orderly organization and collaborative management of mixed-traffic vehicles, reducing the complexity of vehicle interaction, and improving the efficiency of lane-changing organization in accident bottleneck areas.

[0024] By establishing a multi-gap search mechanism that includes geometric gap constraints, collision time constraints, and a comprehensive evaluation function, the optimal target lane-changing gap can be matched for each mixed-traffic subgroup from multiple candidate lane-changing gaps, thereby improving the rationality and safety of lane-changing decisions and reducing vehicle lane-changing conflicts and traffic disturbances.

[0025] A phased collaborative lane-changing control strategy is constructed, which combines longitudinal sliding mode control in the lane-changing triggering stage with lateral and longitudinal collaborative control in the lane-changing execution stage. This strategy enables vehicles to quickly converge to the target position and target lane, improving the stability, robustness, and control accuracy of the lane-changing process.

[0026] The sub-vehicle group collaborative gap expansion strategy is set up to actively adjust the target gap space after the lane change is completed, so as to provide safe lane change conditions for subsequent vehicles and improve the traffic efficiency and lane change success rate in the scenario of continuous lane change of multiple sub-vehicle groups.

[0027] Relying on the cloud-edge-device architecture, the system enables coordinated operation of vehicle status perception, gap search, collaborative decision-making, and control execution. Combined with a vehicle group consistency control mechanism, it allows the vehicle group speed and spacing to quickly return to a stable state after lane changing, effectively improving the traffic efficiency, safety, and operational stability of mixed-traffic groups in accident bottleneck areas.

[0028] Other advantages, objectives, and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination, or may be learned from practice of the invention. The objectives and other advantages of the invention can be realized and obtained through the following description. Attached Figure Description

[0029] To make the objectives, technical solutions, and advantages of the present invention clearer, the preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings, wherein: Figure 1 Flowchart of a centralized collaborative lane-changing control method for mixed-traffic vehicle clusters based on a cloud-edge-device architecture in an accident bottleneck scenario provided by an embodiment of the present invention; Figure 2 This is a schematic diagram of mixed traffic in an accident bottleneck scenario provided by an embodiment of the present invention; Figure 3 This is a schematic diagram of the subgroup division of mixed traffic vehicles provided in an embodiment of the present invention; Figure 4 This is a schematic diagram illustrating the collaborative lane-changing operation mechanism of a mixed-traffic vehicle group under a cloud-edge-device architecture provided in an embodiment of the present invention. Figure 5 This is a schematic diagram of a cooperative lane-changing control strategy for mixed traffic groups provided in an embodiment of the present invention. Detailed Implementation

[0030] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0031] The accompanying drawings are for illustrative purposes only and are schematic diagrams, not actual pictures. They should not be construed as limiting the invention. To better illustrate the embodiments of the invention, some parts in the drawings may be omitted, enlarged, or reduced, and do not represent the actual product dimensions. It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.

[0032] In the accompanying drawings of the embodiments of the present invention, the same or similar reference numerals correspond to the same or similar components. In the description of the present invention, it should be understood that if terms such as "upper," "lower," "left," "right," "front," and "rear" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, they are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, the terms used to describe positional relationships in the drawings are only for illustrative purposes and should not be construed as limiting the present invention. For those skilled in the art, the specific meaning of the above terms can be understood according to the specific circumstances.

[0033] Please see Figure 1 The flowchart below illustrates a centralized collaborative lane-changing control method for mixed-traffic vehicle clusters based on a cloud-edge-device architecture in accident bottleneck scenarios, as provided in this embodiment of the invention. The method includes the following steps: S0 collects vehicle operation status information of mixed traffic groups in the accident bottleneck area through roadside sensing devices and uploads the vehicle operation status information to the cloud control platform; The roadside sensing equipment may include one or more of the following: roadside unit (RSU), millimeter-wave radar, lidar, and video camera; The vehicle operating status information includes at least the vehicle position, vehicle speed, vehicle acceleration, vehicle heading angle, vehicle length, vehicle type, and lane information.

[0034] The S1 cloud control platform constructs a collaborative lane-changing operation mechanism for mixed traffic groups based on a cloud-edge-device architecture in accident bottleneck scenarios. It also establishes rules for dividing mixed traffic sub-groups with connected autonomous vehicles as the lead vehicles based on the composition of mixed traffic vehicles, thereby realizing the collaborative organization and management of mixed traffic groups.

[0035] Based on the vehicle operation status information of the mixed traffic group, the S2 cloud control platform constructs a candidate lane change gap set and establishes a target lane change gap search and matching mechanism that includes geometric gap constraints, collision time constraints and comprehensive evaluation functions to determine the target lane change gap for each mixed traffic subgroup.

[0036] The S3 cloud control platform constructs a vehicle motion model based on the vehicle's kinematic characteristics and establishes a phased collaborative lane-changing control strategy. It performs lane-changing trigger control, lane-changing execution control, and post-lane-changing collaborative gap-expansion control on mixed-traffic sub-groups to achieve orderly lane changing for each mixed-traffic sub-group.

[0037] The S4 cloud control platform acquires the operational status information of the mixed traffic group after lane changing, establishes a vehicle group consistency control mechanism, and coordinates the vehicle speed and spacing of the mixed traffic group to enable the mixed traffic group to quickly restore a consistent operational status and stably pass through the accident bottleneck area after lane changing.

[0038] S5. The cloud control platform generates longitudinal control commands and lateral control commands based on lane change trigger control, lane change execution control, cooperative gap widening control in S3 and vehicle group consistency control in S4, and sends them to the on-board control unit of the target sub-vehicle group leader vehicle via the vehicle-to-everything (V2X) communication channel. The vehicle control unit controls the vehicle's power system, braking system, and steering system to perform corresponding longitudinal acceleration / deceleration control and lateral steering control according to the longitudinal control command and the lateral control command. After the vehicle is controlled, the roadside sensing equipment continuously collects vehicle operation status information and feeds it back to the cloud control platform to update vehicle position, speed, acceleration and vehicle spacing status parameters. The cloud control platform dynamically updates control parameters based on the feedback of vehicle operating status information, and sends the updated control commands back to the vehicle execution terminal, forming a closed-loop collaborative control process of vehicle status perception, collaborative decision-making, control execution and status feedback.

[0039] The steps described above will be explained in detail below with reference to the accompanying drawings.

[0040] As one implementation method, S1 includes the following steps: S11. Constructing a mixed-traffic environment under accident bottleneck scenarios. Please see Figure 2 This is a schematic diagram of a mixed traffic scenario involving an accident bottleneck, provided in an embodiment of the present invention; a two-way, two-lane mixed traffic scenario involving an accident bottleneck is set up. From Figure 2 As can be seen, the road consists of two parts, upper and lower, corresponding to two opposing directions of travel, each containing two lanes (lane 0 and lane 1). A bottleneck section, caused by a traffic accident, is placed on the right side of lane 1 in one of the travel directions, such as the eastbound direction, to simulate the reduction in the number of lanes due to an accident. Its length is... Due to this bottleneck, vehicles originally traveling in lane 1 must merge into lane 0 when approaching the bottleneck area to continue normal traffic flow. If vehicles change lanes haphazardly upstream of the bottleneck, it will significantly increase the propagation of traffic disturbances, leading to a decrease in traffic capacity near the bottleneck area and even inducing severe congestion. Therefore, coordinated organization and control of mixed-traffic vehicles are necessary.

[0041] S12. Establish a mechanism for dividing mixed traffic vehicle groups into subgroups. Please see Figure 3 This is a schematic diagram of subgroup division for mixed-traffic vehicles provided in an embodiment of the present invention. Under the cloud-edge-device architecture, to achieve orderly organization and coordinated control of mixed-traffic vehicles and ensure that the mixed-traffic vehicles in the upstream lanes of bottleneck areas can adjust their driving status in an orderly manner, it is necessary to divide the mixed-traffic vehicles into subgroups. Specifically, the following rules for dividing mixed-traffic vehicles into subgroups are given: The connected autonomous vehicle is used as the lead vehicle of the sub-vehicle group, and all connected human-driven vehicles that follow it are classified as members of the sub-vehicle group until the next connected autonomous vehicle appears. 1) When there are no connected human drivers following behind a connected autonomous vehicle, the connected autonomous vehicle constitutes a separate sub-group of vehicles.

[0042] 2) Based on the above rules, the mixed-traffic vehicle group can be divided into several mixed-traffic sub-groups, with connected automated vehicles as the lead vehicles and connected human-driven vehicles as the following vehicles, and the mixed-traffic sub-groups serve as the basic execution unit for subsequent coordinated lane-changing control.

[0043] S13. Construct a collaborative lane-changing operation mechanism for mixed-vehicle groups under a cloud-edge-device architecture. Please see Figure 4This is a schematic diagram illustrating the collaborative lane-changing mechanism of mixed-traffic vehicle groups under the cloud-edge-device architecture provided in this embodiment of the invention. The edge cloud server acquires raw vehicle operation data through road testing equipment and uploads vehicle status information to the cloud control platform. The operation status information includes at least vehicle position, vehicle speed, vehicle acceleration, vehicle heading angle, vehicle length, vehicle type, and lane information. Based on the collaborative lane-changing control strategy and road area division principles, the cloud control platform functionally partitions the bottleneck section into normal driving areas and collaborative lane-changing areas. Subsequently, the cloud control platform performs sub-vehicle group division for vehicles within the collaborative lane-changing area and sends the area division results and mixed-traffic sub-vehicle group division results to the edge cloud server. Within the collaborative lane-changing area, the edge cloud server, based on the data collected and fused by the road testing equipment and according to the deployed collaborative lane-changing strategy, performs target gap search for each mixed-traffic sub-vehicle group in lane 1, calculates key decision parameters such as lane-changing timing for each vehicle, and feeds them back to the cloud control platform. The cloud control platform generates control commands based on the decision parameters and issues them to mixed-traffic vehicles to achieve longitudinal control during the approach to the target gap, lateral and longitudinal coordinated control during the lane-changing execution phase, longitudinal control after the lane-changing is completed, and vehicle group coordinated control. This ensures that each mixed-traffic sub-group can safely, orderly and efficiently merge into the lane and quickly pass through the accident bottleneck area.

[0044] In this embodiment: RSU / radar / camera installation location: such as on a roadside pole within 200 to 500 meters upstream of the accident bottleneck area.

[0045] Communication protocol: C-V2X PC5 interface (direct communication) or Uu interface (cellular communication), end-to-end latency ≤100ms.

[0046] Control frequency: The cloud control platform control cycle is 10Hz-20Hz, and the edge server sensing data fusion frequency is ≥20Hz.

[0047] Vehicle actuator: The onboard unit (OBU) of the lead vehicle is connected to the vehicle powertrain via a CAN bus.

[0048] Please see Figure 5 This is a schematic diagram of the cooperative lane-changing control strategy for mixed-traffic vehicle groups provided in an embodiment of the present invention. The system utilizes edge servers and road test equipment to collect vehicle status information of the mixed-traffic vehicle groups, establishes a multi-gap search mechanism, and matches the optimal target lane-changing gap for each mixed-traffic subgroup through geometric gap constraints, collision time constraints, and a comprehensive evaluation function.

[0049] As one implementation method, S2 includes the following steps: S21. Construct a candidate lane change gap set The roadside equipment collects information on the operational status of mixed-traffic vehicles within the accident bottleneck area. This vehicle operational status information includes at least vehicle position, vehicle speed, vehicle acceleration, vehicle heading angle, vehicle length, vehicle type, and lane information. An edge server performs fusion processing on the collected data. The roadside equipment can be one or more combinations of roadside units (RSUs), millimeter-wave radar, lidar, and video cameras.

[0050] Please see Figure 5 In Step 1, when a group of mixed vehicles in lane 1 performs coordinated lane changing as a sub-group, the corresponding merging position needs to be determined in the target lane (lane 0). Therefore, in this embodiment, the target lane changing position of the sub-group is defined as the target lane changing gap. In the target lane (lane 0), the longitudinal clearance between any two adjacent vehicles can be used as a candidate lane-changing clearance. All candidate lane-changing clearances constitute a candidate clearance set. Represented as:

[0051] in, The subscript indicates the corresponding candidate lane change gap number. In this embodiment, it represents the candidate lane change gap between the first and second vehicles in lane 0. The total number of vehicles in lane 0 of the coordinated lane-changing area.

[0052] To fully utilize the multiple small-scale gaps in target lane 0 and support the rapid and reliable lane changing of mixed-traffic vehicles, this embodiment proposes a gap search and selection method oriented towards sub-vehicle groups. When the mixed-traffic vehicle group enters the cooperative lane-changing area, the edge cloud server uses collected vehicle operating status information to select a set of candidate gaps. The search and filtering process is performed to select the most suitable candidate gap for the lead car of each subgroup and determine it as the target lane change gap.

[0053] The multi-gap search mechanism established in this embodiment includes three types of constraints and evaluations: Geometric clearance constraints: ensure that the lead car in the sub-car group has enough space to complete the lane change when it reaches the candidate clearance; Time-of-collision constraint (TTC constraint): Ensures the time safety of the lead vehicle in a group of vehicles and the vehicles in front and behind it during lane changing; Candidate gap comprehensive evaluation function: Quantitatively evaluate the candidate gaps that meet the above constraints, and select the optimal gap as the target lane-changing gap. If multiple sub-car groups select the same gap, they will change lanes in order of evaluation value from smallest to largest.

[0054] The above methods can enable safe, orderly, and efficient lane changing for mixed-traffic vehicle groups in the target lane.

[0055] S22 Establish geometric clearance constraints During the gap search process, the target lane-changing gap needs to satisfy the geometric gap constraint to ensure that the lead car in the sub-car group has sufficient space to complete the lane-changing operation when it reaches the target lane-changing position. The specific constraint is expressed as follows:

[0056] in, For the group of vehicles Medium target candidate gap The position of the following vehicle at the moment of lane change; t The current moment; For the group of vehicles Interval between arrivals of the middle car Time required to change lanes; To ensure a safe distance or reserve a buffer distance; For the group of vehicles The length of the lead car. Sub-cart group The position of the lead car at the moment of lane change; Sub-cart group Medium target candidate gap The position of the vehicle in front at the moment of lane change; For the group of vehicles Intermediate gap The length of the vehicle in front.

[0057] This formula is part of the geometric clearance constraint, used to ensure that the lead car in a sub-group has sufficient longitudinal space to complete the lane change when it reaches the candidate lane-changing gap position, preventing collisions with vehicles in front and behind. Left boundary: Minimum safe distance between the lead car in the sub-group and the vehicle behind the gap. Right boundary: Minimum safe distance between the lead car in the sub-group and the vehicle in front of the gap.

[0058] in:

[0059]

[0060]

[0061]

[0062] in: For the group of vehicles Middle car and gap The longitudinal speed difference of the following vehicle; For the group of vehicles The longitudinal speed of the lead car; For the group of vehicles Intermediate gap The longitudinal speed of the following vehicle; For the group of vehicles Middle car and gap The longitudinal acceleration difference of the following vehicle; For the group of vehicles The longitudinal acceleration of the lead car; For the group of vehicles The longitudinal acceleration of the lead car; For the group of vehicles Middle car and candidate gap Distance to the lane change position; Candidate gap The position of the car behind; The position of the lead car in the sub-car group; The lane change trigger position is when the sub-car group The middle car exceeds the gap The distance to the car behind reached Lane changing can be performed at any time.

[0063] This constraint ensures that the vehicle has sufficient space in the longitudinal direction, avoiding insufficient space during lane changes.

[0064] S23 Establish collision time constraints During the candidate lane change gap search process, to ensure the lead car in the sub-group maintains time safety with vehicles in front and behind during lane changes, collision time constraints (TTC constraints) must be met. Specifically, these include: 1. Time constraint for forward collision:

[0065] in, For the group of vehicles Middle car and gap The time of the collision with the vehicle in front; , Position and speed of the lead car in the sub-car group; , The candidate gap is determined by the position and speed of the vehicle ahead.

[0066] 2. Rear vehicle collision time constraint:

[0067] in, For the group of vehicles Middle car and gap The time of the collision with the following vehicle; , The position and speed of the vehicle following the candidate gap.

[0068] 3. Rear vehicle collision time constraint: To ensure lane-changing safety, the collision time between the front and rear vehicles must be greater than the safe headway. :

[0069] S24 Candidate Gap Comprehensive Evaluation Function During the candidate lane change gap search process, there may be more than one candidate gap that simultaneously satisfies both geometric gap constraints and collision time constraints. To determine the most suitable target gap, this embodiment constructs a comprehensive evaluation function for candidate gaps. Each candidate gap is quantitatively evaluated, and the candidate gap with the best evaluation is selected as the target lane change gap. The comprehensive evaluation function expression for the candidate gap is as follows:

[0070] in, , , Let be the weight coefficient, and satisfy... ; The cost of execution.

[0071] When multiple candidate gaps simultaneously satisfy both geometric gap constraints and collision time constraints, select... The gap with the smallest value is taken as the target lane-changing gap. If multiple sub-car groups select the same gap as the target lane-changing gap, then it is necessary to follow the procedure... The values ​​are selected from smallest to largest to change lanes sequentially.

[0072] As one implementation method, S3 includes the following steps: This embodiment establishes a longitudinal kinematic model of the vehicle based on its kinematic characteristics and designs a phased cooperative lane-changing control algorithm, including: longitudinal sliding mode control in the lane-changing triggering stage; lateral and longitudinal cooperative control in the lane-changing execution stage; and a sub-vehicle group cooperative gap-expanding strategy.

[0073] S31. Establish a vehicle kinematic model To achieve coordinated control and collision warning of vehicles in a mixed vehicle fleet, a unified vehicle kinematics description model is established. In accident bottleneck scenarios, a classic point mass motion model is used to model the motion states of connected autonomous vehicles (CAVs) and connected human-driven vehicles (CHVs). Within this framework, the... The kinematic behavior of the vehicle is expressed as follows:

[0074] In the formula, , , They represent the first The vehicle is Position, velocity, and acceleration information at any given moment; Indicates the first The car is The control input at a given time indicates the position of the vehicle in the mixed traffic group. The acceleration of a vehicle, that is, the rate of change of acceleration.

[0075] S32. Longitudinal control algorithm during lane change triggering phase After the mixed-traffic vehicle groups in lane 1 have completed the allocation of the target lane change gap, the mixed-traffic vehicle group selected as the first to execute a lane change at the target lane change gap enters the lane change triggering phase. In this phase, the lead vehicle of the vehicle group needs to move to the lane change triggering position corresponding to the target lane change gap to meet the subsequent lane change execution conditions. In this embodiment, this process is defined as the lane change triggering phase.

[0076] During the lane-change triggering phase, the cloud control platform sends longitudinal control commands to the lead vehicle of the sub-group based on the target lane-change gap position and vehicle operating status. This coordinates the lead vehicle's position, speed, and acceleration, gradually bringing it closer to the vehicle following the target lane-change gap. Once the lead vehicle completes the lane change and enters lane 0, the vehicle immediately following it in the sub-group automatically becomes the new lead vehicle for the remaining un-changed lanes, continuing to receive longitudinal control feedback commands from the cloud control platform. This process repeats until all vehicles in the sub-group have completed the lane change. At any given time, only the current lead vehicle in the sub-group is subject to cloud-controlled longitudinal control; the remaining vehicles maintain autonomous driving and operate according to a car-following model.

[0077] Longitudinal control of the lead car For the selected target lane change clearance car group The longitudinal control input of the lead car is defined as follows:

[0078]

[0079]

[0080] in, For the group of vehicles l The longitudinal control input for the lead car represents the rate of change of vehicle acceleration; , , The acceleration, velocity, and position of the vehicle after the lane change interval; , , The acceleration, velocity, and position of the lead car in the sub-group; , To control the gain coefficient; , The correlation coefficient for the reaching law of sliding mode control; To control the exponent of the power term in sliding mode; For the sliding surface function of the lead car; To suppress the saturation function of chattering, The parameter is on the order of magnitude; D This is the location where the lane change is triggered.

[0081] longitudinal control of following vehicles For the remaining vehicles in the subgroup, the drivers operate autonomously and follow the car-following model. Their longitudinal control inputs are:

[0082] in, For the first k Longitudinal control input for following vehicles; The optimal velocity function; For the first k Predicted vehicle location; For the first k The speed of connected vehicles and human drivers; For the first k The predicted distance between the vehicle and the vehicle in front; The desired vehicle spacing in the lane-changing coordination area; , The difference between relative velocity and acceleration; , , , These are the parameters for the car-following model.

[0083] In this embodiment: the longitudinal control command is an acceleration signal, which is sent from the cloud control platform to the on-board unit (OBU) of the lead vehicle through the C-V2X communication link. The OBU then sends torque / throttle opening commands to the vehicle's longitudinal dynamics actuator (drive motor controller or electronic throttle) through the vehicle's CAN bus, causing the lead vehicle to produce the desired longitudinal acceleration and deceleration motion.

[0084] S33. Lateral control algorithm during lane change triggering phase Please see Figure 5 In Step 2, when the mixed-traffic sub-group meets the lane-changing trigger condition and enters the lane-changing phase, the lead car of the sub-group needs to enter the target lane-changing gap in the lateral direction. The control algorithm in this stage uses lateral sliding mode control to achieve orderly adjustment of the vehicle's lateral position and heading angle, ensuring a safe and smooth lane change.

[0085] For the sub-car group l The lateral acceleration of the head car With front wheel steering angle The control input is defined as follows:

[0086]

[0087]

[0088]

[0089] in, for Time car group The lateral acceleration of the lead car; for Time car group The front wheel angle of the lead car; The longitudinal speed of the lead car in the sub-car group; This refers to the wheelbase of the lead car; To control the gain coefficient; This is the correction amount for the heading angle of the lead vehicle; The heading angle of the lead vehicle; The heading angle for target lane 0; For the sliding surface function of the lateral control of the lead car; , These are the correlation coefficients of the approach law in lateral sliding mode control; Saturation function, used to suppress and control chattering; These are boundary layer parameters; The lateral position of the lead car; The target lateral position (lane change centerline).

[0090] In this embodiment: the lateral control command is the front wheel steering angle signal, which is sent from the cloud control platform to the electric power steering system EPS of the lead vehicle through the C-V2X communication link, so that the lead vehicle produces the desired lateral displacement and heading angle change.

[0091] S34. Sub-vehicle group cooperative gap widening control strategy When the lead vehicle of a sub-group in lane 1 arrives at lane 0, coordinated lane widening control is needed to allow the remaining vehicles in the sub-group to smoothly merge into lane 0 and complete the lane change. During this phase, the lead vehicle in the sub-group decelerates to create a safe distance, providing lane-changing space for the following vehicles.

[0092] Please refer to Step 3 in section 5. Considering the possibility of multiple sub-car groups changing lanes and widening gaps simultaneously, the lead car in a sub-car group with a later target gap number requires a greater deceleration during the gap widening process. Sub-car group 1 The lead vehicle in the middle is in the target gap The longitudinal deceleration at that point can be expressed as:

[0093] In the formula, The reference deceleration of the sub-vehicle group within its target gap; The additional increment in the deceleration of the lead car in subsequent target gaps is defined as:

[0094] in, The maximum permissible deceleration; It is a coefficient of order of magnitude; The total number of target gaps.

[0095] Using the above strategy, the lead car of the sub-group decelerates and widens the gap in sequence according to the target gap, ensuring that subsequent vehicles can safely and orderly merge into the target lane, thereby achieving coordinated control of vehicle spacing during the overall lane change process of the sub-group.

[0096] As one implementation method, S4 includes the following steps: After each mixed-traffic vehicle group completes its lane change and enters the target lane, in order to eliminate speed fluctuations and vehicle spacing deviations caused during the lane change process and enable the entire mixed-traffic vehicle group to quickly return to a stable operating state, this embodiment adopts a centralized cooperative control method to implement consistency control of the mixed-traffic vehicle group. Specifically, it includes the following steps: S41 Vehicle Group Status Perception and Cooperative Control The edge server collects real-time operational status information of each vehicle in the mixed traffic group and uploads it to the cloud control platform. The operational status information includes at least vehicle position, vehicle speed, and vehicle acceleration.

[0097] Based on the received status information of the entire vehicle group, the cloud control platform centrally analyzes the operational status of the mixed-traffic vehicle group and calculates control feedback commands according to the preset collaborative control strategy. Subsequently, the cloud control platform sends the control feedback commands to each connected autonomous vehicle and connected human-driven vehicle to coordinate and control the speed and spacing of the entire mixed-traffic vehicle group, thereby reducing traffic disturbances caused by lane changing and improving the overall operational stability of the vehicle group.

[0098] S42 Vehicle Group Consistency Constraint Control To ensure the consistency and stability of mixed-traffic vehicles in a single lane, consistency constraints are set on the cloud control platform. When centrally calculating the control feedback of all vehicles, the cloud control platform also needs to consider these consistency constraints. The specific expression is as follows:

[0099] in, For the first i The car at any time t Location; To indicate the first i The car at any time t speed; This indicates the desired headway during the operation of the train group.

[0100] The above constraint means: when When the error between the actual headway and the expected headway between two adjacent vehicles in a mixed traffic group approaches zero, and the speed difference between two adjacent vehicles in the mixed traffic group also approaches zero, the entire mixed traffic group can be guaranteed to achieve a consistent operating state.

[0101] In this embodiment: the control feedback command is sent to the on-board unit (OBU) of each connected vehicle via the C-V2X communication link. The OBU sends acceleration / braking / steering commands to the power system, braking system and steering system of each vehicle through the vehicle CAN bus, thereby changing the longitudinal and lateral motion states of each vehicle.

[0102] S43 train group status convergence and stable operation Based on the constraints of vehicle group consistency, the cloud control platform continuously monitors and adjusts the operating status of the mixed traffic group, so that the mixed traffic group gradually returns to a stable following state after the lane change is completed.

[0103] Please see Figure 5 In Step 4, after experiencing the disruption of a traffic bottleneck event, the connected autonomous vehicles and connected human drivers are controlled according to a predetermined model. By continuously reducing the speed error and spacing error between adjacent vehicles, the operating state of the entire mixed traffic group quickly converges to a stable state, achieving consistent control of the mixed traffic group and safely and efficiently passing through the accident bottleneck area.

[0104] Through the aforementioned vehicle group status perception, collaborative control, and consistency constraint mechanisms, traffic fluctuations generated during lane changing can be effectively suppressed, improving the operational stability, traffic efficiency, and traffic safety of mixed-traffic vehicle groups after lane changing.

[0105] The present invention also provides a computer-readable storage medium having a computer program stored thereon. When executed by a processor, the computer program can implement the centralized cooperative lane-changing control method for mixed-traffic clusters based on a cloud-edge-device architecture under accident bottlenecks described in any of the above embodiments, including the following steps: 1. Data Collection The system collects vehicle operation status information of mixed traffic groups through road test equipment and edge servers, including vehicle position, speed, acceleration, heading angle, vehicle length, vehicle type and lane information; the collected data is then uploaded to the cloud control platform for centralized processing.

[0106] 2. Sub-vehicle group division Based on vehicle type and driving status, the mixed traffic group is divided into several sub-groups with connected autonomous vehicles as the lead vehicles and connected human-driven vehicles as the follow vehicles; each sub-group serves as the basic execution unit for subsequent lane change control.

[0107] 3. Determining the target lane change clearance Based on the candidate lane change gap set, the optimal target lane change gap is matched for each sub-car group using geometric gap constraints, collision time constraints, and a comprehensive evaluation function; if multiple sub-car groups choose the same gap, they change lanes in order of increasing evaluation value.

[0108] 4. Phased Coordinated Lane Changing Control Lane change triggering phase: longitudinal sliding mode control brings the lead car of the sub-car group closer to the target lane change gap; lane change execution phase: lateral control enables the vehicle to smoothly merge into the target lane; sub-car group coordinated gap widening: provides safe lane change space for subsequent vehicles, ensuring the safety of lane change sequence and vehicle spacing.

[0109] 5. Vehicle group consistency control After the lane change is completed, the cloud control platform implements consistent control based on the current status of the mixed traffic group, coordinating the speed and spacing of the traffic group, so that the entire traffic group can quickly return to a stable operating state.

[0110] 6. Issuance of control commands The processed control commands are sent to each vehicle in the subgroup to achieve closed-loop execution of longitudinal and lateral control, ensuring the safe and efficient passage of the mixed-traffic group within the accident bottleneck area.

[0111] In this embodiment, a centralized collaborative lane-changing control system for mixed traffic vehicles based on a cloud-edge-device architecture under accident bottlenecks is also proposed. The system is characterized by being composed of a cloud control platform, roadside sensing devices, and a vehicle-to-everything (V2X) communication module.

[0112] The cloud control platform includes a processor and a non-transitory memory, wherein the non-transitory memory stores a computer program, and the processor executes the computer program for: Receive vehicle operating status information uploaded by roadside sensing devices; Based on the vehicle operating status information, a mixed-traffic sub-vehicle group division rule is constructed to determine the target lane-changing gap for each sub-vehicle group. Establish a phased collaborative lane-changing control strategy to generate longitudinal and lateral control commands; Establish a vehicle group consistency control mechanism to coordinate and control the speed and spacing of mixed traffic groups after lane changes; Roadside sensing equipment, including at least one of a roadside unit (RSU), millimeter-wave radar, lidar, or video camera, is used to collect vehicle operating status information of mixed-traffic vehicles. The vehicle-to-everything (V2X) communication module is used to establish a communication connection between the cloud control platform and the mixed-traffic vehicles, and to send the control commands to the on-board unit (OBU) of each connected vehicle via the V2X communication channel, so that the vehicle physically performs the corresponding lateral and longitudinal movement adjustments.

[0113] The system in this embodiment implements an end-to-end physical signal acquisition, decision-making, and execution link through the following five levels of hardware modules, which work together as follows: (1) Roadside perception layer: It consists of multiple roadside units (RSUs), millimeter-wave radar, lidar and video cameras deployed within 200 to 500 meters upstream of the accident bottleneck area. It is used to collect the position, speed, acceleration, heading angle, vehicle length, vehicle type and lane marking information of each vehicle in the mixed traffic group in real time. The sampling frequency of the roadside perception layer is not less than 20Hz, and the end-to-end delay of the raw perception data of a single point from collection to reception by the edge server does not exceed 50ms. (2) Edge computing layer: It consists of edge servers deployed in roadside equipment rooms or mobile edge computing nodes (MECs). The edge servers are equipped with industrial-grade multi-core processors and GPU acceleration cards, which are used to perform time synchronization, coordinate system transformation and multi-sensor target-level fusion processing on roadside perception data. The time synchronization adopts the PTP precise time protocol, and the synchronization accuracy of the whole network does not exceed 1μs. The output frequency of the fused vehicle target list is not less than 20Hz, and the single-frame fusion processing latency does not exceed 20ms. The fused data is transmitted back to the cloud control platform through a fiber optic private network or 5G link. (3) Cloud-based decision layer: Composed of a cloud control platform server, equipped with a non-transitory memory and at least one processor. When the computer program stored in the non-transitory memory is executed by the processor, it implements the steps S1 to S4 of the method described above. The working sequence of the cloud control platform includes: receiving the vehicle target list after fusion by the edge computing layer (reception delay not exceeding 30ms); performing sub-vehicle group division, target lane change gap search and longitudinal / lateral control law calculation (calculation delay not exceeding 50ms); generating longitudinal control commands and lateral control commands and sending them to the on-board unit (OBU) of the head vehicle of the target sub-vehicle group via the vehicle network V2X communication channel (command sending delay not exceeding 100ms); the control cycle T of the cloud control platform is 50ms to 100ms, corresponding to a control frequency of 10Hz to 20Hz. (4) Vehicle-side execution layer: It consists of the on-board unit (OBU) of each connected vehicle. The OBU receives the longitudinal control command and lateral control command issued by the cloud control platform through the C-V2X communication module, and distributes the longitudinal control command to the vehicle power system (including drive motor controller MCU and / or electronic throttle) and braking system (including brake-by-wire ESP-hev) of the lead vehicle through the vehicle CAN bus (using CAN 2.0B protocol, communication rate 500kbps). It also distributes the lateral control command to the electric power steering system EPS of the lead vehicle. The vehicle power system, braking system and steering system physically execute the control command to make the lead vehicle produce the desired longitudinal acceleration and deceleration motion and lateral steering motion. The instruction distribution delay between the OBU and each execution system does not exceed 10ms. (5) Communication layer: It consists of a C-V2X PC5 / Uu dual-mode base station deployed on the cloud control platform and an on-vehicle C-V2X communication terminal deployed on the vehicle side; among them, the PC5 direct communication interface operates in the 5.9GHz frequency band, with an effective direct communication distance of not less than 300 meters, and is used for low-latency direct communication between vehicles and the road; the Uu cellular communication interface is based on 4G / 5G public network or private network and is used for high-bandwidth sensing data backhaul; from the time the cloud control platform issues the control command to the time the target sub-vehicle group head vehicle OBU receives and decodes it, the end-to-end communication latency does not exceed 100ms, the communication reliability is not less than 99.9%, and the packet loss rate does not exceed 0.1%.

[0114] This invention targets mixed vehicle groups led by intelligent connected vehicles, pre-matching each subgroup to a designated target gap. A lead vehicle actively aligns with the vehicle behind it using longitudinal sliding mode control, transforming multi-vehicle spatial competition into precise tracking of a single gap, enabling the entire group to enter lanes in an orderly manner. Longitudinally, a sliding mode controller is used to robustly converge the lead vehicle to the desired gap. Laterally, a chain-like triggering rule based on the preceding vehicle's state is constructed, ensuring that subsequent vehicles only change lanes sequentially when the preceding vehicle has completed its lane change and sufficient safety conditions are met. After a lane change, a unified car-following model is seamlessly handed over, ensuring smooth integration of heterogeneous traffic flows. Relying on a cloud-edge-device architecture, a centralized architecture acquires global traffic status in real time. Vehicle group segmentation, gap assignment, sliding mode and car-following command calculation, and lateral lane-change decision are integrated in the cloud. Each functional module is modularly connected for perception and execution, forming an end-to-end verification system from collaborative decision-making to single-vehicle closed-loop control.

[0115] This invention addresses the problem of coordinated lane changing in mixed-traffic vehicle groups under accident bottleneck scenarios. It establishes a mixed-traffic sub-vehicle group segmentation mechanism with a connected autonomous vehicle as the lead vehicle, and pre-matches target lane-changing gaps for each sub-vehicle group. Through unified organization and coordinated control of the sub-vehicle groups, orderly lane changing of mixed-traffic vehicles is achieved, reducing the number of conflicts between connected autonomous vehicles and connected human-driven vehicles in the mixed-traffic group, thereby reducing vehicle interaction complexity. Through geometric gap constraints and collision time constraints, the lead vehicle of the sub-vehicle group maintains a safe distance and a safe headway with the vehicles in front and behind, preventing collisions during lane changing. It also improves lane-changing efficiency in accident bottleneck areas: by utilizing the exponential approach characteristic of the longitudinal sliding mode control law, the acceleration signal of the lead vehicle is strictly constrained within a limited range during execution, avoiding uncomfortable longitudinal impacts on the vehicle.

[0116] This invention constructs a phased collaborative lane-changing control strategy comprising longitudinal control during the lane-changing triggering phase, lateral control during the lane-changing execution phase, and collaborative gap-expanding control after the lane change. Specifically, longitudinal sliding mode control gradually brings the lead vehicle of the sub-group closer to the target lane-changing gap, lateral control ensures smooth merging of vehicles into the target lane, and the collaborative gap-expanding mechanism provides safe lane-changing space for subsequent vehicles, thereby improving the safety, stability, and traffic efficiency of lane-changing in mixed-traffic groups.

[0117] This invention relies on a cloud-edge-device collaborative architecture to achieve coordinated operation of vehicle status perception, sub-vehicle group division, target gap search, collaborative decision-making, and control execution. Through centralized computing on the cloud control platform and collaborative processing by edge servers, it completes the coordinated lane-changing control of mixed-traffic groups and the consistency control of the vehicle group after lane changing, enabling the vehicle group speed and spacing to quickly return to a stable state after lane changing, thereby improving the overall operating efficiency, safety, and stability of mixed-traffic groups in accident bottleneck areas.

[0118] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A centralized cooperative lane-changing control method for mixed-traffic vehicle clusters based on a cloud-edge-device architecture under accident bottleneck conditions, characterized in that, Executed by a cloud control platform including a processor and non-transitory memory, the method includes the following steps: S0. Collect vehicle operation status information of mixed traffic groups in the accident bottleneck area through roadside sensing devices and upload it to the cloud control platform. The vehicle operation status information includes at least vehicle position, speed, acceleration, heading angle, vehicle length, vehicle type and lane information. S1. The cloud control platform constructs a collaborative lane-changing operation mechanism for mixed traffic groups based on a cloud-edge-device architecture under accident bottleneck scenarios, and establishes rules for dividing mixed traffic sub-groups to achieve collaborative organization and management of mixed traffic groups. S2. The cloud control platform acquires the vehicle operation status information of the mixed traffic group, constructs a candidate lane-changing gap set, and establishes a target lane-changing gap search and matching mechanism to determine the target lane-changing gap for each sub-group of vehicles. S3. The cloud control platform establishes a vehicle kinematic model based on the vehicle's kinematic characteristics and constructs a phased collaborative lane-changing control strategy to perform lane-changing trigger control, lane-changing execution control, and collaborative gap-expanding control on the sub-vehicle group to complete orderly lane changing. S4. The cloud control platform obtains the vehicle group status information after the lane change is completed, establishes a vehicle group consistency control mechanism, and coordinates and controls the speed and spacing of the mixed vehicle group so that the vehicle group can quickly return to a stable operating state after the lane change is completed. S5. The cloud control platform sends control commands to the lead vehicle of the target sub-vehicle group via the vehicle network communication channel, so that the vehicle power system of the lead vehicle performs lateral and longitudinal movement adjustments.

2. The centralized collaborative lane-changing control method for mixed-traffic vehicle groups based on a cloud-edge-device architecture under accident bottleneck conditions as described in claim 1, characterized in that, The rules for dividing mixed-vehicle groups in S1 include: The connected autonomous vehicle is designated as the lead vehicle of the sub-vehicle group, and the connected human-driven vehicles that follow it are classified as members of the sub-vehicle group until the next connected autonomous vehicle appears. When there are no connected human drivers following behind a connected autonomous vehicle, the connected autonomous vehicle constitutes a separate sub-group of vehicles.

3. The centralized cooperative lane-changing control method for mixed-traffic vehicle clusters based on a cloud-edge-device architecture under accident bottleneck conditions as described in claim 1 or 2, characterized in that, The candidate lane change gap set is constructed in S2 as follows: The longitudinal clearance between any two adjacent vehicles in the target lane is used as the candidate lane-changing clearance, and all candidate clearances are combined to form a candidate clearance set. Represented as: in, The subscript indicates the corresponding candidate lane change gap number. This represents the candidate lane change gap between the first and second vehicles. The total number of vehicles in the target lane; The target lane change clearance matching in S2 is achieved through geometric clearance constraints, which satisfy the following: in, For the group of vehicles Medium target candidate gap The position of the following vehicle at the moment of lane change; t The current moment; For the group of vehicles Interval between arrivals of the middle car Time required to change lanes; To ensure a safe distance or reserve a buffer distance; For the group of vehicles The length of the lead car; Sub-cart group The position of the lead car at the moment of lane change; Sub-cart group Medium target candidate gap The position of the vehicle in front at the moment of lane change; For the group of vehicles Intermediate gap The length of the vehicle in front; The geometric clearance constraint ensures that the lead car in the sub-car group has sufficient longitudinal space to complete the lane change operation when it reaches the candidate lane change gap position, preventing the lead car in the sub-car group from colliding with the vehicles in front and behind.

4. The centralized cooperative lane-changing control method for mixed-traffic vehicle clusters based on a cloud-edge-device architecture under accident bottleneck conditions as described in claim 3, characterized in that, The target lane change gap matching in S2 also includes a collision time constraint, which satisfies the following: Forward collision time constraint: in, For the group of vehicles Middle car and gap The time of the collision with the vehicle in front; , The position and speed of the lead car in the sub-car group; , The candidate gap is determined by the position and speed of the vehicle in front. Rear-end collision time constraint: in, For the group of vehicles Middle car and gap The time of the collision with the following vehicle; , The position and speed of the vehicle following the candidate gap; Rear-end collision time constraint: To ensure lane-changing safety, the collision time between the front and rear vehicles must be greater than the safe headway. : The collision time constraint ensures that the lead car in the sub-vehicle group and the vehicles in front and behind maintain time safety during lane changing. When both the collision time of the lead car and the collision time of the rear car are greater than the safe headway, the time safety condition for lane changing operation is confirmed to be met.

5. The centralized collaborative lane-changing control method for mixed-traffic vehicle clusters based on a cloud-edge-device architecture under accident bottleneck conditions as described in claim 4, characterized in that, The target lane change gap matching in S2 also includes a comprehensive evaluation function for candidate gaps: in, , , Let be the weight coefficient, and satisfy... ; To determine the execution cost, select The smallest candidate gap is taken as the target lane change gap; The comprehensive evaluation function quantifies and evaluates candidate gaps that meet geometric gap constraints and collision time constraints, and selects the candidate gap with the lowest execution cost as the target lane-changing gap, thereby reducing vehicle lane-changing conflicts and traffic disturbances.

6. The centralized cooperative lane-changing control method for mixed-traffic vehicle clusters based on a cloud-edge-device architecture under accident bottleneck conditions as described in claim 1, characterized in that, The lane change trigger control in S3 includes: The cloud control platform generates longitudinal control commands based on the position of the target lane change gap and the position, speed, and acceleration differences between the lead car of the sub-car group and the cars following the target lane change gap, and sends them to the lead car of the sub-car group. The lead vehicle of the sub-vehicle group adjusts its longitudinal movement state according to the longitudinal control command, gradually approaching the lane change trigger position corresponding to the target lane change gap; Once the lead car in the sub-car group completes the lane change, the vehicle immediately following in the original sub-car group automatically becomes the new lead car and continues to receive longitudinal control commands from the cloud control platform until all vehicles in the sub-car group have completed the lane change.

7. The centralized cooperative lane-changing control method for mixed-traffic vehicle clusters based on a cloud-edge-device architecture under accident bottleneck conditions as described in claim 6, characterized in that, Lane change execution control in S3 includes: When the lead car of the sub-vehicle group meets the preset lane-changing trigger condition, the lead car of the sub-vehicle group is controlled to move laterally towards the target lane; The lateral control generates steering control commands based on the vehicle's lateral position deviation and heading angle deviation to control the vehicle to gradually converge toward the center line of the target lane. Once the lead vehicle of the sub-vehicle group completes its lane change, subsequent vehicles will perform lane changes in a preset order to allow the entire sub-vehicle group to merge into the target lane.

8. The centralized cooperative lane-changing control method for mixed-traffic vehicle clusters based on a cloud-edge-device architecture under accident bottleneck conditions as described in claim 7, characterized in that, Cooperative clearance expansion control in S3 and vehicle group consistency control in S4 include: Once the lead vehicle in the group of vehicles enters the target lane, it reduces its speed to increase the distance between itself and the vehicle in front, providing space for subsequent vehicles to change lanes. When multiple sub-car groups perform lane changes simultaneously, the gap widening control intensity corresponding to each sub-car group is determined based on the sorting result of the target lane change gap. After all vehicles have completed lane changing, the position and speed information of each vehicle in the mixed traffic group is obtained, and the spacing and speed difference between adjacent vehicles are coordinated and controlled so that the actual spacing between adjacent vehicles approaches the preset spacing and the speed difference between adjacent vehicles approaches zero, thereby restoring the mixed traffic group to a consistent operating state.

9. A centralized cooperative lane-changing control system for mixed-traffic vehicles based on a cloud-edge-device architecture under accident bottleneck conditions, characterized in that, include: The cloud control platform includes a processor and a non-transitory memory, wherein the non-transitory memory stores a computer program, and the processor executes the computer program for: Receive vehicle operating status information uploaded by roadside sensing devices; Construct a mixed-traffic sub-vehicle group division rule, based on the vehicle operation status information, and determine the target lane change gap for each sub-vehicle group according to the division rule; Establish a phased collaborative lane-changing control strategy to generate longitudinal and lateral control commands; Establish a vehicle group consistency control mechanism to coordinate vehicle group speed and spacing; Roadside sensing equipment, including at least one of roadside unit, millimeter-wave radar, lidar or video camera, is used to collect vehicle operating status information of mixed traffic vehicles; The vehicle-to-everything (V2X) communication module is used to establish a communication connection between the cloud control platform and the mixed-traffic vehicles, and to send the control commands to each vehicle so that the vehicle can perform lateral and longitudinal movement adjustments.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method of any one of claims 1 to 8, wherein the cloud control platform acquires vehicle operating status information through roadside sensing devices and sends control commands to the vehicle through the vehicle network communication channel, so that the vehicle's power system, braking system and / or steering system physically execute the corresponding control commands to complete lateral and longitudinal movement adjustments.

Images