Multi-vehicle emergency avoidance cooperative control method and system
By establishing a distributed trajectory negotiation and locking method in multi-vehicle emergency avoidance scenarios through V2V communication, the problem of decision conflict and jitter in multi-vehicle emergency avoidance is solved, and fast and accurate avoidance control is achieved, avoiding secondary collisions and decision instability.
Patent Information
- Application Number
- CN202511561894.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-29
- Publication Date
- 2026-02-06
AI Technical Summary
When multiple autonomous vehicles face the risk of an emergency collision, existing technologies suffer from problems such as decision-making conflicts, decision jitter, lack of a global perspective, and insufficient communication protocols, leading to secondary collisions and unstable decision-making.
By establishing a distributed trajectory negotiation and locking method through V2V communication, vehicles form a temporary decision-making community within milliseconds, negotiate a conflict-free combination of avoidance trajectories, and enforce locking to avoid secondary collisions and decision jitter.
It improves the accuracy and precision of vehicle avoidance control, shortens the control response time, ensures rapid convergence and robustness of the decision-making process, and avoids decision conflicts and secondary collisions.
Smart Images
Figure CN121483085A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of intelligent vehicle control, in particular to a multi-vehicle emergency avoidance cooperative control method and system. BACKGROUND
[0002] In the field of intelligent driving, collision avoidance technology is mainly divided into single-vehicle autonomous avoidance and centralized cooperative avoidance. Single-vehicle autonomous avoidance technology relies on vehicle sensors to perceive the environment and independently decides to execute avoidance actions (deceleration, turning). This is the main way of current L2 / L3 level automatic driving systems. For example, when detecting an obstacle in front, the AEB (automatic emergency braking) system will start braking. The advantage of this approach is simplicity and independence from external communication.
[0003] Centralized cooperative avoidance is to upload the state information of multiple vehicles to a cloud control platform or roadside MEC through V2N (vehicle-to-cloud) communication. The central node performs global optimal trajectory planning and then issues instructions to each vehicle for execution. This approach can theoretically achieve global optimization, but in an emergency scenario, the communication round-trip delay (from vehicle perception to danger, data upload, cloud computing, and then instruction issuance) may be too long (usually in the order of hundreds of milliseconds), missing the best avoidance opportunity.
[0004] With the development of V2V direct communication technology (such as DSRC, C-V2X), preliminary cooperative safety applications have emerged, such as cooperative emergency braking (CEB). When a vehicle in front brakes in an emergency, it broadcasts an event message through V2V, and the vehicle behind can decelerate in advance after receiving it. However, the information is only transmitted in one direction, and complex interactive decisions between multiple vehicles cannot be achieved.
[0005] In the multi-agent path planning (MAPP) method, distributed algorithms such as rule-based (such as social force model), optimization-based (such as distributed model predictive control DMPC), and auction-based are used. However, these algorithms either have too simple rules to handle complex scenarios or have high computational complexity, making it difficult to achieve real-time decision-making in milliseconds on a vehicle computing platform.
[0006] The inventors found in their research that the prior art has the following problems when dealing with "strong interaction" emergency scenarios where multiple autonomous vehicles face collision risks: First, decision conflict and secondary collision risk: Without a cooperative mechanism, each vehicle makes decisions based on its own benefit maximization (i.e., avoiding threats to itself as quickly and safely as possible). This easily leads to "decision conflict". For example, two oncoming vehicles may simultaneously change lanes to the same free lane to avoid the same obstacle, causing a more serious head-on collision.
[0007] Secondly, decision jitter and system instability. When multiple vehicles interact with each other, the avoidance action of one vehicle will change the decision environment of another vehicle, triggering a reaction of the other vehicle, which in turn will affect the first vehicle, forming a "decision-reaction" chain. Without a fast converging negotiation mechanism, vehicles may "jitter" between multiple avoidance options, unable to make a final decision, or cause the entire multi-vehicle system to oscillate and become unstable.
[0008] Thirdly, the "local optimum trap" of lacking a global perspective: the "local optimum" decisions made by each vehicle independently, the combined effect of which for the entire multi-vehicle system, can be a very poor "global suboptimal" or even dangerous solution. For example, to avoid a slight scratch, a vehicle may choose to brake hard, but this results in a serious rear-end collision of the vehicle behind.
[0009] Further, the lack of a communication protocol: existing V2V communication is mainly used for broadcasting its own state (BSM) or simple events, lacking a set of communication protocols specially designed for multi-vehicle negotiation of complex trajectories in emergency scenarios. This set of protocols needs to define: who initiates the negotiation, the content of the negotiation ("decision package" format), the process of the negotiation (multi-round auction, voting, confirmation), and how to ensure the final execution of the negotiation result. SUMMARY Therefore, embodiments of the present application provide a multi-vehicle emergency avoidance cooperative control method and system to solve the above problems, to improve the accuracy of vehicle avoidance control, while shortening the control response time, and improving the accuracy of multi-vehicle cooperative control.
[0010] The purpose of the present application is to solve the problems in the background art, and to provide a distributed trajectory negotiation and locking method for multi-vehicle emergency avoidance scenarios. This method does not rely on a central server, but only uses V2V communication to enable multiple vehicles involved in potential collision risks to form a "temporary decision community" autonomously and quickly, negotiate a set of conflict-free and cooperative combined avoidance trajectories within milliseconds, and once an agreement is reached, it is forced to be locked and executed, thereby effectively resolving emergency conflicts and avoiding secondary collisions and decision jitter.
[0011] In a first aspect, embodiments of the present application provide a multi-vehicle emergency avoidance cooperative control method. The cooperative control method is used for cooperative control of vehicles in a driving intersection area formed by at least two or more vehicles, and includes: According to the intersection area, the first vehicle and the second vehicle pass through the vehicle networking information to obtain the first vehicle predicted driving trajectory information and the second vehicle predicted driving trajectory information; according to the intersection probability of the first vehicle predicted driving trajectory information and the second vehicle predicted driving trajectory information, the first vehicle or the second vehicle sends a negotiation request broadcast to the other party; The vehicle receiving the negotiation request broadcast obtains multiple pieces of candidate low-collision-probability trajectory information according to current vehicle dynamics constraint information and surrounding environment information, and broadcasts the multiple pieces of candidate low-collision-probability trajectory information to other vehicles; the candidate low-collision-probability trajectory information includes candidate trajectory information and trajectory cost information; The vehicle receiving the candidate low-collision-probability trajectory information obtains preferred trajectory information according to trajectory cost information in the received multiple pieces of candidate low-collision-probability trajectory information and the multiple pieces of candidate low-collision-probability trajectory information locally, and broadcasts the preferred trajectory information to other vehicles; The local vehicle obtains current recommended trajectory information according to the received preferred trajectory information and a multiplexing rate of the preferred trajectory information of the local vehicle, and generates vehicle control information according to control of the current recommended trajectory information.
[0012] With reference to the first aspect, in a first possible implementation manner of the first aspect, the method further includes obtaining multiple pieces of first vehicle predicted driving trajectory information and multiple pieces of second vehicle predicted driving trajectory information according to the first vehicle and the second vehicle passing through the intersection area through vehicle networking information; and triggering the first vehicle or the second vehicle to send a negotiation request broadcast to the other party according to an intersection probability of the multiple pieces of first vehicle predicted driving trajectory information and the multiple pieces of second vehicle predicted driving trajectory information.
[0013] With reference to the first aspect, in a second possible implementation manner of the first aspect, the vehicle networking information includes vehicle information and environment information in V2X; and the vehicle information includes vehicle identification information ID, vehicle position information, vehicle current speed information, vehicle acceleration information, and vehicle heading angle information. The step of obtaining the first vehicle predicted driving trajectory information and the second vehicle predicted driving trajectory information according to the first vehicle and the second vehicle passing through the intersection area through vehicle networking information includes: The first vehicle predicted driving trajectory information and the second vehicle predicted driving trajectory information are obtained through a constant steering rate and acceleration model trajectory predictor according to the vehicle identification information ID, the vehicle position information, the vehicle current speed information, the vehicle acceleration information, and the vehicle heading angle information; the first vehicle predicted driving trajectory information and the second vehicle predicted driving trajectory information include trajectory band information in a future prediction time; the trajectory band information includes shape and position information of a trajectory and trajectory width information formed by multiple trajectories.
[0014] With reference to the first aspect, in a third possible implementation manner of the first aspect, the step of triggering the first vehicle or the second vehicle to send a negotiation request broadcast to the other party according to an intersection probability of the first vehicle predicted driving trajectory information and the second vehicle predicted driving trajectory information includes: Based on the trajectory band information in the predicted driving trajectory information of the first vehicle and the trajectory band information in the predicted driving trajectory information of the second vehicle, the overlap of trajectory bands within a future set time is such that the future set time is less than the future predicted time. If the overlap of the trajectory bands is greater than the set overlap threshold, the first vehicle or the second vehicle will be triggered to broadcast a negotiation request to the other party. The negotiation request includes: the communication command identifier of the air interface protocol of the vehicle V2X protocol, the communication identifier field of this communication, the initiator identifier of this communication, the list of recipients of this communication, and the validity period of the message or request.
[0015] In conjunction with the first aspect, this application provides a fourth possible implementation of the first aspect, wherein the step of a vehicle receiving a negotiation request broadcast, obtaining multiple candidate low-collision probability trajectory information based on current vehicle dynamics constraint information and surrounding environment information, and broadcasting the multiple candidate low-collision probability trajectory information to other vehicles includes: The vehicle that receives the negotiation request broadcast verifies its identity as a receiving member according to the list of receiving members for this communication. If the verification is successful, it joins the communication according to the communication identifier field and forms a decision-making community communication group with the vehicle that sent the information. Under the premise that the simulation meets the vehicle dynamics and kinematic constraints, multiple candidate low-collision probability trajectory information is generated by the discretized control input forward simulation method. The multiple candidate low-collision probability trajectory information includes candidate trajectory information and trajectory cost information. The trajectory cost information is obtained by the distance between the trajectory endpoint and the original target point, the acceleration and angular acceleration on the trajectory, the minimum distance between the trajectory and the static obstacle, and the weight coefficients of each parameter. The vehicle currently packages multiple alternative low-collision probability trajectory information into a decision package and broadcasts it to other vehicles.
[0016] In conjunction with the first aspect, this application provides a fourth possible implementation of the first aspect, wherein the steps of a vehicle receiving candidate low-collision probability trajectory information, obtaining preferred trajectory information based on the trajectory cost information in the received multiple candidate low-collision probability trajectory information and the local multiple candidate low-collision probability trajectory information, and broadcasting the preferred trajectory information to other vehicles include: For vehicles that receive alternative low-collision probability trajectory information, the combined cost of alternative trajectories of other vehicles and local vehicles is obtained based on the trajectory cost information from the multiple alternative low-collision probability trajectory information received and the multiple alternative low-collision probability trajectory information locally; based on the combined cost, the preferred trajectory information corresponding to the preferred combined cost value is obtained.
[0017] In conjunction with the first aspect, this application provides another possible implementation of the first aspect, wherein the step of obtaining the current recommended trajectory information by the local vehicle based on the received preferred trajectory information and the reuse rate of the preferred trajectory information of the local vehicle includes: Based on the received preferred trajectory information and the preferred trajectory information of local vehicles, the local vehicles obtain the trajectory reuse rate; If the reuse rate is greater than the set threshold, the current recommended trajectory information is obtained; otherwise, a forced avoidance information command is generated. Local vehicles generate vehicle control information based on the mandatory avoidance information command.
[0018] In conjunction with the first aspect, this application provides another possible implementation of the first aspect, wherein the local vehicle is controlled based on the current recommended trajectory information, and the generated vehicle control information includes: The local vehicle generates a lock execution command based on the current recommended trajectory information; the lock execution command includes the current recommended trajectory information; the local vehicle broadcasts the lock execution command to other vehicles.
[0019] Secondly, embodiments of this application also provide a multi-vehicle emergency avoidance cooperative control system. This cooperative control system is used for cooperative vehicle control within an intersection area formed by the driving trajectories of at least two or more vehicles. It includes: a triggering unit, a candidate trajectory generation unit, a distributed negotiation unit, and a trajectory locking unit; wherein... The triggering unit is configured to obtain the predicted driving trajectory information of the first vehicle and the second vehicle based on the vehicle network information of the first vehicle and the second vehicle in the intersection area; and to trigger the first vehicle or the second vehicle to send a negotiation request broadcast to the other party based on the intersection probability of the predicted driving trajectory information of the first vehicle and the predicted driving trajectory information of the second vehicle. The alternative trajectory generation unit is configured to generate multiple alternative low-collision probability trajectories based on the current vehicle dynamics constraint information and surrounding environment information, and broadcast these multiple alternative low-collision probability trajectories to other vehicles; the alternative low-collision probability trajectory information includes alternative trajectory information and trajectory cost information. The distributed negotiation unit is configured such that when a vehicle receives candidate low-collision probability trajectory information, it obtains the preferred trajectory information based on the trajectory cost information in the multiple candidate low-collision probability trajectory information received and the multiple candidate low-collision probability trajectory information locally, and broadcasts the preferred trajectory information to other vehicles. The trajectory locking unit is configured to allow the local vehicle to obtain the current recommended trajectory information based on the received preferred trajectory information and the reuse rate of the preferred trajectory information of the local vehicle; the local vehicle controls the vehicle based on the current recommended trajectory information and generates vehicle control information.
[0020] In conjunction with the second aspect, this application provides a first possible implementation of the second aspect, wherein the vehicle network information includes vehicle information and environmental information in V2X; the vehicle information includes vehicle identification information ID, vehicle location information, vehicle current speed information, vehicle acceleration information, and vehicle heading angle information; The alternative trajectory generation unit is further configured to obtain the predicted driving trajectory information of the first vehicle and the predicted driving trajectory information of the second vehicle based on the vehicle network information of the first vehicle and the second vehicle in the intersection area, including the following steps: Based on vehicle identification information (ID), vehicle location information, vehicle current speed information, vehicle acceleration information, and vehicle heading angle information, a trajectory predictor using a constant steering rate and acceleration model is used to obtain the predicted driving trajectory information of the first vehicle and the predicted driving trajectory information of the second vehicle. The predicted driving trajectory information of the first vehicle and the predicted driving trajectory information of the second vehicle include trajectory band information within the future prediction time. The trajectory band information includes the shape and position information of the trajectory and the trajectory width information formed by multiple trajectories.
[0021] This invention establishes a multi-vehicle cooperative network within an intersection area. Based on vehicle dynamics constraints and surrounding environmental information, multiple vehicles generate optimal trajectories through a finite number of negotiations. This avoids the efficiency problems associated with centralized computation, improves the accuracy of vehicle avoidance control, and shortens control response time. Furthermore, through distributed negotiation, all relevant vehicles reach a consensus before taking action, effectively avoiding decision conflicts and secondary collisions. The finite multi-round negotiation protocol and "lock-in" mechanism ensure that the decision-making process converges in a very short time. The negotiation process aims to find a solution that is better for the community as a whole, rather than merely satisfying individual optimality, achieving a Pareto improvement effect. The entire negotiation process is based entirely on direct V2V communication, without relying on any central server or roadside MEC. Therefore, the method or system is robust and independent of infrastructure.
[0022] Furthermore, the multi-vehicle emergency avoidance cooperative control method provided in this application embodiment can also trigger a negotiation request broadcast based on the intersection probability of multiple first vehicle predicted driving trajectory information and multiple second vehicle predicted driving trajectory information, thereby improving the accuracy of control.
[0023] Furthermore, the multi-vehicle emergency avoidance cooperative control method provided in this application embodiment can also obtain the predicted driving trajectory information of the first vehicle and the predicted driving trajectory information of the second vehicle through a trajectory predictor of a constant steering rate and acceleration model based on the vehicle identification information ID, vehicle position information, vehicle current speed information, vehicle acceleration information, and vehicle heading angle information, thereby improving the accuracy of trajectory prediction.
[0024] Further, the multi-vehicle emergency avoidance cooperative control method provided by the embodiments of the present application can also trigger the first vehicle or the second vehicle to send a negotiation request broadcast to the other party by setting a coincidence threshold, thereby reducing the frequency of negotiation request broadcast, triggering communication again when negotiation trajectory is really needed, thereby reducing the consumption of system resources and improving the accuracy of effective control.
[0025] Further, the multi-vehicle emergency avoidance cooperative control method provided by the embodiments of the present application can also select a predicted trajectory for each predicted vehicle through the parameters of the alternative trajectory information and the trajectory cost information, so as to obtain an avoidance predicted trajectory under the condition of referring to the actual control difficulty of each vehicle.
[0026] Further, the multi-vehicle emergency avoidance cooperative control method provided by the embodiments of the present application further comprises obtaining preferred trajectory information corresponding to a preferred combination cost value according to the combination cost, thereby improving the accuracy and effectiveness of the preferred trajectory information.
[0027] Further, the multi-vehicle emergency avoidance cooperative control method provided by the embodiments of the present application further comprises obtaining current recommended trajectory information or generating forced avoidance information commands through the reuse rate, thereby improving the safety of avoidance control.
[0028] Further, the multi-vehicle emergency avoidance cooperative control method provided by the embodiments of the present application further comprises generating a lock execution instruction according to the current recommended trajectory information, thereby facilitating subsequent vehicle control.
[0029] In order to make the above objectives, characteristics and advantages of the present application more apparent and easy to understand, the following preferred embodiments are described in detail below, and the accompanying drawings are described as follows. BRIEF DESCRIPTION OF DRAWINGS
[0030] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.
[0031] Figure 1 A flow chart of the multi-vehicle emergency avoidance cooperative control method provided by the embodiments of the present application is shown.
[0032] Figure 2 A schematic diagram of the multi-round distributed trajectory bidding and negotiation steps provided by the embodiments of the present application is shown.
[0033] Figure 3 A composition schematic diagram of the multi-vehicle emergency avoidance cooperative control system provided by the embodiments of the present application is shown. Detailed Implementation
[0034] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.
[0035] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.
[0036] The purpose of this invention is to solve the problems mentioned in the background art and provide a distributed trajectory negotiation and locking method for multi-vehicle emergency avoidance scenarios. This method does not rely on a central server, but only through V2V communication, enabling multiple vehicles involved in potential collision risks to autonomously and quickly form a "temporary decision-making community." Within milliseconds, a set of conflict-free and cooperative combined avoidance trajectories is negotiated, and once a consensus is reached, the trajectories are forcibly locked and executed, thereby effectively resolving emergency conflicts and avoiding secondary collisions and decision jitter.
[0037] This invention is applied to the control of autonomous vehicles and can also be used in assisted driving situations. The triggering unit of the hardware system implementing this method can be deployed on roadside equipment or intersection equipment. Its data communication conforms to 4G, 5G, and V2X data formats and communication standards. The multi-vehicle emergency avoidance cooperative control method in the embodiments of this application is used for the cooperative control of vehicles within an area where at least two or more vehicles are driving and their driving trajectories intersect.
[0038] The distributed trajectory negotiation and locking method proposed in this invention is based on a fast negotiation protocol using a "commitment-consistency" model. When specific triggering conditions are met, the relevant vehicles automatically enter negotiation mode, referring to... Figure 1 This is a flowchart of the multi-vehicle emergency avoidance cooperative control method provided in the embodiments of this application. As shown in the figure, the method includes the following steps: Step S101: Based on the vehicle-to-everything (V2X) information of the first and second vehicles within the intersection area, predictive driving trajectory information of the first and second vehicles is obtained. Based on the intersection probability of the first and second vehicle's predicted driving trajectories, either the first or second vehicle is triggered to broadcast a negotiation request to the other.
[0039] Negotiation Trigger and Community Formation: Each vehicle continuously monitors the V2X information it receives. When a vehicle (initiator) predicts that in the future T time (e.g., 2 seconds), its predicted trajectory overlaps with at least N other vehicles (e.g., N > 1) and the collision probability exceeds a threshold, it immediately triggers negotiation. The initiator broadcasts a "Negotiation Request (NR)" message to these related vehicles, and the vehicles receiving the NR automatically join to form a temporary "Decision Community (DC)".
[0040] In the embodiment, the intersection area described above can be a potential conflict area defined by traffic data collected by intersection cameras or roadside devices, including the dynamic path intersection range of vehicles, pedestrians and non-motor vehicles. The Internet of Vehicles information shares the areas identified in the position, speed, acceleration and steering intention in real time through V2V or V2I communication protocol.
[0041] Further, through the collected images of the interaction area or according to the Beidou satellite positioning data of the vehicle combined with the images, the Internet of Vehicles V2X information of the first vehicle and the second vehicle (at least two vehicles) or more vehicles is obtained by identifying the vehicle contour or identification in the images. The Internet of Vehicles V2X information can be obtained from the roadside device, or directly obtained through the vehicle-mounted communication unit.
[0042] Thus, the future predicted trajectory information is obtained through the V2X information of the first vehicle and the second vehicle or more vehicles.
[0043] Step S102, the vehicle receiving the negotiation request broadcast obtains a plurality of candidate low collision probability trajectory information according to the current vehicle dynamics constraint information and the surrounding environment information and broadcasts the plurality of candidate low collision probability trajectory information to other vehicles. The candidate low collision probability trajectory information includes candidate trajectory information and trajectory cost information.
[0044] Candidate safe trajectory set generation: Each vehicle in the community quickly generates a set containing a plurality of (e.g., 3-5) candidate safe trajectories based on its own dynamics constraints and surrounding environment. Each candidate trajectory is attached with a "cost" value, which is a scalar that comprehensively considers the degree of deviation from the original path, the size of acceleration / angle acceleration (comfort), potential secondary risk, etc.
[0045] Step S103, the vehicle receiving the candidate low collision probability trajectory information obtains the preferred trajectory information according to the trajectory cost information in the plurality of candidate low collision probability trajectory information received and the plurality of candidate low collision probability trajectory information locally, and broadcasts the preferred trajectory information to other vehicles.
[0046] In the embodiment, refer to Figure 2is a schematic diagram of multi-round distributed trajectory bidding and negotiation steps, as shown, including the following steps: Figure 2 Step S201, the first round - intention broadcast. Each vehicle broadcasts its "decision package (DP)" containing all candidate trajectories and their costs to all members in the community through V2V. Step S202, the second round - conflict detection and bidding. After receiving the DP of all other members, each vehicle simulates the execution of all possible trajectory combinations locally. For each of its own candidate trajectories, it checks whether there is a conflict with each of the candidate trajectories of other vehicles. Then, it calculates a "comprehensive score" for each of its own candidate trajectories, which takes into account not only its own cost, but also the "friendliness" of other vehicle trajectories that do not conflict with it (i.e., how low the other cost is). Step S203, the third round - scheme publication and voting. Each vehicle selects the "combined trajectory scheme" (i.e., a set containing one of its own trajectories and one of the other non-conflicting trajectories) that it considers optimal according to the calculation of the last round, and broadcasts this scheme. Step S104, the local vehicle obtains the current recommended trajectory information according to the received preferred trajectory information and the multiplexing rate of the local vehicle's preferred trajectory information. The local vehicle controls according to the current recommended trajectory information to generate vehicle control information.
[0047] Step S204, the fourth round - consensus achievement and locking. Each vehicle collects all the schemes published by the members. If there is a scheme that is commonly recommended (i.e., "voted") by more than a certain percentage (such as 51% or more, configurable) of members, then this scheme is considered a "consensus scheme". All members immediately lock the trajectory assigned to themselves in the scheme and enter the execution state.
[0048] Lock execution and state broadcast: Once the trajectory is locked, the vehicle's planning and control module will enforce the trajectory and no longer accept new negotiation requests until the trajectory is executed or the danger is resolved. At the same time, the vehicle will continue to broadcast its state as "locked" to inform other vehicles that it is executing a coordinated trajectory and avoid external interference. If consensus is not reached within a specified time (such as 500 milliseconds), the system automatically downgrades and executes the most conservative avoidance strategy (such as emergency braking) preset by each vehicle.
[0049] With reference to the first aspect, in a first possible implementation manner of the first aspect, the vehicle information comprises a vehicle state vector. The vehicle state vector is maintained by each vehicle in real time, and comprises a vehicle identifier ID, a vehicle position p, a vehicle current speed v, a vehicle acceleration a, and a vehicle heading angle θ.
[0050] With reference to the first aspect, in a second possible implementation manner of the first aspect, the vehicle information comprises vehicle information and environment information in V2X. The vehicle information comprises a vehicle identifier ID, a vehicle position p, a vehicle current speed v, a vehicle acceleration a, and a vehicle heading angle θ.
[0051] The step of obtaining the first vehicle predicted driving track information and the second vehicle predicted driving track information according to the vehicle-to-vehicle information of the first vehicle and the second vehicle in the intersection region comprises the following steps. The first vehicle predicted driving track information and the second vehicle predicted driving track information are obtained by using a constant steering rate and acceleration model trajectory predictor according to the vehicle identifier ID, the vehicle position p, the vehicle current speed v, the vehicle acceleration a, and the vehicle heading angle θ. The first vehicle predicted driving track information and the second vehicle predicted driving track information comprise track band information in a future prediction time. The track band information comprises shape and position information of a track, and track width information formed by a plurality of tracks.
[0052] Specifically, in the step S101, that is, in the negotiation triggering and community forming process, the vehicle information comprises a vehicle state vector. The vehicle state vector is maintained by each vehicle in real time, and comprises a vehicle identifier ID, a vehicle position p, a vehicle current speed v, a vehicle acceleration a, and a vehicle heading angle θ.
[0053] The trajectory prediction comprises: each vehicle uses a trajectory predictor based on a constant steering rate and acceleration (CTRA) model to predict a track band (considering uncertainty) of the vehicle and other vehicles (obtaining states through BSM) in a future T_pred (for example, 3 seconds).
[0054] In an embodiment, the triggering condition is that a prediction module of Car_A finds that a predicted track band of Car_A overlaps with predicted track bands of Car_B and Car_C in a future t_c<2.0s, and a collision probability P_coll of the overlapping region is greater than 0.8. Car_A becomes the initiator.
[0055] For example, an NR message is sent: Car_A immediately broadcasts a “negotiation request (NR)” message through V2V.
[0056] NR message structure: { "msg_type": "NR", "community_id": "uuid-1234-abcd", "initiator_id": "Car_A", "members": ["Car_B", "Car_C"], "deadline_ms": 500} For example: community formation: Car_B and Car_C receive the NR, verify that they are indeed members, immediately enter the "Negotiating" state, and join the decision community (DC) with ID uuid-1234-abcd.
[0057] In combination with the first aspect, the third possible implementation manner of the first aspect is provided, and the step of triggering the first vehicle or the second vehicle to send a negotiation request broadcast to the other party according to the intersection probability of the first vehicle predicted trajectory information and the second vehicle predicted trajectory information includes: According to the trajectory band information in the first vehicle predicted trajectory information and the trajectory band information in the second vehicle predicted trajectory information, the trajectory band coincidence degree in the future set time, and the future set time is less than the future prediction time.
[0058] If the trajectory band coincidence degree is greater than the set coincidence degree threshold, the first vehicle or the second vehicle is triggered to send a negotiation request broadcast to the other party. The negotiation request includes: a communication instruction identifier of an air interface protocol of a vehicle V2X protocol, a communication identifier field of this communication, an initiator identifier of this communication, a receiving member list of this communication, and a validity period of the message or request.
[0059] In combination with the first aspect, the fourth possible implementation manner of the first aspect is provided, and the step of receiving the negotiation request broadcast, obtaining a plurality of alternative low collision probability trajectory information according to the current vehicle dynamics constraint information and the surrounding environment information, and broadcasting the plurality of alternative low collision probability trajectory information to other vehicles includes: The vehicle receiving the negotiation request broadcast verifies whether the local is a receiving member according to the receiving member list of this communication, and after the verification is passed, the vehicle joins this communication according to the communication identifier field, and forms a decision community communication group with the vehicle sending the information.
[0060] Under the premise that the simulation meets the vehicle dynamics and kinematics constraints, a plurality of alternative low collision probability trajectory information is generated by a discrete control input forward simulation method. The plurality of alternative low collision probability trajectory information includes alternative trajectory information and trajectory cost information. The trajectory cost information is obtained by the distance between the trajectory endpoint and the original target point, the acceleration and angular acceleration on the trajectory, and the minimum distance between the trajectory and the static obstacle and the weight coefficient of each parameter.
[0061] In one embodiment of the present application, the alternative safety trajectory set generation (performed locally and in parallel at each vehicle) generates a set of alternative trajectories (alternative low collision probability trajectories) TS_i = {traj_i,1, traj_i,2,...} for each planning module of Car_A, Car_B, Car_C, respectively, under the premise of satisfying the vehicle dynamics and kinematics constraints.
[0062] The trajectory generation method thereof: for example, a discrete control input (such as [left turn 10°, straight, right turn 10°] x [accelerate, constant speed, decelerate]) can be used for forward simulation to generate multiple candidate trajectories.
[0063] For example, a cost function Cost(traj) calculates the cost of each generated trajectory traj. Cost(traj) = w1* C_dev + w2 * C_dyn + w3 *C_risk C_dev: the distance between the trajectory end point and the original target point, which punishes deviation.
[0064] C_dyn: the integral of acceleration and angular acceleration on the trajectory, which punishes uncomfortable jerks.
[0065] C_risk: the reciprocal of the minimum distance between the trajectory and static obstacles (such as road edges, buildings), which punishes secondary risks.
[0066] w1, w2, w3: weight coefficients. The current vehicle packs the multiple alternative low collision probability trajectory information into a decision package and broadcasts it to other vehicles.
[0067] In combination with the first aspect, the fourth possible implementation manner of the first aspect is provided, and the vehicle receiving the alternative low collision probability trajectory information obtains the preferred trajectory information according to the trajectory cost information in the received multiple alternative low collision probability trajectory information and the local multiple alternative low collision probability trajectory information, and broadcasts the intention broadcast to other vehicles. The step of obtaining the preferred trajectory information includes: The vehicle receiving the alternative low collision probability trajectory information obtains the combined cost of the alternative trajectory of the other vehicle and the alternative trajectory of the local vehicle according to the trajectory cost information in the received multiple alternative low collision probability trajectory information and the local multiple alternative low collision probability trajectory information. According to the combined cost, the preferred trajectory information corresponding to the preferred combined cost value is obtained.
[0068] In conjunction with the first aspect, this application provides another possible implementation of the first aspect, wherein the step of obtaining the current recommended trajectory information by the local vehicle based on the received preferred trajectory information and the reuse rate of the preferred trajectory information of the local vehicle includes: The local vehicle calculates the trajectory reuse rate based on the received preferred trajectory information and its own preferred trajectory information. If the reuse rate is greater than a set threshold, the current recommended trajectory information is obtained. Otherwise, a mandatory avoidance information command is generated. The local vehicle then generates vehicle control information based on the mandatory avoidance information command.
[0069] Multi-round distributed trajectory bidding and negotiation. Negotiation round 1 (T=0~T1ms): Intent to broadcast: Each vehicle packages its set of alternative trajectories and corresponding costs into a "decision package (DP)" and broadcasts it.
[0070] DP message structure { "msg_type": "DP", "community_id": "uuid-1234-abcd", "source_id": "Car_A", "trajectories": [ {"id": "A1", "path": [...], "cost": 10.5}, {"id": "A2", "path": [...], "cost": 12.8}, {"id": "A3", "path": [...], "cost": 8.2} ] When T=T1ms, all cars have received the DP from the other two cars.
[0071] Negotiation Round 2 (T=Ti~T2ms): Conflict Detection and Solution Evaluation (Local Calculation) Taking Car_A as an example, it now has three DPs: DP_A, DP_B, and DP_C.
[0072] It constructs a three-dimensional "combination space" with a size of |TS_A| x |TS_B| x |TS_C|.
[0073] For each combination (traj_A,i, traj_B,j, traj_C,k) in space, perform collision detection. If any two trajectories collide in spacetime, the combination is invalid.
[0074] For all valid combinations, compute its "combination cost" C_combo = Cost(A,i) + Cost(B,j) + Cost(C,k).
[0075] Car_A finds the one that minimizes C_combo, say (A2, B1, C3). This combination is Car_A's "best solution".
[0076] Round 3 (T=T2~T3ms): Solution announcement and voting Car_A, Car_B, Car_C broadcast their respective "best solutions" simultaneously.
[0077] Vote message structure: { "msg_type": "Vote", "community_id": "uuid-1234-abcd", "source_id":"Car_A", "proposed_solution": {"Car_A": "A2", "Car_B": "B1", "Car_C": "C3"}}.
[0078] Suppose Car_C also computes the same result, broadcasts {"Car_A": "A2", "Car_B": "B1", "Car_C": "C3"}.
[0079] Car_B might compute a different solution, broadcasts {"Car_A": "A2", "Car_B": "2", "Car_C": "1"}.
[0080] Round 4 (T=T3~T4ms): Consensus reached and locked Each car collects all the Vote messages and tallies the votes.
[0081] Car_A finds that solution S1 = (A2, B1, C3) gets two votes from Car_A and Car_C. Solution S2 = (A2, B2, C1) gets one vote from Car_B.
[0082] Suppose the consensus threshold is floor(N / 2) + 1 = floor(3 / 2) + 1 = 2.
[0083] Solution S1's vote count 2 >= 2, consensus reached! All three cars immediately agree on S1 as the final "consensus solution".
[0084] Car_A locks its own trajectory to A2, Car_B locks to B1, and Car_C locks to C3.
[0085] In conjunction with the first aspect, this application provides another possible implementation of the first aspect, wherein the local vehicle controls the vehicle based on the current recommended trajectory information, and the generation of vehicle control information includes: the local vehicle generating a lock execution command based on the current recommended trajectory information. The lock execution command includes the current recommended trajectory information. The local vehicle broadcasts the lock execution command to other vehicles.
[0086] Locking execution and status broadcasting Status switch: All three vehicles switch their status to "Locked".
[0087] Forced execution: The vehicle's underlying control module (such as the MPC controller) begins to precisely track the locked trajectory traj_A2, traj_B1, traj_C3.
[0088] Status Broadcast: During the subsequent execution, the three vehicles will add a status field with the value "Locked" and the community_id to their BSM messages. This informs other vehicles in the area (including newly entered ones) that the convoy is performing cooperative avoidance and should not interfere.
[0089] The negotiation timeout handling includes: if no solution receives enough votes before the T=50ms deadline, the negotiation fails. All vehicles immediately execute the preset, safest "default action," such as braking simultaneously at maximum deceleration. Secondly, this application also provides a multi-vehicle emergency avoidance cooperative control system. This cooperative control system is used for cooperative vehicle control within an area where at least two vehicles are driving and their driving trajectories intersect. (Refer to...) Figure 3 A schematic diagram of the components of a collaborative control system for multi-vehicle emergency avoidance, as shown below. Figure 3 As shown, it includes: a triggering unit 101, a candidate trajectory generation unit 201, a distributed negotiation unit 301, and a trajectory locking unit 401. Among them, Triggering unit 101 is configured to obtain predicted driving trajectory information of the first vehicle and the second vehicle based on vehicle network information of the first vehicle and the second vehicle within the intersection area. Based on the intersection probability of the predicted driving trajectory information of the first vehicle and the second vehicle, trigger the first vehicle or the second vehicle to broadcast a negotiation request to the other.
[0090] The alternative trajectory generation unit 201 is configured to receive the negotiation request broadcast vehicle, and obtain a plurality of alternative low collision probability trajectory information according to the current vehicle dynamics constraint information and the surrounding environment information, and broadcast the plurality of alternative low collision probability trajectory information to other vehicles. The alternative low collision probability trajectory information includes alternative trajectory information and trajectory cost information.
[0091] The distributed negotiation unit 301 is configured to receive the alternative low collision probability trajectory information of the vehicle, and obtain the preferred trajectory information according to the trajectory cost information in the plurality of received alternative low collision probability trajectory information and the plurality of alternative low collision probability trajectory information locally, and broadcast the preferred trajectory information to other vehicles.
[0092] The trajectory locking unit 401 is configured to obtain the current recommended trajectory information according to the received preferred trajectory information and the multiplexing rate of the preferred trajectory information of the local vehicle. The local vehicle generates vehicle control information according to the current recommended trajectory information.
[0093] In combination with the second aspect, the second aspect provides a first possible implementation manner of the second aspect. The Internet of Vehicles information includes vehicle information and environment information in V2X. The vehicle information includes vehicle identification information ID, vehicle position information, vehicle current speed information, vehicle acceleration information, and vehicle heading angle information.
[0094] The alternative trajectory generation unit 201 is further configured to obtain the first vehicle predicted driving trajectory information and the second vehicle predicted driving trajectory information according to the first vehicle and the second vehicle passing through the Internet of Vehicles information in the intersection area, and the step includes: According to the vehicle identification information ID, the vehicle position information, the vehicle current speed information, the vehicle acceleration information, and the vehicle heading angle information, the first vehicle predicted driving trajectory information and the second vehicle predicted driving trajectory information are obtained through a constant steering rate and acceleration model trajectory predictor. The first vehicle predicted driving trajectory information and the second vehicle predicted driving trajectory information include trajectory band information in a future prediction time. The trajectory band information includes trajectory shape and position information and trajectory width information formed by a plurality of trajectories.
[0095] Therefore, the present application can produce the following beneficial effects: 1. Effectively avoid decision conflict and secondary collision: through distributed negotiation, all related vehicles reach an agreement before action, ensuring that the final executed combined trajectory is globally verified and has no internal conflict. This fundamentally solves the problem of secondary collision caused by independent decision.
[0096] 2. Ensures the rapid convergence and stability of decision-making: The limited multi-round negotiation protocol and "locking" mechanism designed in the application ensure that the decision-making process can converge in a very short time (usually within tens of milliseconds), avoiding the problem of infinite "shaking" between multiple options. Once locked, the system enters the deterministic execution phase, showing high stability.
[0097] 3. Achieves the synergistic effect of "Pareto improvement": The negotiation process aims to find a solution that is better for the whole community, not just to satisfy individual optimality. This enables the system to jump out of the "local optimal trap" and find a coordinated avoidance solution that can ensure the safety of all vehicles while considering comfort and efficiency.
[0098] 4. Strong robustness, independent of infrastructure: The entire negotiation process is based entirely on V2V direct communication and does not rely on any central server or roadside MEC. This makes the method effective anywhere with V2V coverage, including tunnels, remote highways, and other places without network coverage, with high reliability and scenario adaptability.
[0099] 5. Clear protocol, easy to standardize: The application defines clear negotiation trigger conditions, message formats (NR, DP), interaction processes, and state machines, providing a solid technical foundation for developing industry-wide coordinated avoidance standards.
[0100] The multi-vehicle emergency avoidance coordination control system provided by the embodiments of the application can be specific hardware on a device or software or firmware installed on a device, etc. The device provided by the embodiments of the application has the same implementation principle and technical effects as the foregoing method embodiments. For brevity and conciseness, the device embodiment part is not mentioned in the foregoing method embodiment. Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the system, device, and unit described above can be referred to the corresponding processes in the foregoing method embodiments, which will not be repeated here.
[0101] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division, and actual implementation can have another division manner. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some communication interfaces, devices or units, and can be electrical, mechanical or other forms.
[0102] The units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, i.e. may be located in one place, or may be distributed on multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment scheme.
[0103] In addition, the functional units in the embodiments provided in the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit.
[0104] The functions, if realized in the form of software functional units and sold or used as independent products, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application can be embodied in the form of software products, and the computer software products are stored in a storage medium, including a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.
[0105] It should be noted that: similar reference numbers and letters represent similar items in the following drawings, therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings, in addition, the terms "first", "second", "third" and the like are only used to distinguish description, and cannot be understood as indicating or implying relative importance.
[0106] Finally, it should be noted that: the above-described embodiments are only specific embodiments of the present application, used to illustrate the technical solutions of the present application, and not to limit them, the protection scope of the present application is not limited thereto, although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any skilled person familiar with the technical field can modify or easily think of changes to the technical solutions recorded in the foregoing embodiments within the technical scope disclosed by the present application, or make equivalent replacement to part of the technical features; and these modifications, changes or replacements do not make the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application. All should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.
Claims
1. A cooperative control method for multi-vehicle emergency avoidance, the cooperative control method being used for cooperative control of vehicles in at least two or more vehicles in driving whose driving trajectories form an intersection area, characterized by, It includes: According to the intersection area within the first vehicle and the second vehicle through the Internet of Vehicles information, the first vehicle prediction driving track information and the second vehicle prediction driving track information are obtained; According to the intersection probability of the first vehicle prediction driving track information and the second vehicle prediction driving track information, the first vehicle or the second vehicle sends negotiation request broadcast to the other party; The vehicle receiving the negotiation request broadcast obtains a plurality of alternative low collision probability track information according to the current vehicle dynamics constraint information and the surrounding environment information, and broadcasts the plurality of alternative low collision probability track information to other vehicles; The alternative low collision probability track information includes alternative track information and track cost information. The vehicle receiving the alternative low collision probability track information obtains the preferred track information according to the track cost information in the received plurality of alternative low collision probability track information and the local plurality of alternative low collision probability track information, and broadcasts the preferred track information to other vehicles. The local vehicle obtains the current recommended track information according to the received preferred track information and the multiplexing rate of the local vehicle's preferred track information. The local vehicle generates vehicle control information according to the current recommended track information control.
2. The method according to claim 1, wherein Also includes, according to the intersection area within the first vehicle and the second vehicle through the Internet of Vehicles information, a plurality of first vehicle prediction driving track information and a plurality of second vehicle prediction driving track information are obtained; According to the intersection probability of the plurality of first vehicle prediction driving track information and the plurality of second vehicle prediction driving track information, the first vehicle or the second vehicle sends negotiation request broadcast to the other party.
3. The method according to claim 1 or 2, wherein The vehicle information in the V2X includes vehicle identification information ID, vehicle position information, vehicle current speed information, vehicle acceleration information, vehicle heading angle information. The step of obtaining the first vehicle prediction driving track information and the second vehicle prediction driving track information according to the intersection area within the first vehicle and the second vehicle through the Internet of Vehicles information includes: According to the vehicle identification information ID, the vehicle position information, the vehicle current speed information, the vehicle acceleration information and the vehicle heading angle information, the first vehicle prediction driving track information and the second vehicle prediction driving track information are obtained through the constant steering rate and acceleration model trajectory predictor; The first vehicle prediction driving track information and the second vehicle prediction driving track information include track band information within the future prediction time; The track band information includes the shape and position information of the track, and the track width information formed by a plurality of tracks.
4. The method of claim 3, wherein, The step of triggering the first vehicle or the second vehicle to send negotiation request broadcast to the other party according to the intersection probability of the first vehicle prediction driving track information and the second vehicle prediction driving track information includes: According to the track band information in the first vehicle prediction driving track information and the track band information in the second vehicle prediction driving track information, the track band coincidence degree within the future set time, the future set time is less than the future prediction time; If the overlap of the trajectory bands is greater than a set overlap threshold, the first vehicle or the second vehicle is triggered to broadcast a negotiation request to the other party. The negotiation request includes: the communication command identifier of the air interface protocol of the vehicle V2X protocol, the communication identifier field of this communication, the initiator identifier of this communication, the list of receiving members of this communication, and the validity period of the message or request.
5. The method of claim 4, wherein, The step of the vehicle receiving the negotiation request broadcast, obtaining multiple candidate low-collision probability trajectory information based on the current vehicle dynamics constraint information and surrounding environment information, and broadcasting the multiple candidate low-collision probability trajectory information to other vehicles includes: The vehicle that receives the negotiation request broadcast verifies whether it is a receiving member according to the list of receiving members for this communication. If the verification is successful, it joins the communication according to the communication identifier field and forms a decision-making community communication group with the vehicle that sent the information. Under the premise that the simulation satisfies the vehicle dynamics and kinematic constraints, multiple candidate low-collision probability trajectory information is generated through a discretized control input forward simulation method. The multiple candidate low-collision probability trajectory information includes candidate trajectory information and trajectory cost information. The trajectory cost information is obtained by the distance between the trajectory endpoint and the original target point, the acceleration and angular acceleration on the trajectory, the minimum distance between the trajectory and static obstacles, and the weight coefficients of each parameter. The current vehicle packages the multiple alternative low-collision probability trajectory information into a decision package and broadcasts it to other vehicles.
6. The method of claim 4, wherein, The step of the vehicle that receives the candidate low-collision probability trajectory information to obtain preferred trajectory information and broadcast the preferred trajectory information to other vehicles, based on the trajectory cost information in the received multiple candidate low-collision probability trajectory information and the local multiple candidate low-collision probability trajectory information, includes: The vehicle that receives the alternative low-collision probability trajectory information obtains the combined cost of the alternative trajectories of other vehicles and the local vehicle based on the trajectory cost information from the multiple alternative low-collision probability trajectory information received and the multiple alternative low-collision probability trajectory information locally; and obtains the preferred trajectory information corresponding to the preferred combined cost value based on the combined cost.
7. The method of claim 1, wherein, The step of obtaining the current recommended trajectory information for the local vehicle based on the received preferred trajectory information and the reuse rate of the preferred trajectory information of the local vehicle includes: The local vehicle obtains the trajectory reuse rate based on the received preferred trajectory information and the preferred trajectory information of the local vehicle. If the reuse rate is greater than a set threshold, the current recommended trajectory information is obtained; otherwise, a forced avoidance information command is generated. The local vehicle generates vehicle control information based on the mandatory avoidance information command.
8. The method of claim 1, wherein, The local vehicle is controlled based on the current recommended trajectory information, and the generated vehicle control information includes: The local vehicle generates a lock execution command based on the current recommended trajectory information; the lock execution command includes the current recommended trajectory information; the local vehicle broadcasts the lock execution command to other vehicles.
9. A cooperative control system for multi-vehicle emergency avoidance, the cooperative control system being used for cooperative control of vehicles in driving of at least two or more vehicles whose driving trajectories form an intersection area, characterized by, It includes: a triggering unit, a candidate trajectory generation unit, a distributed negotiation unit, and a trajectory locking unit; among which, The triggering unit is configured to obtain first vehicle predicted driving track information and second vehicle predicted driving track information according to the first vehicle and the second vehicle passing through the intersection area through vehicle networking information; and trigger the first vehicle or the second vehicle to send a negotiation request broadcast to the other party according to an intersection probability of the first vehicle predicted driving track information and the second vehicle predicted driving track information. The alternative track generation unit is configured to obtain a plurality of alternative low-collision-probability track information according to current vehicle dynamics constraint information and surrounding environment information and broadcast the plurality of alternative low-collision-probability track information to other vehicles; and the alternative low-collision-probability track information includes alternative track information and track cost information. The distributed negotiation unit is configured to obtain preferred track information according to track cost information in the received plurality of alternative low-collision-probability track information and the plurality of alternative low-collision-probability track information locally, and broadcast the preferred track information to other vehicles. The track locking unit is configured to obtain current recommended track information according to the received preferred track information and multiplexing rate of the preferred track information of a local vehicle; and generate vehicle control information according to the current recommended track information.
10. The cooperative control system for multi-vehicle emergency avoidance according to claim 9, wherein, The vehicle networking information includes vehicle information and environment information in V2X; the vehicle information includes vehicle identification information ID, vehicle position information, vehicle current speed information, vehicle acceleration information, and vehicle heading angle information. The alternative track generation unit is further configured to obtain the first vehicle predicted driving track information and the second vehicle predicted driving track information according to the first vehicle and the second vehicle passing through the intersection area through the vehicle networking information, including: obtaining the first vehicle predicted driving track information and the second vehicle predicted driving track information through a constant steering rate and acceleration model track predictor according to the vehicle identification information ID, the vehicle position information, the vehicle current speed information, the vehicle acceleration information, and the vehicle heading angle information; the first vehicle predicted driving track information and the second vehicle predicted driving track information include track band information in a future prediction time; the track band information includes shape and position information of a track, and track width information formed by a plurality of tracks.