Intelligent vehicle ramp confluence decision-making method considering advanced confluence of rear vehicles into main line and vehicles behind main line

Abstract

The invention belongs to the technical field of intelligent vehicles, and particularly relates to an intelligent vehicle ramp confluence decision-making method considering advanced confluence of a rear vehicle into a main line and a vehicle behind the main line, which comprises the following steps: acquiring a road level position, a vehicle speed and a longitudinal acceleration of a vehicle, and if the three meet set conditions, judging that the vehicle enters an active confluence decision-making stage; meanwhile, the vehicle carries out panoramic scanning on an acceleration lane behind the vehicle, and potential cooperative vehicles are screened out according to a set mode; behavior characteristics of the potential cooperative vehicles are further monitored, and when set conditions are met, it is judged that the potential cooperative vehicles have an advanced afflux behavior, and safety evaluation is started immediately; and if the potential cooperative vehicle is judged to be safely converged in advance and the collision time of the own vehicle and the potential cooperative vehicle is greater than or equal to 0.8 s, the own vehicle completes convergence with an acceleration curve greater than that of the potential cooperative vehicle, and otherwise, a conservative convergence strategy is executed. The method provided by the invention can improve the success rate, safety and efficiency of intelligent vehicle afflux.

Description

Technical Field

[0001] This invention belongs to the field of intelligent vehicle technology, specifically relating to an intelligent vehicle ramp merging decision method that considers the early merging of following vehicles into the main line and vehicles behind the main line. In particular, it is a method that triggers or optimizes the merging decision of the vehicle by identifying, evaluating and utilizing the preemptive merging behavior of following vehicles. Background Technology

[0002] In highway ramp merging scenarios, intelligent vehicles need to identify the possible gaps in the target lane (also known as the main line / main line lane / target lane) and merge safely. Traditional merging decision algorithms are mainly based on predicting the motion state of vehicles in the target lane (i.e., the vehicles in front and behind in the main line lane) and determining the merging opportunity by calculating indicators such as the time to collision (TTC). For example, Chinese invention patent application number 202010075592.9, entitled "A Risk Decision-Making Device and Method for Vehicles Autonomously Merging into the Main Road from the Acceleration Lane," discloses that it calculates the real-time merging risk level based on the remaining distance of the ramp vehicle from the end of the acceleration lane, the real-time speed of the ramp vehicle, the speed of vehicles behind the vehicle on the main road, the distance between the vehicle and the vehicles behind, and the angle between the vehicle and the vehicles behind, and decides the merging method of the vehicle based on the real-time merging risk level.

[0003] However, the merging algorithms for intelligent vehicles tend to be conservative overall, often favoring conservative merging decisions in scenarios with low merging risk. This negatively impacts the overall merging success rate. For example, in heavy traffic, existing methods may fail to provide a sufficiently large safety gap in the target lane for extended periods, forcing the vehicle to slow down or even stop at the end of the acceleration lane, severely affecting traffic efficiency and safety. Furthermore, traditional algorithms treat other participants in the traffic environment as independent or even opposing entities, ignoring potential interactions between vehicles. For instance, when an intelligent vehicle merges with other vehicles in the acceleration lane, other vehicles behind it may sometimes merge prematurely. In this case, since the rear vehicle has already completed its merging process and risk assessment of vehicles behind it in the main lane, and has executed its merging, its premature merging, assuming no significant conflict in its decision-making process, effectively creates a physical obstruction or psychological deterrent for vehicles in the main lane. This provides the intelligent vehicle with a brief but effective merging window, helping it merge into the main lane earlier and avoiding merging failures. Existing technologies are completely unable to identify and utilize this valuable "non-active cooperation signal," and cannot further improve the success rate of autonomous merging of intelligent vehicles.

[0004] In view of this, the present invention is hereby proposed. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and propose an intelligent vehicle ramp merging decision method that considers the early merging of following vehicles into the main line and vehicles behind the main line. This method can sense and understand the "demonstrative" preemptive merging behavior of following vehicles, regard this behavior as a high-value positive cooperation signal, and actively and safely trigger or optimize the merging decision of the vehicle based on this signal, thereby improving the success rate, safety and efficiency of merging.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: This invention provides an intelligent vehicle ramp merging decision-making method that considers vehicles merging into the main line ahead of others and vehicles behind the main line. The method includes acquiring the vehicle's road-level position, speed, and longitudinal acceleration. If all three parameters meet preset conditions, the vehicle is determined to have entered the active merging decision-making stage. Simultaneously, the vehicle activates millimeter-wave radar and visual sensors to perform a panoramic scan of the acceleration lane behind it, filtering potential partner vehicles according to a preset method. The behavioral characteristics of potential partner vehicles are further monitored. When preset conditions are met, it is determined that a potential partner vehicle is merging ahead of others, and a safety assessment is immediately initiated. If the potential partner vehicle is determined to be merging safely ahead of others, and the collision time between the vehicle and the potential partner vehicle is greater than or equal to 0.8 seconds, the vehicle completes the merge with an acceleration curve greater than that of the potential partner vehicle; otherwise, a conservative merging strategy is executed. Among them, potential cooperating vehicles are the nearest neighboring vehicles in the same lane behind the vehicle and whose movement trajectories are related to the vehicle's own.

[0007] Furthermore, when the vehicle simultaneously meets the following conditions: its road-level position is in the acceleration lane, its speed is greater than 40 km / h, and its longitudinal acceleration is greater than 0 m / s², 2 At that time, the vehicle system determines that the vehicle has entered the active merging decision-making stage.

[0008] Furthermore, the method for screening potential partner vehicles is as follows: Millimeter-wave radar and vision sensors located on the left rear and directly behind the vehicle perform a panoramic scan of the acceleration lane within a range of 50-200 meters behind the vehicle, locking onto objects behind the vehicle, within the same lane, and at a distance from the vehicle. d relative Satisfy: 20m≤ d relative The nearest vehicle within a range of ≤100m.

[0009] Furthermore, potential cooperating vehicles are continuously monitored at a frequency of 30~100Hz. When all of the following conditions are met simultaneously, the onboard system determines that a potential cooperating vehicle has prematurely merged into the main line: Lateral acceleration of potential cooperative vehicles ay BV >0.3m / s 2The front of the vehicle has already crossed the center line of the target lane on the main line.

[0010] Furthermore, the security assessment method is as follows: The onboard system uses a side-rear millimeter-wave radar to track vehicles (MVs) behind in the target lane on the main line and calculates the time to collision (TTC) between potential collaborating vehicles and MVs, and conducts a safety assessment based on the time to collision (TTC). The formula for calculating TTC is: in, d relative , mv ( t Let t be the longitudinal relative distance between the potential cooperating vehicle and the MV. v relative , mv ( t Let t be the longitudinal relative velocity of the two vehicles at time t.

[0011] Furthermore, during the process of the potential cooperating vehicle moving laterally from the start of its movement to cross the center line of the main target lane, if the potential cooperating vehicle and the following vehicle in the main target lane always satisfy the collision time TTC(t)>3s, it is considered to be a safe early merging; otherwise, the vehicle ignores this behavior and executes a conservative merging strategy.

[0012] Furthermore, the process of the vehicle merging by referring to the acceleration curve of the potential cooperating vehicle is as follows: the on-board system acquires the longitudinal acceleration sequence of the potential cooperating vehicle during the time period from the start of its lateral movement to the moment the front of the vehicle crosses the center line of the target lane on the main line. And calculate the average longitudinal acceleration during its merging process. Simultaneously, the vehicle's trajectory planner generates the same merging path by referencing the motion trajectories of potential collaborating vehicles, and the vehicle's trajectory is greater than... The acceleration, using the merging path as the route, accelerates and merges into the target lane of the main line.

[0013] Furthermore, the average longitudinal acceleration The calculation formula is as follows: in, t 1 represents the time when potential partner vehicles begin lateral movement. t 0 represents the time it takes for the front of a potential partner vehicle to cross the center line of the target lane on the main line.

[0014] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention is the first to use "behavior of following vehicles" as the core decision input, which can discover and utilize "hidden" merging opportunities that traditional algorithms cannot identify, significantly reducing the waiting time of vehicles at the end of the ramp, improving the traffic efficiency of the merging area, avoiding traffic congestion, making the decision logic of the intelligent driving system closer to that of experienced human drivers, making the behavior more predictable and cooperative, improving the driving experience and the predictability of surrounding vehicles.

[0015] 2. This invention is highly feasible and applicable, and has low integration costs. Specifically, the method of this invention is mainly based on existing vehicle sensors (cameras, radar, vision sensors) and computing platforms, without relying on mandatory V2X communication, and is easy to integrate and deploy on existing intelligent driving systems. Attached Figure Description

[0016] The accompanying drawings are incorporated in and form part of this specification, and together with the description serve to explain the principles of the invention.

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of the intelligent vehicle ramp merging decision method of the present invention; Figure 2 This is a schematic diagram provided by the present invention showing the autonomous vehicle in the active merging decision-making stage and with potential cooperating vehicles present. Figure 3 A schematic diagram illustrating the potential early merging behavior of a cooperative vehicle, as provided by this invention; Figure 4 This is a schematic diagram illustrating the cooperative merging behavior between the vehicle and the following vehicle provided by the present invention. Detailed Implementation

[0019] Exemplary embodiments will now be described in detail. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples consistent with some aspects of the invention as detailed in the appended claims.

[0020] The vehicle system (also referred to as the system) on which this embodiment relies includes an autonomous driving domain controller, vehicle sensors (including forward, rear and side millimeter-wave radar and vision sensors (groups) including but not limited to surround view cameras and forward / rear cameras), vehicle high-precision map database, vehicle GPS / IMU (inertial measurement unit) integrated navigation unit, drive-by-wire system, brake-by-wire system, and steering-by-wire system; The autonomous driving domain controller includes a perception fusion module and a decision planning module connected to it. The visual sensors transmit the monitored data to the perception fusion module via a high-speed communication link (such as Ethernet). The perception fusion module transmits the calculated data (based on the vehicle system's internal communication bus (such as CAN bus, Ethernet bus, or the domain controller's internal high-speed data bus)) to the decision planning module. The decision planning module includes a behavior decision unit and a trajectory planner, which work together to generate control commands and are connected to the drive-by-wire system, brake-by-wire system, and steering-by-wire system via the CAN bus. The perception fusion module uses a post-fusion algorithm to comprehensively calculate the perception results processed independently by various sensors such as radar, and employs algorithms such as Kalman filtering for target tracking and state estimation.

[0021] The decision planning module is used to determine whether the vehicle has entered the active merging decision stage, whether it needs to activate the millimeter-wave radar and vision sensor, filter potential cooperating vehicles according to the set method, determine whether potential cooperating vehicles have engaged in early merging behavior, initiate a safety assessment, determine whether potential cooperating vehicles have safely merged in advance, calculate the collision time between the vehicle and potential cooperating vehicles, and determine whether to allow the vehicle to merge into the main line with the planned path and acceleration (i.e., determine whether to activate the merging strategy of this invention or the conservative merging strategy), etc., which are the parts that need to be calculated and determined in this embodiment.

[0022] The specific data flow is as follows: data from vehicle sensors (millimeter-wave radar / surround-view camera) is transmitted to the perception fusion module for processing via the CAN bus, and then the decision planning module makes judgments and generates commands to control systems such as drive-by-wire, brake-by-wire, and steering-by-wire.

[0023] The parameters required for collection by the vehicle-mounted system and their sources are as follows: This includes matching the vehicle's GPS receiver with an in-vehicle high-precision map database to accurately obtain the road-level location of the vehicle as it is traveling; The inertial measurement unit (IMU) and the vehicle's CAN bus are used to acquire dynamic parameters such as the vehicle's longitudinal velocity, acceleration, and heading angle in real time.

[0024] The parameters of the cooperating vehicles behind include the distance, relative speed and azimuth of the vehicles behind, which are accurately detected by rearward millimeter-wave radar. The lateral acceleration is calculated based on the radar data and combined with Kalman filtering. The visual sensor simultaneously uses precise perception of lane lines to help determine the relative position of vehicles behind the vehicle in the same lane as the vehicle itself, in order to confirm whether they are in the same acceleration lane. The vehicle uses surround-view cameras to detect lane lines and determine whether it has crossed the lane lines.

[0025] Parameters of vehicles behind the main line, including relative distance and relative speed obtained through side-rear millimeter-wave radar.

[0026] The parameters required by the decision planning module include the time to collision (TTC) calculated based on radar data, the longitudinal acceleration sequence of potential cooperating vehicles that are continuously tracked and recorded, and the merging path generated by the trajectory planner, which is sent to the drive-by-wire system via the CAN bus as control commands.

[0027] The fusion and processing of the aforementioned multi-source sensor data provides a complete information foundation for the entire collaborative decision-making process. Example

[0028] Please see Figures 1-4 This embodiment provides an intelligent vehicle ramp merging decision method that considers vehicles merging into the main line in advance and vehicles behind the main line, including the following steps: The specific implementation steps of the method are as follows: Step 1: Import Scene Activation and Vehicle Status Judgment This step is a prerequisite for the method's initiation. First, the vehicle's GPS receiver is matched with a high-precision map database to accurately confirm that the vehicle is traveling in the acceleration lane of the highway. Simultaneously, the inertial measurement unit (IMU) and the vehicle's CAN bus are used to acquire dynamic parameters such as the vehicle's longitudinal velocity, acceleration, and heading angle in real time. When the decision-making and planning module determines that the vehicle is in the acceleration lane and its speed and acceleration indicate an active pursuit of merging opportunities, the subsequent cooperative decision-making process is formally activated. This ensures that the entire method only initiates in the correct road scenario and when the vehicle is actively seeking a merging opportunity, avoiding invalid or erroneous calculations at inappropriate times.

[0029] Specifically, when the vehicle's CV speed meets the requirements vego CV >40km / h and longitudinal acceleration meets aego CV >0m / s 2 The decision planning module determines that the vehicle has entered the "active merging decision stage" and then activates the identification of rear cooperating vehicles to screen out potential cooperating vehicles.

[0030] Step Two: Rear Collaborating Vehicle Recognition In this step, the vehicle system uses the fusion of radar and visual perception data from numerous traffic participants to filter out the nearest neighboring vehicles that are in the same lane behind the vehicle and whose movement trajectories are related. These vehicles are then officially identified as "potential cooperating vehicles," laying the foundation for subsequent fine-grained identification of their specific behaviors.

[0031] Specifically, the decision-making and planning module activates the millimeter-wave radar and vision sensors on the left rear and directly behind the vehicle to scan the acceleration lane within a 150-meter range behind the vehicle. Using DBSCAN target clustering and Kalman filter tracking algorithms, it identifies targets behind the vehicle, within the same lane, and at a distance from the vehicle. d relative Satisfying 20m≤ d relative The nearest vehicle within a ≤100m range (BV, e.g.) Figure 2 The vehicle is identified as a "potential partner vehicle" and continuously tracked. If no potential partner vehicle is identified, the original conservative inbound strategy (traditional inbound decision) is executed.

[0032] It should be noted that millimeter-wave radar returns data from hundreds or thousands of point clouds. DBSCAN target clustering works by grouping point clouds that are physically close and have similar velocity characteristics into a single object. In other words, it turns points into "vehicles".

[0033] Kalman filtering can predict a target's state in the next moment based on its position and velocity in the previous moment. By matching the predictions with the observations in consecutive frames, the system can assign a unique ID to this "clustered object." Among all tracked IDs, the system uses lane information provided by a high-precision map to filter out the closest targets that are "in the same lane," have a longitudinal distance of "20m~100m," and are the closest to the target. In other words, Kalman filtering is responsible for "numbering" the vehicle and continuously monitoring it.

[0034] Step 3: Rear vehicle merging in advance - behavior recognition This step aims to accurately capture the initial intent of a potential collaborating vehicle to merge. By continuously tracking the vehicle and utilizing multi-dimensional information such as sudden increases in lateral displacement and changes in heading angle, the onboard sensors determine that the vehicle is not simply deviating from its lane, but rather performing a deliberate "pre-merge" maneuver. The purpose of this identification is to promptly and accurately discover this crucial opportunity for collaboration.

[0035] Specifically, the decision-making and planning module continuously monitors the behavioral characteristics of potential partner vehicles (BVs) at a frequency of 50Hz. The system determines that a vehicle has begun premature merging when all of the following conditions are met simultaneously: The lateral acceleration of the vehicle was calculated using radar data. ay BVand satisfy ay BV >0.3m / s 2 The direction is pointing towards the target lane.

[0036] Simultaneously, based on the lane line detection results from the visual sensor and the radar target coordinates, it is confirmed that the front of the vehicle has crossed the center line of the main lane (e.g., Figure 3 ).

[0037] Step 4: Safety Assessment of Potential Partner Vehicle Merging Behavior Upon detecting an unauthorized merging attempt, the decision-making and planning module immediately initiates a dynamic safety assessment. At this point, the onboard sensors shift their focus to the vehicle behind in the target lane. Using a side-facing millimeter-wave radar, the module continuously calculates the real-time time-to-collision (TTC) between the potential merging vehicle (BV) and the following vehicle (MV) in the main lane, verifying whether the merging was successfully completed without posing an imminent collision with the following vehicle (MV). Only merging demonstrations confirmed as "safe and successful" are adopted by the system as valid and trustworthy cooperation signals.

[0038] Specifically, once a potential collaborating vehicle is detected to be merging ahead of schedule, the decision-making and planning module immediately initiates a safety assessment. It uses a side-rear millimeter-wave radar to track the vehicle MV behind it in the target lane of the main line, calculates the time-to-collision (TTC) between the potential collaborating vehicle BV and MV, and performs a safety assessment based on the TTC. The formula for calculating TTC is: in, d relative , mv ( t Let t be the longitudinal relative distance between the potential cooperating vehicle and the MV. v relative , mv ( t Let t be the longitudinal relative velocity of the two vehicles at time t.

[0039] If, during the process of the potential cooperating vehicle BV moving laterally from the moment it begins to cross the center line of the target lane, the collision time TTC(t) between the potential cooperating vehicle BV and the following vehicle MV in the target lane always satisfies TTC(t) > 3s, then it is considered a safe and successful early merging. Conversely, if, during the merging process, the calculated TTC(t) ≤ 3s at any moment, the decision planning module determines that BV's decision is reckless or erroneous, this cooperation signal is invalid, and the vehicle will ignore this behavior and continue to execute the original conservative merging strategy (the original conservative merging strategy adopts the merging strategy mentioned in the background art, entitled "A Risk Decision-Making Device and Method for Vehicles Autonomously Merging into the Main Road from the Acceleration Lane").

[0040] Step 5: Autonomous Vehicle Collaboration Injection Decision Generation Once the potential partner vehicle (BV)'s premature merging behavior passes safety verification, the decision-making and planning module will logically infer that the potential partner vehicle (BV) has already merged into the target lane on the main line, and its physical presence constitutes a physical obstruction or psychological deterrent to vehicles following on the main line, thus creating a brief but effective merging window for the vehicle. Based on this reasoning, the decision-making and planning module, while ensuring that the collision time to collision (TTC) between the vehicle and the potential partner vehicle (BV) meets the safe dynamic time difference of 0.8 seconds, breaks through the limitations of the original conservative merging strategy and immediately generates an "Immediately Execute Partner Merging" command. This command is sent to the vehicle's drive-by-wire system via the vehicle's CAN bus, triggering the merging action. In short, it transforms favorable changes in the external environment into a decision-making advantage for the vehicle, decisively seizing the merging opportunity.

[0041] Step Six: Collaborative Merge Trajectory and Velocity Planning Simultaneously with issuing the merging command, the decision-making and planning module enters the refined trajectory and velocity planning stage. The trajectory planner references the spatial path adopted by a successfully merging BV and its longitudinal acceleration characteristics throughout the critical merging period, planning a smooth merging path for the autonomous vehicle. This path is then sent to the drive-by-wire system via the CAN bus, driving the autonomous vehicle to complete the merging with an acceleration curve greater than that of the BV. This ensures seamless dynamic coordination between the autonomous vehicle's merging behavior and the BV, guaranteeing not only safety but also improving the smoothness of the merging process and the overall efficiency of traffic flow.

[0042] Specifically, during the time period from when the BV begins to move laterally to when the front of the vehicle crosses the center line of the target lane. T merge =[ t 0, t Within [1], the decision planning module obtains its longitudinal acceleration sequence. And calculate the average longitudinal acceleration during its merging process. : Simultaneously, the vehicle's trajectory planner generates the same merging path by referencing the motion trajectories of potential collaborating vehicles, and the vehicle's trajectory is greater than... The acceleration, using the merging path as the route, accelerates into the target lane of the main line (e.g. Figure 4 This minimizes disruption to mainline traffic and ensures passenger comfort.

[0043] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention.

[0044] It should be understood that the present invention is not limited to the content already described above, and various modifications and changes can be made without departing from its scope. The scope of the present invention is limited only by the appended claims.

Claims

1. A smart vehicle ramp merging decision method that considers vehicles merging into the main line in advance and vehicles behind the main line, characterized in that, This includes acquiring the vehicle's road-level position, speed, and longitudinal acceleration. If all three conditions are met, the vehicle is determined to have entered the active merging decision-making stage. At the same time, the vehicle activates millimeter-wave radar and vision sensors to perform a panoramic scan of the acceleration lane behind it and select potential cooperating vehicles according to the set method. Further monitor the behavioral characteristics of potential partner vehicles. When the set conditions are met, determine that the potential partner vehicle has engaged in premature merging behavior and immediately initiate a safety assessment. If it is determined that the potential cooperating vehicle is safely merging in, and the collision time between the vehicle and the potential cooperating vehicle is greater than or equal to 0.8s, then the vehicle completes the merging with an acceleration curve greater than that of the potential cooperating vehicle; otherwise, a conservative merging strategy is executed. Among them, potential cooperating vehicles are the nearest neighboring vehicles in the same lane behind the vehicle and whose movement trajectories are related to the vehicle's own.

2. The intelligent vehicle ramp merging decision method according to claim 1, characterized in that, When the vehicle simultaneously meets the following conditions: its road-level position is in the acceleration lane, its speed is greater than 40 km / h, and its longitudinal acceleration is greater than 0 m / s² 2 At that time, the vehicle system determines that the vehicle has entered the active merging decision-making stage.

3. The intelligent vehicle ramp merging decision method according to claim 1, characterized in that, The method for screening potential partner vehicles is as follows: Millimeter-wave radar and vision sensors located on the left rear and directly behind the vehicle perform a panoramic scan of the acceleration lane within a range of 50-200 meters behind the vehicle, locking onto objects behind the vehicle, within the same lane, and at a distance from the vehicle. d relative Satisfy: 20 m≤ d relative The nearest vehicle within a range of ≤100 m.

4. The intelligent vehicle ramp merging decision method according to claim 1, characterized in that, The system continuously monitors potential cooperating vehicles at a frequency of 30-100Hz. When all of the following conditions are met simultaneously, the onboard system determines that a potential cooperating vehicle has prematurely merged into the main line: Lateral acceleration of potential cooperative vehicles ay BV > 0.3m / s 2 Meanwhile, the front of the vehicle has crossed the center line of the target lane on the main line.

5. The intelligent vehicle ramp merging decision method according to claim 1, characterized in that, The security assessment method is as follows: The onboard system uses a side-rear millimeter-wave radar to track vehicles (MVs) behind in the target lane on the main line and calculates the time to collision (TTC) between potential collaborating vehicles and MVs, and conducts a safety assessment based on the time to collision (TTC). The formula for calculating TTC is: in, d relative , mv ( t Let t be the longitudinal relative distance between the potential cooperating vehicle and the MV. v relative , mv ( t Let t be the longitudinal relative velocity of the two vehicles at time t.

6. The intelligent vehicle ramp merging decision method according to claim 5, characterized in that, If, during the process of a potential cooperating vehicle moving laterally from the moment it begins to cross the center line of the target lane, the collision time TTC(t) between the potential cooperating vehicle and the vehicle behind it in the target lane always satisfies MV > 3s, then it is considered a safe early merging; otherwise, the vehicle ignores this action and executes a conservative merging strategy.

7. The intelligent vehicle ramp merging decision method according to claim 1, characterized in that, The process of the merging vehicle completing the merging with an acceleration curve greater than that of the potential cooperating vehicle is as follows: the on-board system acquires the longitudinal acceleration sequence of the potential cooperating vehicle during the time period from the start of its lateral movement to the moment its front crosses the center line of the target lane. And calculate the average longitudinal acceleration during its merging process. Simultaneously, the vehicle's trajectory planner generates the same merging path by referencing the motion trajectories of potential collaborating vehicles, and the vehicle's trajectory is greater than... The acceleration, using the merging path as the route, accelerates and merges into the target lane of the main line.

8. The intelligent vehicle ramp merging decision method according to claim 7, characterized in that, The average longitudinal acceleration The calculation formula is as follows: in, t 0 represents the time when the potential partner vehicle begins to move laterally. t 1 represents the time it takes for the front of a potential partner vehicle to cross the center line of the target lane on the main line.

Images