Vehicle lane change control method and related device

Abstract

The application discloses a vehicle lane changing control method and related equipment, and relates to the technical field of vehicle control. The method comprises the following steps: if the remaining distance between a target vehicle and the latest lane changing point is less than a preset distance threshold, the target vehicle is subjected to deceleration control based on the remaining distance; in the deceleration control process, whether a safety gap meeting a preset safety condition exists is determined based on obstacle operation information in a lane to be merged into; if the safety gap does not exist, the target vehicle is further subjected to deceleration according to a target deceleration parameter, so as to build the safety gap meeting the preset safety condition; and a lane changing path of the target vehicle from a current lane to the lane to be merged into is planned based on the longitudinal interval of the safety gap, so that the target vehicle completes lane changing. The application can realize safe, continuous and efficient lane changing of a vehicle in a congestion scenario, thereby improving the lane changing success rate within a limited distance.

Description

Technical Field

[0001] This application relates to the field of vehicle control technology, and more specifically, to a vehicle lane change control method and related equipment. Background Technology

[0002] With the continuous development of intelligent driving technology, autonomous vehicles are increasingly being used in complex traffic scenarios such as urban roads, highways, and ramp merging. Lane change control, as a crucial component of intelligent driving systems for path planning and behavioral decision-making, directly impacts vehicle safety, continuity, and traffic efficiency. In real-world road environments, vehicles frequently face complex conditions such as congested traffic, limited target lane space, and limited lane change distances. This is especially true in scenarios such as approaching lane closures, merging, or exit changes, where vehicles need to complete safe and efficient lane changes within a limited longitudinal distance. Therefore, research on lane change control methods for complex traffic scenarios is of great significance.

[0003] In related technologies, lane change control typically relies on the vehicle's own motion state and the available space in the target lane to make decisions, executing a lane change operation only after detecting a available lane change gap. However, existing methods often passively wait for the target lane to naturally form a safe gap, which can easily lead to problems such as insufficient remaining distance and limited speed adjustment space when the vehicle approaches the end of the lane change. Furthermore, when the target lane is congested, if a gap meeting the safety conditions cannot be detected in time, the vehicle is prone to lane change failures or repeated lane change attempts. In other words, related technologies suffer from low efficiency in utilizing safe gaps during vehicle lane change control, low lane change success rates, and insufficient adaptability to complex congested scenarios. Summary of the Invention

[0004] In the summary section of this application, the relevant technical solutions are described in general terms, and a series of simplified concepts are introduced. These concepts will be further elaborated in the detailed embodiments section. This summary section should not be construed as limiting the key or essential technical features of the claimed solutions, nor is it intended to limit the scope of protection of the claimed solutions.

[0005] The vehicle lane change control method and related equipment provided in this application can achieve safe, continuous and efficient lane changes for vehicles in congested scenarios by decelerating in advance based on the latest lane change point, dynamically searching or actively constructing a safety gap, and combining the safety gap constraint for path planning, thereby improving the success rate of lane changes within a limited distance.

[0006] In a first aspect, this application provides a vehicle lane change control method, comprising: if the remaining distance between a target vehicle and the latest lane change point is less than a preset distance threshold, then decelerating the target vehicle based on the remaining distance; during the deceleration control process, determining whether there is a safe gap that meets preset safety conditions based on obstacle movement information in the lane to be merged; if there is no safe gap, controlling the target vehicle to further decelerate according to a target deceleration parameter to construct a safe gap that meets the preset safety conditions; and planning a lane change path for the target vehicle to laterally cut into the lane to be merged from the current lane based on the longitudinal interval of the safe gap, so as to control the target vehicle to complete the lane change.

[0007] In some implementations, the deceleration control of the target vehicle based on the remaining distance includes: determining a target deceleration based on the remaining distance and the current speed of the target vehicle; if the remaining distance is greater than a target distance threshold, generating a first acceleration limit command based on the target deceleration and a preset comfort deceleration upper limit, wherein the lower limit of acceleration limited by the first acceleration limit command is the larger of the target deceleration and the preset comfort deceleration upper limit; if the remaining distance is less than or equal to the target distance threshold and the absolute value of the target deceleration is less than a preset hard stop threshold, generating a hard stop line command, wherein the hard stop line command is used to instruct the target vehicle to stop at a preset stopping buffer distance from the latest lane change point; when the remaining distance is less than or equal to the target distance threshold and the absolute value of the target deceleration is greater than or equal to the preset hard stop threshold, generating a second acceleration limit command, wherein the second acceleration limit command is used to instruct the target vehicle to continue decelerating at the target deceleration.

[0008] In some implementations, if the remaining distance between the target vehicle and the latest lane change point is less than a preset distance threshold, then decelerating the target vehicle based on the remaining distance includes: acquiring the current speed of the target vehicle and the obstacle movement information in the lane to be merged into; classifying and identifying merging endpoint scenarios, stationary obstacle scenarios, and ordinary congestion scenarios based on the current speed, the longitudinal position of the latest lane change point, and the obstacle movement information; if any scenario is identified and the remaining distance is less than the preset distance threshold, then decelerating the target vehicle based on the remaining distance.

[0009] In some implementations, the classification and identification of merging endpoint scenarios, stationary obstacle scenarios, and ordinary congestion scenarios based on the current vehicle speed, the longitudinal position of the latest lane change point, and the obstacle operation information includes: when the congestion scenario identification status of the previous judgment period is a non-congestion state, determining the entry threshold set as the lag threshold set; when the congestion scenario identification status of the previous judgment period is a congestion state, using the exit threshold set as the lag threshold set, wherein the value of each threshold in the exit threshold set is greater than the corresponding value in the entry threshold set; and classifying and identifying merging endpoint scenarios, stationary obstacle scenarios, and ordinary congestion scenarios based on the lag threshold set, the current vehicle speed, the longitudinal position of the latest lane change point, and the obstacle operation information.

[0010] In some implementations, the classification and identification of merging endpoint scenarios, stationary obstacle scenarios, and ordinary congestion scenarios based on the lag threshold set, the current vehicle speed, the longitudinal position of the latest lane change point, and the obstacle operation information includes: determining that the merging endpoint scenario identification is successful when the longitudinal distance between the latest lane change point and the target vehicle is less than a first preset buffer distance and the current vehicle speed is less than a first preset speed threshold; determining that the stationary obstacle scenario identification is successful when there is a stationary obstacle with a longitudinal speed less than a preset stationary threshold in the lane to be merged, the longitudinal distance between the stationary obstacle and the target vehicle is less than a safe following distance determined based on the current vehicle speed, and the current vehicle speed is less than a second preset speed threshold; determining that the ordinary congestion scenario identification is successful when the global congestion indicator is valid, the speed of the vehicle in front in this lane is less than a preset slow threshold and the following distance is less than a preset following distance, and the current vehicle speed is less than a third preset speed threshold; wherein, the values ​​of the first preset speed threshold, the second preset speed threshold, and the third preset speed threshold in the entry threshold set are respectively less than the corresponding values ​​in the exit threshold set.

[0011] In some implementations, determining whether a safe gap meeting preset safety conditions exists based on obstacle movement information within the lane to be merged includes: selecting a set of target obstacles located within the lane to be merged from the obstacles within the lane to be merged; sorting the obstacles in the target obstacle set according to their longitudinal positions to determine a reference obstacle, wherein the reference obstacle is the target obstacle in the target obstacle set whose longitudinal position is closest to the target vehicle; generating candidate following gaps and candidate leading gaps based on the longitudinal intervals of the reference obstacle, wherein the candidate following gap is the longitudinal space between the tail of the adjacent obstacle in front of the reference obstacle and the head of the reference obstacle, and the candidate leading gap is the longitudinal space between the tail of the reference obstacle and the head of the adjacent obstacle behind the reference obstacle; and performing safety assessments on the candidate following gaps and the candidate leading gaps respectively to determine whether a safe gap meeting the preset safety conditions exists.

[0012] In some implementations, the step of performing safety assessments on the candidate following gap and the candidate leading gap to determine whether a safe gap exists that meets the preset safety conditions includes: generating multiple acceleration samples for each candidate gap using a preset acceleration sampling set; performing longitudinal kinematic predictions on the target vehicle at multiple preset prediction times based on each acceleration sample to obtain the collision time between the target vehicle and the corresponding front and rear obstacles at each preset prediction time; identifying candidate gaps whose collision time at all preset prediction times is not less than a preset safe collision time threshold as candidate safe gaps; selecting the acceleration sample with the smallest absolute value from the acceleration samples corresponding to the candidate safe gaps as the target acceleration; and determining the candidate safe gap corresponding to the target acceleration as a safe gap that meets the preset safety conditions.

[0013] In some implementations, controlling the target vehicle to further decelerate according to a target deceleration parameter to construct a safe gap that meets the preset safety conditions includes: determining a minimum additional deceleration based on the longitudinal velocity of the reference obstacle, constrained by a preset safe collision time threshold; determining the larger of the minimum additional deceleration and a preset comfort deceleration upper limit as the target deceleration parameter; predicting the longitudinal positions of the target vehicle and the reference obstacle in the merging lane based on the target deceleration parameter to obtain a first predicted longitudinal position of the target vehicle and a second predicted longitudinal position of the reference obstacle after a preset construction time; constructing a candidate gap interval based on the first predicted longitudinal position and the second predicted longitudinal position, combined with a preset safety buffer distance; and determining the candidate gap interval as a safe gap that meets the preset safety conditions when the longitudinal length of the candidate gap interval is greater than a preset gap length.

[0014] Secondly, this application also provides a vehicle lane change control device, comprising: a deceleration control unit, configured to decelerate the target vehicle based on the remaining distance if the remaining distance between the target vehicle and the latest lane change point is less than a preset distance threshold; a gap judgment unit, configured to determine whether a safe gap meeting preset safety conditions exists based on obstacle movement information in the lane to be merged during the deceleration control process; a gap construction unit, configured to control the target vehicle to further decelerate according to target deceleration parameters to construct a safe gap meeting the preset safety conditions if the safe gap does not exist; and a lane change control unit, configured to plan a lane change path for the target vehicle to laterally cut into the lane to be merged from the current lane based on the longitudinal interval of the safe gap, so as to control the target vehicle to complete the lane change.

[0015] Thirdly, this application also provides a vehicle, including: a memory and a processor, the processor being configured to execute a computer program stored in the memory to implement the steps of the vehicle lane change control method described in the first aspect.

[0016] Fourthly, this application also provides a computer-readable storage medium storing computer-executable instructions or a computer program, which, when executed by a processor, implement the steps of the vehicle lane change control method described in the first aspect.

[0017] Fifthly, this application also provides a computer program product, including a computer program or computer-executable instructions, which, when executed by a processor, implement the steps of the vehicle lane change control method provided in the embodiments of this application.

[0018] In summary, this application determines the remaining distance between the target vehicle and the latest lane-change point, and initiates deceleration control in advance when the remaining distance is less than a preset distance threshold. This transforms the lane-change decision from a passive response near the lane-change endpoint to advance planning based on distance constraints. Since the vehicle can reserve deceleration and gap search time before reaching the latest lane-change point, it avoids the problem of insufficient longitudinal adjustment space due to entering the lane-change area too late, thereby improving the feasibility of completing a lane change within the limited distance. During deceleration control, the application further combines the operational information of obstacles in the lane to be merged with the determination of whether a safe gap exists that meets safety conditions. This means that the lane-change process no longer relies solely on the naturally formed available space between vehicles, but rather selects a suitable merging timing through real-time analysis of the target lane's traffic status. Because the safe gap determination is performed simultaneously with the deceleration process, the vehicle can continuously evaluate lane-change conditions during speed adjustment, thereby improving the accuracy of safe gap identification. The system improves the stability of lane-changing decisions. When no safe gap exists that meets safety conditions, the system controls the target vehicle to decelerate further according to the target deceleration parameters, allowing the vehicle's motion to actively influence the surrounding traffic space distribution, thereby creating a safe gap that meets safety conditions. Compared to simply waiting for a natural gap to appear in the target lane, this system can create lane-changing conditions by adjusting the vehicle's own speed, reducing the probability of lane-changing failure due to congestion in the target lane, and avoiding unstable behavior caused by the vehicle continuously attempting to change lanes when there is no effective gap. After obtaining a safe gap, the system plans the lane-changing path for the vehicle to laterally enter the lane to be merged from the current lane based on the longitudinal interval corresponding to the safe gap. This combines lane-changing path planning with safe gap constraints. Since the safe gap serves as the spatial basis for the vehicle's lateral entry, the vehicle can complete the lane change within the range that meets safety conditions, reducing repeated adjustments during the lane-changing process and improving the continuity, safety, and ability to merge into the scene through congestion. In summary, the vehicle lane change control method provided in this application can achieve safe, continuous and efficient lane changes in congested scenarios by decelerating in advance based on the latest lane change point, dynamically searching or actively constructing safety gaps, and combining safety gap constraints for path planning, thereby improving the success rate of lane changes within a limited distance. Attached Figure Description

[0019] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit this specification. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 A schematic flowchart illustrating a vehicle lane change control method provided in an embodiment of this application; Figure 2This is a schematic diagram of the composition structure of a vehicle lane change control device provided in an embodiment of this application; Figure 3 This is a schematic diagram of the composition structure of a vehicle provided in an embodiment of this application. Detailed Implementation

[0020] The terms used in the specification, claims, and drawings of this application, such as "first," "second," "third," "fourth," etc. (if any), are used to distinguish similar objects and not to describe a specific order or sequence. Therefore, it is to be understood that these terms can be used interchangeably where appropriate, allowing the described embodiments to be used in different orders, unless specifically required by the illustrations or description. Furthermore, the terms "is" and "has," and any variations thereof, are intended to cover, non-exclusively, all possible constituent elements. For example, a process, method, system, product, or apparatus comprising several steps or units is not necessarily limited to the steps or units explicitly listed, but may also include other steps or units not explicitly listed, or steps or units inherent to the process, method, product, or apparatus.

[0021] In this application, a "module" or "unit" refers to a computer program or part of a computer program that has a specific function and works in conjunction with other related parts to achieve a predetermined goal. These modules or units can be implemented by software, hardware (e.g., processing circuitry or memory), or a combination of both. One or more processors or memories can implement one or more modules or units. Furthermore, each module or unit can also be part of a larger module or unit.

[0022] The technical solutions of this application will be described in detail below with reference to the accompanying drawings of the embodiments. It should be noted that the described embodiments are only a part of this application, and not all embodiments. In the following description, the "some embodiments" mentioned are only a subset of all possible embodiments, which may be the same or different subsets, and different embodiments can be combined with each other without conflict.

[0023] Figure 1 This is a schematic flowchart of a vehicle lane change control method provided in an embodiment of this application. For example, see [link to example]. Figure 1 The vehicle lane change control method provided in this application embodiment may include the following steps 101 to 104: Step 101: If the remaining distance between the target vehicle and the latest lane change point is less than a preset distance threshold, then the target vehicle is decelerated based on the remaining distance.

[0024] In some examples, the target vehicle is a self-driving vehicle currently equipped with an intelligent driving assistance system and executing lane change decisions. This vehicle is equipped with an intelligent driving perception and planning control system that can collect its own driving status parameters and lane position parameters in real time. For example, a conventionally mass-produced passenger car equipped with Navigation on Autopilot-City (NOA-City) or Navigation on Autopilot-Highway (NOA-Highway) is the target vehicle when it is driving normally on the road and triggers the lane change decision logic.

[0025] The latest lane change point is the longitudinal critical termination position where a vehicle is allowed to complete a legal and safe lane change operation in its current lane. It serves as the benchmark position for determining whether a vehicle has entered an extreme lane change situation. The latest lane change point can be calculated and output in real time by the system's built-in motion search module. The calculation logic is to superimpose the target vehicle's current longitudinal driving position with the minimum lane change distance parameter. This position is comprehensively matched with road rules constraints such as road markings, ramp merging boundaries, navigation route change nodes, and merging termination areas. Once a vehicle passes this position, it will be unable to legally complete a lane change, which can easily lead to safety risks such as driving over the line, illegal lane changes, and collisions with surrounding vehicles. For example, in the scenario of merging on a highway ramp, the longitudinal position corresponding to the compliant lane change termination marking at the end of the road is the latest lane change point in this situation. In the scenario of navigation lane change in urban roads, the longitudinal termination position corresponding to the critical distance for route change is also the latest lane change point.

[0026] The remaining distance is the straight-line longitudinal distance between the target vehicle's current longitudinal position and the latest lane change point. It only counts the effective and compliant lane change distance within the vehicle's driving lane and is a criterion for triggering pre-deceleration control. It can be calculated in real time by the intelligent driving positioning system collecting the vehicle's current position and combining it with the latest lane change point position output by the motion search module, using the longitudinal coordinate difference. For example, if the target vehicle's current longitudinal position is 1000 meters and the corresponding latest lane change point longitudinal position is 1200 meters, then the current vehicle's remaining distance is 200 meters. As the vehicle continues to move forward, the remaining distance will decrease dynamically in real time.

[0027] The preset distance threshold is a pre-calibrated and fixed extreme lane change warning distance threshold used to determine whether the vehicle has entered the pre-control zone of the latest lane change point. It is the judgment limit for activating deceleration control. It can be combined with measured calibration data of urban and highway congested lane change conditions. In this embodiment, the preset distance threshold can be fixed at 250 meters. This threshold can be adapted to various extreme lane change scenarios such as congested ramps and dense traffic merging and merging. It can ensure that the vehicle has sufficient time for deceleration, gap search and lane change adjustment, effectively avoiding the risk of insufficient lane change distance. For example, when the calculated real-time remaining distance value is less than 250 meters, the pre-deceleration control logic of this step is triggered immediately; when the remaining distance is greater than or equal to 250 meters, the vehicle maintains normal cruising state and does not initiate forced deceleration intervention.

[0028] The process of decelerating the target vehicle based on the remaining distance is a pre-control behavior that uses the real-time dynamically updated remaining distance as the core input benchmark, combined with the real-time driving status of the vehicle, to actively intervene in the vehicle speed control system and reduce the speed of the target vehicle by issuing deceleration constraint commands.

[0029] For example, in the complete identification logic of the latest lane change point, the vehicle's built-in motion search module (motionsearcher) outputs two sets of core parameters in real time: lane change type identifier (lane_change_type) and minimum lane change distance (min_lc_distance). Based on these two sets of parameters, the determination of whether the vehicle has entered the pre-triggering interval of the latest lane change point is completed. This application embodiment sets a globally fixed determination logic, only determining that the vehicle has entered the pre-triggering control range of the extreme lane change when the minimum lane change distance meets the interval constraint of being greater than 0 meters and less than the minimum lane change distance threshold (kMinLcDistance). The minimum lane change distance threshold is uniformly set to 250 meters. The longitudinal coordinate of the latest lane change point (lane_change_end_s) is calculated using the target vehicle's own longitudinal coordinate (ego_sl.s). The minimum lane change distance is superimposed on the real-time output, and the longitudinal boundary of the lane change termination is accurately locked through coordinate calculation. At the same time, the lane change type identifier output by the motion search module distinguishes two types of forced lane change conditions: if the lane change type identifier is the merging end identifier (kMergeEnd), the corresponding condition is the main road merging lane change type (TYPE_MERGE_CHANGE) or the main road exit lane change type (TYPE_MAINROAD_EXIT_CHANGE); if the lane change type identifier is the navigation end identifier (kNaviEnd), the corresponding condition is the regular route switching lane change type (TYPE_DEFAULT_ROUTE_CHANGE) or the road centering lane change type (TYPE_CENTER_CHANGE). By classifying and marking the lane change inducement, the condition classification basis is provided for subsequent differentiated deceleration control and gap search logic.

[0030] For example, when the intelligent driving system of the target vehicle enters the lane change decision state, the motion search module continuously outputs the latest lane change point position and minimum lane change distance parameters corresponding to the current operating conditions in real time; the vehicle positioning system collects its own longitudinal driving coordinates in real time, calculates the current remaining distance through coordinate difference, and continuously compares the real-time remaining distance with the preset 250-meter distance threshold. When the real-time remaining distance is detected to be less than 250 meters for the first time, it is immediately determined that the vehicle has entered the latest lane change point limit lane change control zone, and then the pre-deceleration control logic is activated; throughout the process, the dynamically changing remaining distance is used as the core control basis. The larger the remaining distance, the smoother the deceleration intervention intensity; the smaller the remaining distance, the higher the deceleration control precision. The target vehicle is continuously and gradually decelerated to make the vehicle speed gradually adapt to the remaining limited lane change distance.

[0031] By implementing step 101, the vehicle can actively adjust its speed before reaching the lane change constraint position, reserving sufficient longitudinal space for subsequent safety gap search and lane change path planning. Compared with the method of making a lane change decision after approaching the latest lane change point, it can avoid the problem of insufficient remaining distance causing the vehicle to be unable to complete speed adjustment or insufficient lane change space, and improve the feasibility of the vehicle completing a lane change within the limited distance.

[0032] Step 102: During the deceleration control process, based on the obstacle movement information in the lane to be merged, determine whether there is a safe gap that meets the preset safety conditions.

[0033] In some examples, the lane to be merged into is the target driving lane that the target vehicle needs to enter and change into. It is a compliant target lane determined by the intelligent driving system based on the navigation route, road merging rules, and lane driving guidance information. It is also the final driving lane after the vehicle completes the lane change. The system can identify road lane lines and lane traffic attributes through sensing devices such as onboard cameras and LiDAR, and combine the results of global navigation path planning to lock in a unique compliant lane to be merged into, excluding invalid lanes such as emergency lanes, lanes where lane changes are prohibited, and oncoming lanes. For example, when a vehicle needs to merge into a right-hand ramp while driving on a highway, the right-hand ramp lane is the lane to be merged into. When driving on urban roads, the adjacent lane that needs to be switched to the left to adapt to changes in the navigation route is the lane to be merged into in the corresponding driving condition.

[0034] Obstacle operation information refers to the real-time driving and positional status of all traffic-related objects within the merging lane. It forms the data basis for determining whether the space within the lane meets the conditions for lane changing. This information can be collected and output in real time by the vehicle's environmental perception system, encompassing the obstacle's longitudinal position, lateral position, speed, direction, and size. The perception system continuously scans the entire merging lane, filtering complete operational data for all dynamic and stationary obstacles, including motor vehicles and non-motor vehicles. For example, the position and speed data of slow-moving vehicles, stationary vehicles with malfunctions, and closely following vehicles within the merging lane all fall under obstacle operation information. By collecting this information in real-time at high frequency, the traffic flow distribution and obstacle operation status within the merging lane can be fully reconstructed.

[0035] Preset safety conditions are standardized criteria pre-calibrated to determine whether a local space within a lane supports a target vehicle's safe lane change. They serve as the basis for distinguishing between effective safe spaces and ineffective dangerous spaces, and can be adapted to extreme conditions such as congested lane changes and close-range forced lane changes. The calibration target is to ensure that the vehicle's lane change process has no collision risk and that its driving state is stable. This provides a unified and stable judgment standard for gap detection and can avoid misjudgment problems caused by dynamic traffic flow interference. For example, in the embodiments of this application, the preset safety condition is a collision time safety threshold judgment, which requires that the candidate gap meets the minimum safety standard in both the front and rear directions during the entire predicted driving time of the vehicle. This completely avoids the risk of rear-end collisions, scrapes, and side collisions with obstacles in the lane to be merged into. Only spaces that meet this standard can be considered compliant and usable spaces.

[0036] A safety gap is the effective longitudinal passage space between two adjacent obstacles in the merging lane that allows a target vehicle to fully cut in while meeting preset safety conditions throughout the process. The safety gap is the only effective area for a target vehicle to laterally cut into the merging lane. The spatial dimensions, relative position, and dynamic stability must all meet the lane change safety requirements. It is the spatial carrier for a vehicle to complete a compliant and safe lane change. For example, the longitudinal blank area formed between two vehicles in front and behind in the merging lane can be considered a safety gap if the space is large enough, the relative speed is stable, and there is no risk of collision. This allows the target vehicle to smoothly complete the lane change and cut-in action.

[0037] For example, during the entire process of the target vehicle performing pre-deceleration control based on the remaining distance, the environmental perception system continuously collects the overall road condition information of the lane to be merged into, updates the operational information of all obstacles in the lane in real time, and fully records the position, speed, and distribution status of each obstacle. Based on the real-time collected obstacle operation information, the system performs a full-area traversal screening of the longitudinal space of the entire lane to be merged into, extracting various types of vehicle gap spaces in the lane one by one. At the same time, the system calls the built-in preset safety conditions throughout the process to verify the safety of each extracted lane gap space, judging whether each space meets the lane change safety benchmark requirements. After multiple layers of safety judgment, the system accurately selects all effective safety gaps that meet the standards in the lane to be merged into, completing the gap detection and identification work. This provides accurate data support and spatial basis for subsequent vehicles to directly use the safety gaps to complete lane changes, or to perform gap construction operations when there are no effective gaps. This can improve the identification accuracy and utilization efficiency of lane change space in congested scenarios.

[0038] By implementing step 102, the vehicle can determine the appropriate timing for a lane change based on the real-time traffic conditions of the target lane. Since the safety clearance judgment is performed simultaneously with the vehicle deceleration process, it avoids decision lag caused by relying solely on fixed lane change strategies or passively waiting for lane gaps, thereby improving the accuracy of safety clearance identification and the reliability of lane change decisions.

[0039] Step 103: If no safe gap exists, control the target vehicle to further decelerate according to the target deceleration parameters to create a safe gap that meets the preset safety conditions.

[0040] In some examples, the target deceleration parameter is the optimal deceleration control amount that the target vehicle needs to perform to actively build a compliant safety gap based on the initial deceleration. Specifically, it is the longitudinal deceleration value of the vehicle and is a control parameter that guides the vehicle to perform a second precise deceleration. For example, in the case of dense traffic congestion, the minimum additional deceleration required to meet the safety gap construction requirement is calculated to be -1.5 meters per second squared. After correction by combining the upper limit of the vehicle's comfort deceleration, the final target deceleration parameter is determined to be -1.5 meters per second squared. The vehicle will smoothly perform the second deceleration operation according to this fixed deceleration.

[0041] The process of controlling the target vehicle to further decelerate according to the target deceleration parameters to construct a safe gap that meets the preset safety conditions involves terminating the conventional passive gap search logic when there is no readily available safe gap in the lane to be merged. Instead, the target vehicle is subjected to secondary precise deceleration control based on the calculated target deceleration parameters. By changing the longitudinal speed and position of the target vehicle, the longitudinal relative distance between the target vehicle and the reference obstacle in the lane to be merged is increased, actively creating an effective lane-changing space that meets the preset safety conditions. This represents a control upgrade from passive gap finding to active gap creation. For example, when the distance between vehicles in front and behind in the lane to be merged is small and there is no cutting space that meets the safety standards, the target vehicle continuously and smoothly decelerates according to the calibrated target deceleration parameters. Its own speed gradually decreases, its longitudinal position continues to lag, and the relative distance with the reference obstacle in front continues to increase, ultimately forming a new safe gap that is of sufficient size, stable in state, and meets the safety judgment standards.

[0042] For example, after the target vehicle completes the pre-deceleration control and safety clearance detection, if it is determined that none of the candidate spaces in the merging lane meet the preset safety conditions, it enters the active clearance construction control process. Immediately, based on the real-time motion state of the vehicle and the running state of the reference obstacle in the merging lane, and using the legally mandated safe collision time as a constraint benchmark, it accurately calculates the minimum additional deceleration required to create a safe clearance, and completes parameter correction through a built-in comfort deceleration threshold, locking in the final executable target deceleration parameters. Responding to this parameter command, it performs a smooth and precise secondary deceleration on top of the original deceleration, maintaining a constant speed throughout. During deceleration, there are no instantaneous braking or sudden speed changes; the future longitudinal positions of the target vehicle and the reference obstacle in the merging lane are predicted in real time, and the changes in the relative distance between the two vehicles are continuously monitored; as the target vehicle decelerates smoothly, the longitudinal safety distance between the vehicle and the obstacle in front continues to expand, gradually forming a new safety gap with sufficient longitudinal dimensions and meeting the conditions for full-area safety judgment. This completes the active gap-creating operation under congested and gapless conditions, completely solving the problem of no available lane-changing gap in dense traffic scenarios, and providing reliable space support for subsequent vehicles to cut laterally into the merging lane in one go and complete a stable and safe lane change.

[0043] By implementing step 103, when there is no safe gap that meets the preset safety conditions, the target vehicle is further controlled to decelerate according to the target deceleration parameters, so that the target vehicle can actively adjust the relative motion relationship between itself and the obstacle in the lane to be merged into, thereby constructing a safe gap that meets the safety conditions. Compared with waiting for the target lane to naturally form a usable gap, the vehicle can actively create lane-changing conditions by controlling its own speed, reducing the probability of lane-changing failure due to insufficient safe gap in congested scenarios, and reducing the instability caused by repeated attempts to change lanes.

[0044] Step 104: Based on the longitudinal interval of the safety gap, plan the lane change path for the target vehicle to laterally cut into the lane to be merged from the current lane, so as to control the target vehicle to complete the lane change.

[0045] In some examples, the longitudinal interval of the safety gap is the effective longitudinal space range within the lane to be merged into that has been determined to be a safety gap. It is the legal and safe longitudinal constraint range for the vehicle to perform a lateral lane change and cut-in action, and is used to limit the effective longitudinal distance range for the vehicle to complete the lane change. For example, if the effective starting longitudinal position of the safety gap formed by searching or actively constructing is 17 meters and the effective ending longitudinal position is 45 meters, then the currently available longitudinal interval of the safety gap is a continuous longitudinal road segment from 17 meters to 45 meters. The vehicle needs to complete the complete lateral lane change action within this longitudinal interval.

[0046] A lane change path is a two-dimensional driving trajectory of a target vehicle smoothly transitioning from its current lane position to the center of the lane it is to merge into. It is the only execution trajectory for the vehicle to complete the lane change, and it simultaneously satisfies longitudinal spatial constraints, lateral switching constraints, and vehicle kinematic constraints. The lane change path can be generated by the vehicle trajectory planning system based on the longitudinal interval of the safety gap, the current lane position, the position of the lane to merge into, and the vehicle's kinematic parameters. The trajectory is solved entirely based on the confirmed safety space, avoiding obstacle interference areas. For example, if the target vehicle is initially located at the center of the current lane, a smooth, laterally continuous transition trajectory with controllable acceleration changes and no collision risk is planned within the longitudinal interval corresponding to the safety gap, so that the vehicle can smoothly complete the lane change during longitudinal travel. This smooth transition trajectory is the compliant lane change path.

[0047] For example, after the target vehicle obtains an effective safety gap, the trajectory planning module first reads the complete longitudinal interval corresponding to the safety gap and sets this interval as the effective execution window for the vehicle's lateral lane change action, limiting the vehicle to complete lane switching only within this longitudinal distance range. Combining the target vehicle's real-time driving status, the current lane reference position, and the lane width parameters of the lane to be merged into, the module matches the vehicle's allowed lateral kinematic constraints and solves a continuous, smooth, and abrupt overall switching trajectory within the longitudinal interval of the safety gap. The trajectory planning process relies entirely on the verified safety space, does not generate trajectories in dangerous areas outside the gap, and does not use trial trajectories with multiple small adjustments. After the trajectory is generated, the vehicle control system continuously outputs lateral control commands according to the complete planned trajectory, and, in conjunction with the vehicle's stable longitudinal driving speed, enables the target vehicle to smoothly shift laterally from the initial position of the current lane and cut into the center position of the lane to be merged into in one go.

[0048] By implementing step 104, the lane change path is matched with the actual available safe space. Since the safety gap serves as the spatial constraint basis for the vehicle's lateral entry process, it can ensure that the vehicle can complete continuous lane changes while meeting safety conditions, reduce the number of path adjustments during the lane change process, and improve the stability, safety, and traffic capacity of the vehicle's lane change process in complex traffic scenarios.

[0049] In summary, this embodiment of the application determines the remaining distance between the target vehicle and the latest lane-changing point, and initiates deceleration control in advance when the remaining distance is less than a preset distance threshold. This effectively transforms the lane-changing decision from a passive response near the lane-changing endpoint to advance planning based on distance constraints. Since the vehicle can reserve deceleration and gap search time before reaching the latest lane-changing point, it avoids the problem of insufficient longitudinal adjustment space due to entering the lane-changing area too late, thereby improving the feasibility of completing a lane change within the limited distance. During the deceleration control process, the system further combines the running information of obstacles in the lane to be merged to determine whether there is a safe gap that meets the safety conditions. This means that the lane-changing process no longer relies solely on the naturally formed available space between vehicles, but rather selects a suitable merging time through real-time analysis of the traffic state of the target lane. Because the safe gap judgment is performed simultaneously with the deceleration process, the vehicle can continuously evaluate the lane-changing conditions during speed adjustment, thereby improving the accuracy of safe gap identification. The system improves the stability of lane-changing decisions. When no safe gap exists that meets safety conditions, the system controls the target vehicle to decelerate further according to the target deceleration parameters, allowing the vehicle's motion to actively influence the surrounding traffic space distribution, thereby creating a safe gap that meets safety conditions. Compared to simply waiting for a natural gap to appear in the target lane, this system can create lane-changing conditions by adjusting the vehicle's own speed, reducing the probability of lane-changing failure due to congestion in the target lane, and avoiding unstable behavior caused by the vehicle continuously attempting to change lanes when there is no effective gap. After obtaining a safe gap, the system plans the lane-changing path for the vehicle to laterally cut into the lane to be merged into based on the longitudinal interval corresponding to the safe gap. This combines lane-changing path planning with safe gap constraints. Since the safe gap serves as the spatial basis for the vehicle's lateral cutting, the vehicle can complete the lane change within the range that meets safety conditions, reducing repeated adjustments during the lane-changing process and improving the continuity, safety, and ability to merge into congested scenarios during the lane-changing process. In summary, the vehicle lane change control method provided in this application can achieve safe, continuous and efficient lane changes in congested scenarios by decelerating in advance based on the latest lane change point, dynamically searching or actively constructing safety gaps, and combining safety gap constraints for path planning, thereby improving the success rate of lane changes within a limited distance.

[0050] In some embodiments, the aforementioned deceleration control of the target vehicle based on the remaining distance may include: determining a target deceleration based on the remaining distance and the current speed of the target vehicle; if the remaining distance is greater than a target distance threshold, generating a first acceleration limit command based on the target deceleration and a preset comfort deceleration upper limit, wherein the lower limit of acceleration limited by the first acceleration limit command is the larger of the target deceleration and the preset comfort deceleration upper limit; if the remaining distance is less than or equal to the target distance threshold and the absolute value of the target deceleration is less than a preset hard stop threshold, generating a hard stop line command, wherein the hard stop line command is used to instruct the target vehicle to stop at a preset stopping buffer distance from the latest lane change point; when the remaining distance is less than or equal to the target distance threshold and the absolute value of the target deceleration is greater than or equal to the preset hard stop threshold, generating a second acceleration limit command, wherein the second acceleration limit command is used to instruct the target vehicle to continue decelerating at the target deceleration.

[0051] In some examples, the current vehicle speed is the real-time longitudinal speed of the target vehicle during its journey. The current speed data can be obtained in real time by the fusion acquisition of the target vehicle's onboard inertial measurement unit (IMU) and the vehicle controller. It has the characteristics of high frequency update, high precision, and low latency, and can truly reflect the instantaneous driving status of the vehicle. For example, when the target vehicle is driving at a constant speed under high-speed merging conditions, the real-time longitudinal speed is 12 meters per second, which is the current speed of the target vehicle.

[0052] The target deceleration is the theoretical longitudinal deceleration required for the target vehicle to achieve a smooth stop or compliant deceleration under the constraints of the current remaining distance and current speed. It can be obtained by solving the vehicle's uniform deceleration kinematic model, with the constraint that the vehicle's speed is controllable when it reaches the latest lane change point, and is dynamically calculated by combining the real-time remaining distance and current speed. For example, when the target vehicle's current speed is 12 meters per second and the effective braking distance is 80 meters, the theoretical deceleration calculated by the kinematic formula is -0.9 meters per second squared, which is the target deceleration corresponding to the current operating condition.

[0053] The target distance threshold is a pre-defined distance judgment benchmark used to distinguish between medium-to-long-distance deceleration conditions and close-distance extreme deceleration conditions before the latest lane change point. It is a dividing parameter for achieving graded differentiated speed control. In this embodiment, the target distance threshold can be 20 meters. For example, when the real-time remaining lane change distance is greater than 20 meters, it is judged as a medium-to-long-distance lane change condition, and the comfort deceleration control logic is executed; when the real-time remaining lane change distance is less than or equal to 20 meters, it is judged as a close-distance extreme lane change condition, and the stop line and extreme deceleration judgment logic is entered.

[0054] The preset comfort deceleration limit is the maximum permissible deceleration limit calibrated to ensure a good driving experience. It is used to constrain the braking intensity during normal lane change deceleration and avoid driving bumps and discomfort caused by large deceleration. In this embodiment, the preset comfort deceleration limit can be set to -2.0 meters per second squared.

[0055] The first acceleration limit command is an acceleration constraint control command issued to the vehicle control system when the target vehicle is in a medium-to-long-distance lane change situation. It is used to achieve smooth and comfortable pre-deceleration control. The value selection logic of the first acceleration limit command is to select the larger value between the target deceleration and the preset comfort deceleration upper limit to ensure that the deceleration force is gentle and controllable and avoid excessive braking. For example, if the target deceleration is -0.9 m / s² and the preset comfort deceleration upper limit is -2.0 m / s², the larger value of -0.9 m / s² is selected as the constraint value to generate the corresponding first acceleration limit command and control the vehicle to decelerate smoothly.

[0056] The preset hard stop threshold is a calibrated deceleration judgment threshold that distinguishes between gentle deceleration and severe deceleration at close range. It is used to determine whether there are safe stopping conditions under close-range conditions and serves as the basis for triggering the hard stop line command. In this embodiment, the preset hard stop threshold can be set to -1.0 m / s². For example, when the absolute value of the target deceleration calculated under close-range conditions is less than 1.0 m / s², it means that the vehicle requires a relatively small deceleration force and has safe stopping conditions. When the absolute value of the target deceleration is greater than or equal to 1.0 m / s², it means that the deceleration requirement is severe and it is not suitable to directly force a stop.

[0057] A hard stop line command is a mandatory stopping instruction issued when a target vehicle is in a close-range, extreme lane-changing situation and the deceleration requirement is mild. It is used to lock the vehicle's final driving position and prevent the risk of exceeding the lane-changing limit. The purpose of the hard stop line command is to restrain the target vehicle to come to a smooth stop at a designated position near the latest lane-changing point, waiting for a subsequent lane-changing opportunity and avoiding the collision risk of close-range high-speed lane changes. For example, if the remaining lane-changing distance is 15 meters and the target deceleration is -0.8 meters per second squared, which meets the conditions for mild deceleration at close range, a hard stop line command is issued to control the vehicle to stop smoothly at the designated buffer position.

[0058] The preset parking buffer distance is the designated safe parking position for vehicles. This position has a fixed safety margin from the latest lane change point to prevent vehicles from crossing the line or driving beyond the boundary, ensuring that the parking state is compliant and safe. In this embodiment, the preset parking buffer distance can be 1.0 meter. For example, when the longitudinal position of the latest lane change point is 200 meters, the vehicle will stop smoothly at the preset parking buffer position at 199 meters, with a 1-meter safety margin from the lane change termination boundary.

[0059] The second acceleration limit command is a high-intensity acceleration constraint command issued when the target vehicle is in a close-range, extreme lane change situation and has a strong deceleration requirement. It is used to replace the hard stop line command to avoid vehicle instability caused by instantaneous strong braking. The second acceleration limit command directly uses the calculated target deceleration as the control basis to continuously control the vehicle's deceleration and postpones the deceleration control to the next control cycle to ensure the stability of the vehicle's driving posture. For example, if the remaining lane change distance is 15 meters and the target deceleration is -1.2 meters per second squared, which exceeds the preset hard stop threshold, the second acceleration limit command is generated to control the vehicle to continuously and smoothly decelerate according to the target deceleration.

[0060] For example, after the target vehicle enters the pre-control zone before the latest lane change point, the current speed and remaining lane change distance of the target vehicle are collected in real time, and the corresponding target deceleration is calculated in real time through the vehicle kinematics model. First, the remaining distance is compared with the target distance threshold to distinguish between medium-to-long distance and short-distance conditions. When the remaining distance is greater than 20 meters (a medium-to-long distance condition), the target deceleration is compared with the preset comfort deceleration upper limit, and the larger deceleration value is selected as the control value. A first acceleration limit command is generated to control the vehicle to complete the pre-deceleration in a comfortable and smooth manner, ensuring the safety of lane change preparation while preserving the maximum driving comfort. When the remaining distance is less than or equal to 20 meters... When the target deceleration is less than the preset hard stop threshold, the system further compares the target deceleration with the preset hard stop threshold. If the absolute value of the target deceleration is less than the preset hard stop threshold, it indicates that the current deceleration is moderate and the stopping safety is sufficient. The system immediately issues a hard stop line command to control the target vehicle to stop smoothly at a preset stopping buffer distance of one meter from the latest lane change point, locking the vehicle's position to avoid the risk of exceeding the lane change boundary. If the absolute value of the target deceleration is greater than or equal to the preset hard stop threshold, it indicates that the instantaneous stopping braking intensity is too large, which may easily cause vehicle jerking and attitude instability. In this case, a second acceleration limit command is generated to control the vehicle to continue to decelerate according to the target deceleration, and the precise speed control logic is carried over to the next control cycle for iterative optimization.

[0061] Through the implementation of the above embodiments, the target deceleration is determined based on the remaining distance and the current speed of the target vehicle, enabling the deceleration control to match the actual motion state of the vehicle and the remaining available space, thereby improving the accuracy of the deceleration control. Furthermore, based on the relationship between the remaining distance and the target distance threshold, either an acceleration limit command or a hard stop line command is used for control. This allows the vehicle to gradually reduce its speed according to a comfortable deceleration when the distance is far, to achieve position constraint through a hard stop line when the distance is close and the deceleration requirement is small, and to continue deceleration control when the distance is close and the deceleration requirement is large. This ensures that the lane-changing space requirement is met while also taking into account the vehicle's driving comfort and control safety, avoiding problems such as delayed lane-changing timing or insufficient speed adjustment caused by a single deceleration strategy.

[0062] In some embodiments, the aforementioned step 101 may include: acquiring the current speed of the target vehicle and the obstacle operation information in the lane to be merged into; classifying and identifying the merging endpoint scenario, the stationary obstacle scenario, and the ordinary congestion scenario based on the current speed, the longitudinal position of the latest lane change point, and the obstacle operation information; if any scenario is identified and the remaining distance is less than a preset distance threshold, then decelerating the target vehicle based on the remaining distance.

[0063] In some examples, the merging endpoint scenario refers to the restricted congestion conditions formed at the end of a forced lane change, such as when vehicles merge into a main road from a ramp or exit from a main road. This is a high-risk scenario involving short-distance forced lane changes. In this scenario, the remaining distance for vehicles to legally change lanes is drastically reduced, and traffic flow is dense and interwoven. Vehicles must complete the lane change within a limited longitudinal range, otherwise they will lose the conditions for a legal lane change. For example, if a vehicle is at the end of a highway ramp, and the latest lane change point is close to the end of the road markings, the available lane change buffer distance is greatly reduced. At the same time, the vehicle is driving at a low speed following other vehicles, and the dense traffic flow makes it impossible to quickly pass through the lane change area. The driving conditions of the vehicle at this time are the merging endpoint scenario.

[0064] A stationary obstacle scenario is a congested situation where there is a stationary or nearly stationary obstacle in front of the merging lane or the vehicle's path, hindering the vehicle's normal lane change. This is a type of obstacle intervention-based extreme lane change scenario. In this scenario, the obstacle in front has no obvious speed and continuously occupies the target lane change space, forcing the vehicle to complete the deceleration and lane change operation within a limited distance. It is very easy to encounter the problem of insufficient space near the lane change point. For example, there is a disabled parked vehicle, road construction obstacle, or stationary vehicle waiting to pass in front of the merging lane. The obstacle's speed is close to zero, and the longitudinal distance between the obstacle and the target vehicle is within the safe following range. The vehicle cannot complete the lane change normally at high speed. The corresponding driving condition is a stationary obstacle scenario.

[0065] Typical congestion scenarios are regular lane-changing situations caused by dense traffic flow and slow overall speed on urban roads or highways. This is a typical lane-changing scenario of overall traffic congestion. In this scenario, the overall traffic speed is low, the vehicle in front in the current lane is continuously moving at a low speed, the following distance is close, the traffic efficiency is low, and the gap between the target lane changes frequently. Conventional lane-changing methods are prone to problems such as missing the opportunity to change lanes or failing to change lanes back. For example, during morning and evening rush hours, the entire traffic flow on urban main roads is slow, the vehicle in front in the current lane is continuously moving at a low speed, the following distance is close, the traffic flow in the target lane is dense, and there are few safe gaps that can be directly used. It is difficult for vehicles to successfully merge into the target lane using conventional cruising lane-changing methods. The vehicle is in a typical congestion scenario under these circumstances.

[0066] A dual-condition judgment can be set. The first condition is that any one of the three types of congestion scenarios mentioned above is recognized effectively, confirming that the current situation is a restricted congestion lane-changing condition. The second condition is that the remaining distance between the target vehicle and the latest lane-changing point is less than the system's preset distance threshold, confirming that the vehicle has entered the extreme lane-changing pre-control zone. Only when both conditions are met simultaneously will the pre-deceleration control strategy be activated, dynamically adjusting the target vehicle's speed based on the real-time remaining distance. If there is only a congestion scenario but the remaining distance is sufficient, or if the remaining distance is insufficient but the road conditions are smooth, no forced deceleration intervention will be triggered, ensuring driving efficiency under normal traffic conditions and achieving differentiated control effects such as precise speed control in congestion-restricted conditions and normal cruising in smooth conditions. For example, if the current vehicle is identified as being in a normal congestion scenario, and the preset distance threshold of less than 250 meters in real-time is detected, both conditions are met simultaneously, and then progressive pre-deceleration control is activated based on the current remaining distance, reserving sufficient safety margin for subsequent gap search and lane-changing operations.

[0067] For example, during intelligent driving, the vehicle's perception and positioning systems continuously collect real-time data on the target vehicle's current speed, the movement information of all obstacles in the lane to be merged, and the longitudinal position of the latest lane change point. Based on the collected multi-dimensional data, the system simultaneously and in parallel classifies and identifies three types of congestion scenarios, verifying whether the current driving condition meets the criteria for a merging endpoint scenario, a stationary obstacle scenario, or a normal congestion scenario. When any of the three scenarios is deemed valid, the system further retrieves the real-time remaining lane change distance parameter and compares it with a preset distance threshold. If the remaining distance is less than the preset distance threshold, the system accurately determines that the vehicle is in a congestion-restricted extreme lane change condition and immediately enters the pre-deceleration control process. Relying on the real-time dynamically updated remaining distance parameter, the system performs refined and gradual deceleration control on the target vehicle, gradually reducing the vehicle's speed to adapt to the low-speed lane change requirements of congestion scenarios. If none of the three scenarios are valid, or the remaining distance is greater than the preset distance threshold, the system determines that there is no risk of extreme lane change, and the vehicle maintains normal cruising without additional deceleration intervention.

[0068] By implementing the above embodiments, the current speed of the target vehicle, the longitudinal position of the latest lane change point, and the running information of obstacles in the lane to be merged are obtained. Based on the above information, the merging endpoint scenario, the stationary obstacle scenario, and the ordinary congestion scenario are classified and identified. This enables the vehicle to determine whether it needs to enter the lane change control process in advance according to different traffic environments. Compared with a uniform lane change strategy that does not distinguish specific scenarios, it can improve the adaptability of lane change control to different congestion conditions. The vehicle only performs deceleration and gap planning when there is a lane change requirement and the corresponding scenario conditions are met, thereby improving the effectiveness of the control strategy.

[0069] In some embodiments, the aforementioned classification and identification of merging endpoint scenarios, stationary obstacle scenarios, and ordinary congestion scenarios based on current vehicle speed, the longitudinal position of the latest lane change point, and obstacle operation information may include: when the congestion scenario identification status in the previous judgment period is a non-congestion state, determining the entry threshold set as a lag threshold set; when the congestion scenario identification status in the previous judgment period is a congestion state, determining the exit threshold set as a lag threshold set, wherein the value of each threshold in the exit threshold set is greater than the corresponding value in the entry threshold set; and classifying and identifying merging endpoint scenarios, stationary obstacle scenarios, and ordinary congestion scenarios based on the lag threshold set, current vehicle speed, the longitudinal position of the latest lane change point, and obstacle operation information.

[0070] In some examples, the previous judgment cycle is the control operation cycle that completes a full congestion scenario judgment and is the basic time unit for state iteration. The scene recognition logic can be continuously run by the vehicle's onboard intelligent driving system at a fixed operation frequency. Each complete process of parameter acquisition, condition judgment, and state output constitutes a judgment cycle. The previous judgment cycle refers to the preceding adjacent cycle of the current operation cycle. For example, if the intelligent driving system iterates with a fixed judgment cycle of 100 milliseconds and is currently executing the fifth cycle judgment, then the fourth cycle is the previous judgment cycle.

[0071] The congestion scenario identification status is the qualitative result of the operating condition output after the end of the previous judgment period. It is used to characterize whether the current lane change operation of the target vehicle belongs to the restricted congestion lane change scenario and is the status basis of the dynamic switching threshold system. For example, if the system identifies that the vehicle meets the judgment conditions of ordinary congestion scenario in the previous judgment period, then the corresponding congestion scenario identification status is effective congestion status.

[0072] Non-congested state is one of the state types in congestion scene identification. It means that in the previous judgment period, the target vehicle's driving conditions did not meet the judgment conditions of any one of the following scenarios: merging endpoint scenario, stationary obstacle scenario, or ordinary congestion scenario. The vehicle is in a normal driving condition with smooth traffic and no pressure to change lanes under mandatory restrictions. For example, if the vehicle is driving at a constant speed on an open road, with smooth traffic flow, no nearby obstacles, no mandatory lane change constraints at the endpoint, and no congestion scenario was triggered in the previous period, then the current state is marked as non-congested state.

[0073] Congestion state is another type of congestion scene recognition state. In the previous judgment period, the target vehicle's driving condition met the judgment conditions of any one of the following scenarios: merging endpoint scenario, stationary obstacle scenario, and ordinary congestion scenario. The vehicle is in a restricted congestion lane change condition. For example, in the previous period, the system identified that the vehicle met the judgment conditions of the merging endpoint scenario, and continued to lock the condition as congestion state, keeping the restricted lane change control logic effective.

[0074] The entry threshold set is a preset set of strict judgment parameters used to switch from a non-congested state to a congested state. It includes the speed threshold and distance threshold corresponding to the merging endpoint scenario, the stationary obstacle scenario, and the ordinary congestion scenario. For example, the entry threshold set is configured with a low speed threshold of 3 meters per second for the merging scenario and a low speed threshold of 6.8 meters per second for the ordinary congestion scenario. The vehicle can only enter the congestion recognition state when the vehicle speed is lower than the corresponding value and the distance condition is matched.

[0075] The exit threshold set is a preset set of lenient judgment parameters used to restore a congested state to a non-congested state. It also includes speed and distance thresholds corresponding to the three types of congestion scenarios. All parameter values ​​in the exit threshold set are greater than the corresponding parameter values ​​in the entry threshold set, making the judgment criteria more lenient. This is used to maintain the triggered congestion state and prevent frequent changes in the scenario state caused by small parameter fluctuations in critical conditions. For example, the exit threshold set is configured with a low-speed threshold of 4 meters per second for merging scenarios and a low-speed threshold of 8.0 meters per second for ordinary congestion scenarios. This is more lenient than the entry threshold set, and can maintain the congestion state even when the vehicle operating conditions improve slightly.

[0076] The hysteresis threshold set is a set of real-time judgment parameters dynamically filtered based on historical operating conditions, serving as the final judgment benchmark for scene identification in the current cycle. Based on the congestion scene identification status of the previous judgment cycle, an adaptive matching threshold system is used. In non-congestion states, the entry threshold set is loaded; in congestion states, the exit threshold set is loaded, ultimately forming a hysteresis threshold set with hysteresis anti-shake characteristics. For example, if the previous cycle was non-congestion, the current cycle uses a strict entry threshold set as the hysteresis threshold set; if the previous cycle was continuously congested, the current cycle uses a lenient exit threshold set as the hysteresis threshold set.

[0077] For example, the global congestion identification status of the previous judgment period can be read and cached first. If the vehicle was in a non-congested state in the previous period, the lag threshold set of the current period is configured as the entry threshold set. A new round of scene identification is carried out with strict judgment criteria. Only when the vehicle speed, distance, and obstacle status fully meet the congestion conditions will the vehicle be judged to enter the congestion-restricted lane change condition, avoiding the accidental triggering of deceleration intervention in smooth conditions. If the vehicle was already in a congested state in the previous period, the lag threshold set of the current period is configured as the exit threshold set. The congestion state is maintained with relatively lenient judgment criteria. Even if the vehicle speed and distance fluctuate slightly, the current situation is still identified as a congested lane change scenario, and the pre-deceleration control logic continues to be effective. Based on the dynamically matched lag threshold set, combined with the real-time collected vehicle current speed, longitudinal position of the latest lane change point, and obstacle operation information of the lane to be merged, the classification and verification of merging endpoint scenarios, stationary obstacle scenarios, and ordinary congestion scenarios are completed in parallel.

[0078] Through the implementation of the above embodiments, the entry threshold set or exit threshold set is selected according to the congestion scene identification status of the previous judgment period, and the congestion scene identification is carried out by using the hysteresis threshold mechanism where the exit threshold is greater than the entry threshold. This prevents the vehicle from frequently switching control states due to small changes in speed or distance when the traffic state is close to the critical condition. This reduces state jitter in the congestion scene identification process, improves the continuity and stability of the lane change decision process, and enables the vehicle to adapt to the dynamically changing road traffic environment more reliably.

[0079] In some embodiments, the aforementioned classification and identification of merging endpoint scenarios, stationary obstacle scenarios, and ordinary congestion scenarios based on a set of hysteresis thresholds, current vehicle speed, longitudinal position of the latest lane change point, and obstacle operation information may include: determining that the merging endpoint scenario identification is successful when the longitudinal distance between the latest lane change point and the target vehicle is less than a first preset buffer distance and the current vehicle speed is less than a first preset speed threshold; determining that the stationary obstacle scenario identification is successful when there is a stationary obstacle with a longitudinal speed less than a preset stationary threshold in the lane to be merged, the longitudinal distance between the stationary obstacle and the target vehicle is less than the safe following distance determined based on the current vehicle speed, and the current vehicle speed is less than a second preset speed threshold; determining that the ordinary congestion scenario identification is successful when the global congestion indicator is valid, the speed of the vehicle in front in this lane is less than a preset slow threshold and the following distance is less than a preset following distance, and the current vehicle speed is less than a third preset speed threshold; wherein, the values ​​of the first preset speed threshold, the second preset speed threshold, and the third preset speed threshold in the entry threshold set are respectively less than their corresponding values ​​in the exit threshold set.

[0080] In some examples, the first preset buffer distance is a critical longitudinal distance benchmark parameter used to determine the restricted working condition at the merging endpoint. It measures the urgency of the target vehicle's distance to the lane change termination boundary and is the distance constraint condition for the merging endpoint scenario to be established. For example, the first preset buffer distance corresponding to the entry threshold set is 50 meters, and the first preset buffer distance corresponding to the exit threshold set is 60 meters. By differentiating distance thresholds, the smooth maintenance of the merging scenario state and the precise entry can be achieved.

[0081] The first preset speed threshold is a critical benchmark parameter for determining the vehicle speed at the merging endpoint scenario. It is used to determine whether vehicles in the merging area are in a low-speed congestion-restricted state. Different fixed values ​​can be configured in the entry threshold set and the exit threshold set, with the entry threshold set value being lower than the exit threshold set value. This ensures that the scenario entry standard is stricter and the exit standard is more lenient. For example, if the first preset speed threshold value corresponding to the entry threshold set is 3.0 meters per second and the first preset speed threshold value corresponding to the exit threshold set is 4.0 meters per second, it can effectively identify low-speed congestion conditions at the merging endpoint, while avoiding frequent scene state switching caused by small fluctuations in vehicle speed.

[0082] The process of determining the merging endpoint scenario recognition is valid only when the longitudinal distance between the latest lane change point and the target vehicle is less than the first preset buffer distance and the current vehicle speed is less than the first preset speed threshold. This process ensures that the merging endpoint congestion scenario is valid only when the vehicle is near the lane change termination boundary and the vehicle speed is in a low-speed congestion state. This accurately matches the restricted short-distance forced lane change conditions at the merging end of ramps and main roads. For example, if the current threshold set is loaded, the vehicle is 45 meters away from the latest lane change point, which is less than the first preset buffer distance of 50 meters, and the vehicle's current speed is 2.5 meters per second, which is less than the first preset speed threshold of 3.0 meters per second, then the merging endpoint scenario recognition is immediately determined to be valid.

[0083] The preset stationary threshold is a critical speed parameter used to distinguish between the stationary state and the dynamic driving state of an obstacle. It serves as a speed screening condition for determining stationary obstacle scenarios and is used to screen stationary or nearly stationary obstacles in the lane to be merged into, eliminating interference from normally moving dynamic vehicles. For example, in this embodiment, the preset stationary threshold is set to 0.3 meters per second. When the longitudinal speed of an obstacle is lower than this value, the obstacle can be determined to be a stationary obstacle and included in the obstacle statistics range for scenario determination.

[0084] The safe following distance is a longitudinal safety protection distance dynamically calculated based on the real-time driving speed of the target vehicle. It is used to determine whether a stationary obstacle poses a safety hindrance to the vehicle's lane change. The safe following distance is positively correlated with the vehicle's current speed. The higher the speed, the greater the corresponding safe following distance, adapting to the braking safety margin requirements at different speeds. For example, using an 8-second safety time margin for distance conversion, when the vehicle's current speed is 6 meters per second, the corresponding safe following distance is 48 meters. At the same time, a minimum distance safety threshold of 100 meters is set, and the maximum value of the two is taken as the effective safe following distance.

[0085] The second preset speed threshold is a critical benchmark parameter for determining vehicle speed in stationary obstacle scenarios. It is used to determine whether a vehicle is in a congested and restricted lane-changing state when there is a stationary obstacle. The second preset speed threshold also distinguishes between two sets of values: an entry threshold set and an exit threshold set. The value in the entry threshold set is strictly less than the value in the exit threshold set, matching the hysteresis anti-shake control logic. For example, the second preset speed threshold corresponding to the entry threshold set is 4.2 meters per second, and the second preset speed threshold corresponding to the exit threshold set is 5.0 meters per second. This can accurately identify low-speed congestion caused by stationary obstacles blocking lane-changing space.

[0086] The process of determining the validity of a stationary obstacle scene identification is only valid when three conditions are met simultaneously: the presence of a stationary obstacle in the merging lane with a longitudinal speed less than a preset stationary threshold, the longitudinal distance between the stationary obstacle and the target vehicle being less than the safe following distance determined based on the current vehicle speed, and the current vehicle speed being less than a second preset speed threshold. This allows for accurate identification of extreme conditions where a static obstacle ahead blocks lane-changing space. For example, if there is a stationary vehicle in the merging lane with a speed of 0.2 meters per second, and the longitudinal distance between this vehicle and the target vehicle is 40 meters, which is less than the safe following distance corresponding to the current vehicle speed, and the current speed of the target vehicle is 3.8 meters per second, which is less than the second preset speed threshold corresponding to the entry threshold set, then the stationary obstacle scene can be determined to be valid.

[0087] The preset slow-moving threshold is a critical speed parameter for determining that the vehicle in front in the current lane is in a congested, low-speed driving state. It serves as a condition for determining the state of the vehicle in front in a typical congested scenario, and is used to filter out low-speed following conditions caused by traffic congestion, distinguishing between normal traffic flow and congested, slow-moving traffic flow. For example, in this embodiment of the application, the preset slow-moving threshold can be set to 5.0 meters per second. When the speed of the vehicle in front is lower than this value, it is determined that the vehicle in front is in a congested, slow-moving state.

[0088] The preset following distance is a critical longitudinal distance parameter for determining whether a vehicle is in a dense following congestion state. It is used to measure the density of traffic flow in the lane and is a spatial constraint condition in ordinary congestion scenarios. It is used to identify congestion conditions caused by close following. For example, in this embodiment of the application, the preset following distance can be 20 meters. When the longitudinal following distance between the vehicle and the vehicle in front in the lane is less than this value, it is determined that the vehicle is in a dense following state.

[0089] The third preset speed threshold is a critical benchmark parameter for vehicle speed in ordinary congestion scenarios, used to determine that the target vehicle is in a low-speed congestion driving state. The third preset speed threshold is also configured with a dual threshold system, with the value of the entry threshold set being less than the value of the exit threshold set, to meet the control requirements of hysteresis anti-shake. For example, the third preset speed threshold corresponding to the entry threshold set is 6.8 meters per second, and the third preset speed threshold corresponding to the exit threshold set is 8.0 meters per second, which is suitable for the speed characteristics of regular congested traffic flow on urban roads and highway main roads.

[0090] The global congestion indicator is a condition indicator generated by the intelligent driving system based on road traffic big data and real-time road conditions. A valid indicator means that the current road segment is in a state of full-area congestion. When the global congestion indicator is valid, the speed of the vehicle in front in this lane is less than the preset slow threshold and the following distance is less than the preset following distance, and the current speed is less than the third preset speed threshold, the process of determining that a normal congestion scenario has been established is a comprehensive judgment based on three dimensions: the global road congestion status, the driving status of the vehicle in front, and the driving status of the vehicle itself. This allows for accurate identification of the normal congestion lane-changing conditions of slow-moving traffic and dense following throughout the entire area. For example, if the global congestion indicator is active, the speed of the vehicle in front in this lane is 4.0 meters per second, the following distance is 15 meters, and the current speed of the vehicle itself is 6.0 meters per second, which is less than the third preset speed threshold corresponding to the entry threshold set, then a normal congestion scenario can be determined to have been established.

[0091] For example, when the target vehicle is running in the lane change decision condition at the latest lane change point, it first performs dynamic matching of the lag threshold set based on the congestion status of the previous judgment period to determine the entry threshold set or exit threshold set to be called in the current period; then, it synchronously reads all condition data such as the vehicle's real-time longitudinal position, the longitudinal position at the latest lane change point, the current vehicle speed, the obstacle operation information of the lane to be merged, the status of the vehicle in front in this lane, and the global congestion indicator, and initiates the quantitative judgment process for three types of congestion scenarios in parallel; in the merging endpoint scenario judgment process, it compares the real-time remaining lane change distance with the first preset buffer distance and the real-time vehicle speed with the first preset speed threshold, and the judgment scenario takes effect if both conditions are met simultaneously; in the stationary obstacle scenario judgment process, it first filters stationary obstacles with speeds lower than the preset stationary threshold. Obstacles are identified, and the relative distance to the obstacle is compared with the real-time safe following distance, and the vehicle speed is compared with the second preset speed threshold. Once all three conditions are met, the scenario is confirmed. In the process of determining a normal congestion scenario, the validity of the global congestion indicator, the speed of the vehicle in front is compared with the preset slow-moving threshold, the following distance is compared with the preset following distance, and the vehicle speed is compared with the third preset speed threshold are checked in sequence. Once all conditions are met, the normal congestion scenario is determined to be effective. At the same time, the dual-threshold hysteresis rule is strictly followed. The various speed thresholds take smaller values ​​and are more strictly judged in the entry threshold set, and larger values ​​and are more leniently judged in the exit threshold set. This ensures that the standards are rigorous when the vehicle switches from a smooth driving condition to a congested driving condition, and the state transitions smoothly when the vehicle returns to a smooth driving condition from a congested driving condition. This can effectively prevent the problem of frequent jitter in scenario recognition under critical conditions.

[0092] Through the implementation of the above embodiments, based on information such as the distance to the latest lane change point, vehicle speed, the state of stationary obstacles, the running state of the vehicle in front, and the following distance, targeted identification is performed for merging endpoint scenarios, stationary obstacle scenarios, and ordinary congestion scenarios, enabling vehicles to accurately determine the sources of different types of lane change constraints. Furthermore, by setting speed thresholds and distance thresholds corresponding to different scenarios, and combining the difference between the entry threshold and the exit threshold to achieve delayed judgment, the possibility of scenario misjudgment and frequent switching in complex traffic environments can be reduced, thereby improving the reliability and adaptability of vehicle lane change control.

[0093] In some embodiments, the aforementioned determination of whether a safe gap meeting preset safety conditions exists based on obstacle operation information within the lane to be merged may include: filtering a set of target obstacles located within the lane to be merged from the obstacles within the lane to be merged; sorting the obstacles in the target obstacle set according to their longitudinal positions to determine a reference obstacle, wherein the reference obstacle is the target obstacle in the target obstacle set whose longitudinal position is closest to the target vehicle; generating candidate following gaps and candidate leading gaps based on the longitudinal intervals of the reference obstacle, wherein the candidate following gap is the longitudinal space between the tail of the adjacent obstacle in front of the reference obstacle and the head of the reference obstacle, and the candidate leading gap is the longitudinal space between the tail of the reference obstacle and the head of the adjacent obstacle behind the reference obstacle; and performing safety assessments on the candidate following gaps and candidate leading gaps respectively to determine whether a safe gap meeting preset safety conditions exists.

[0094] In some examples, the target obstacle set is a dataset of all valid obstacles strictly within the driving range of the lane to be merged, selected from all obstacles in the lane to be merged. It is the only valid data source for subsequent gap calculation and safety assessment. All external obstacles that deviate from the lateral range of the lane to be merged and do not obstruct or interfere with the lane change are not included in the target obstacle set, which can effectively avoid irrelevant obstacles from interfering with the gap determination results. For example, multiple vehicles traveling normally in the lane to be merged are included in the target obstacle set. Non-motorized vehicles outside the lane, vehicles traveling in adjacent long-distance lanes, and other lane-outside obstacles are filtered out and do not participate in subsequent gap calculation.

[0095] The reference obstacle is the effective obstacle in the target obstacle set that is closest to the current position of the target vehicle, determined by longitudinal distance. It serves as the benchmark obstacle for generating candidate gaps. The longitudinal coordinates of all obstacles in the target obstacle set can be sorted in ascending order, and the obstacle with the smallest longitudinal distance can be selected as the reference obstacle. For example, if the target obstacle set contains three vehicle obstacles distributed in the lane to be merged, after sorting by longitudinal distance, the slow-moving vehicle in front of the vehicle that is closest to the vehicle is the reference obstacle for the current cycle.

[0096] The candidate following gap is a longitudinal space that can be cut into behind a reference obstacle. It is a safe alternative zone that the target vehicle can cut into behind the reference obstacle. It can be calculated based on the difference in longitudinal boundary coordinates between adjacent obstacles. The spatial range is defined as a continuous longitudinal passage area between the rear position of the adjacent obstacle in front of the reference obstacle and the head position of the reference obstacle. For example, if the longitudinal position of the head of the reference obstacle is 150 meters and the longitudinal position of the rear of the adjacent obstacle in front of it is 130 meters, the 20-meter continuous longitudinal space formed between the two is the candidate following gap under the current working condition.

[0097] The candidate lead gap is a longitudinal space that can be cut into ahead of a reference obstacle. It is a safe zone in which the target vehicle can overtake the reference obstacle and cut into front of it. It can be calculated based on the difference in longitudinal boundary coordinates between the reference obstacle and the adjacent obstacle behind it. The spatial range is defined as a continuous longitudinal passage area between the rear position of the reference obstacle and the head position of the adjacent obstacle behind it. For example, if the longitudinal position of the rear of the reference obstacle is 170 meters and the longitudinal position of the head of the adjacent obstacle behind it is 190 meters, the 20-meter continuous longitudinal space formed between the two is the candidate lead gap under the current operating conditions.

[0098] The process of conducting safety assessments on candidate following gaps and candidate leading gaps to determine whether there are safe gaps that meet preset safety conditions involves independently verifying the safety of two types of candidate gaps at different locations, comprehensively covering both preceding and following lane change merging scenarios. The spatial stability and collision risk adaptability of each candidate gap are verified through unified preset safety conditions. As long as either type of candidate gap meets the preset safety standards, it can be determined that there is a usable safe gap in the merging lane, providing spatial support for subsequent vehicles to directly complete lane changes. If neither type of candidate gap meets the safety requirements, it is determined that there is currently no usable gap, and the subsequent gap construction process can proceed.

[0099] For example, the system can first activate a precise obstacle filtering logic to remove all invalid obstacles that deviate laterally from the lane to be merged into, and then filter out all valid obstacles that are completely within the lane range, thus constructing a real-time updated target obstacle set. Then, all obstacles within the target obstacle set are sorted in ascending order according to their longitudinal distance, precisely locking the obstacle closest to the target vehicle as a reference obstacle. This reference obstacle is then used to match the boundary positions of adjacent obstacles in front and behind. The system reads the tail coordinates of the adjacent obstacle in front of the reference obstacle, the head coordinates of the reference obstacle, the tail coordinates of the reference obstacle, and the head coordinates of the adjacent obstacle behind it. Based on the coordinate differences, corresponding candidate following gaps and candidate leading gaps are generated, fully covering the two types of merging spaces in front of and behind the reference obstacle. After the gaps are generated, a standardized safety assessment is conducted independently for the two types of candidate gaps, verifying whether each gap meets the preset safety conditions. Finally, based on the assessment results, it is determined whether there is a directly usable safe gap in the current merging lane, achieving refined and comprehensive screening of lane-changing space in congested traffic scenarios.

[0100] By implementing the above embodiments, a set of target obstacles is selected from the obstacles in the lane to be merged, and the reference obstacle closest to the target vehicle is determined according to the longitudinal position of the obstacle. This reduces the influence of irrelevant obstacles on the judgment of safe clearance and improves the efficiency of safe clearance search. Furthermore, candidate following clearances and candidate leading clearances are generated based on the longitudinal interval of the reference obstacle, enabling the vehicle to select from multiple potential merging spaces instead of relying on a single lane change space. This expands the scope of safe clearance search and improves the vehicle's lane change opportunities and safe clearance utilization efficiency in congested traffic.

[0101] In some embodiments, the aforementioned safety assessment of candidate following gaps and candidate leading gaps to determine whether there is a safe gap that meets preset safety conditions may include: generating multiple acceleration samples for each candidate gap using a preset acceleration sampling set; performing longitudinal kinematic prediction of the target vehicle at multiple preset prediction times based on each acceleration sample to obtain the collision time between the target vehicle and the corresponding front and rear obstacles at each preset prediction time; determining candidate gaps whose collision time at all preset prediction times is not less than a preset safe collision time threshold as candidate safe gaps; selecting the acceleration sample with the smallest absolute value from the acceleration samples corresponding to the candidate safe gaps as the target acceleration; and determining the candidate safe gap corresponding to the target acceleration as a safe gap that meets preset safety conditions.

[0102] In some examples, the preset acceleration sampling set is a pre-calibrated set of discrete acceleration values ​​covering the normal acceleration and deceleration range of a vehicle. It is used to simulate various feasible longitudinal speed adjustment states during lane changing. It can include multiple sets of uniformly distributed discrete acceleration and deceleration values, which can fully cover the legal speed adjustment range of the vehicle in congested lane changing scenarios and avoid the problem of missing the optimal gap caused by single acceleration evaluation. For example, the preset acceleration sampling set includes multiple positive acceleration values ​​and negative deceleration values, which fully covers the normal driving conditions of small and medium speed adjustment of the vehicle, and adapts to the driving characteristics of low-speed fine adjustment in congested traffic flow.

[0103] Acceleration samples are independent, discrete acceleration values ​​extracted one by one from a preset acceleration sampling set. They are specific control quantities used to simulate the longitudinal motion state of a vehicle during the single-wheel clearance safety assessment. Each acceleration sample corresponds to a vehicle speed regulation strategy. By traversing all acceleration samples, the safety performance of candidate clearances under different speed regulation states can be comprehensively verified, avoiding the one-sidedness of the assessment caused by a single speed regulation state. For example, independent values ​​such as -1.0 m / s², -0.5 m / s², and 0.5 m / s² can be extracted from the preset acceleration sampling set. Each value is an independent acceleration sample.

[0104] The preset prediction time is a set of future time nodes set in advance to cover the complete time domain interval of the vehicle lane change process. It is the time benchmark for realizing full-time domain safety verification, covering the entire process of the vehicle from the start of deceleration and gap adaptation to the completion of lane change. It can realize dynamic risk detection of the entire lane change process and avoid the problem that single-point time verification cannot identify potential collision hazards in the long term. For example, in the embodiments of this application, the preset prediction time covers a continuous time interval of 0.5 seconds to 4.0 seconds, and sets multiple prediction time nodes with a fixed step size to fully cover the action cycle of the vehicle's extreme lane change.

[0105] Longitudinal kinematics prediction is a dynamic calculation process based on vehicle kinematics models, combined with fixed acceleration samples and preset prediction times, to predict the future longitudinal position and speed of a target vehicle. Longitudinal kinematics prediction only focuses on the longitudinal dimension of the vehicle's lane travel and does not involve lateral offset calculations. It is used to accurately predict the future time-domain trajectory and position of the vehicle under different speed conditions. Longitudinal kinematics prediction can be calculated in real time based on the classic vehicle uniformly accelerated motion model, relying on real-time vehicle speed, acceleration samples, and prediction time to complete the solution. For example, based on the target vehicle's current initial speed of 8 meters per second and an acceleration sample of -1 meter per square second, the longitudinal position and speed of the vehicle at the next two seconds can be predicted, completing the longitudinal kinematics extrapolation at a single moment.

[0106] The collision time between the target vehicle and the corresponding obstacles in front and behind at each preset prediction time is a parameter calculated based on the relative motion relationship between the target vehicle and the obstacles at the front and rear boundaries of the candidate gap at each preset prediction time node. The collision time (Time To Collision, TTC) is a safety indicator in the field of vehicle intelligent driving safety assessment, used to quantify the collision risk level between the vehicle and surrounding obstacles. The larger the collision time value, the higher the driving safety margin. It can be calculated in real time based on the predicted longitudinal position and relative speed of the vehicle and the obstacle, by the ratio of relative distance to relative speed, to solve the bidirectional collision time between the target vehicle and the obstacle in front of the gap and the obstacle behind the gap at the current time. For example, if the relative longitudinal distance between the target vehicle and the obstacle in front is 10 meters and the relative speed is 2 meters per second at the prediction time of two seconds in the future, the corresponding collision time is 5 seconds, which meets the basic safe driving requirements.

[0107] The candidate safety gap is a gap that has been preliminarily determined to meet the lane change safety conditions after safety verification in the full time domain and multiple speed adjustment states. The candidate safety gap has a bidirectional collision time that continuously meets the system's preset safety collision time threshold at all preset prediction times, and there is no instantaneous collision risk at any moment. For example, if a candidate following gap has a bidirectional collision time that is greater than the preset safety collision time threshold of 2.5 seconds in all prediction times from 0.5 seconds to 4.0 seconds, then the gap can be determined as a candidate safety gap.

[0108] The target acceleration is the optimal speed regulation parameter with the smallest absolute value selected from all qualified acceleration samples corresponding to the candidate safety gaps. The selection logic of the target acceleration is based on the optimal driving comfort. Among all safe and feasible speed regulation strategies, the acceleration parameter with the smallest speed fluctuation and the most stable driving state is selected to minimize the driving jolt caused by sudden changes in vehicle speed while ensuring lane change safety. It can traverse all qualified acceleration samples, compare the absolute values ​​of each sample, and output the acceleration parameter corresponding to the minimum value. For example, if a candidate safety gap corresponds to two sets of qualified acceleration samples, -0.5 m / s² and -0.8 m / s², the smaller absolute value of -0.5 m / s² is selected as the target acceleration.

[0109] For example, firstly, a preset acceleration sampling set can be retrieved, and multiple sets of discrete acceleration samples can be obtained. For each set of candidate gaps, all acceleration samples are traversed one by one to perform simulated speed adjustment verification. For each set of acceleration samples, based on the vehicle's longitudinal kinematics model, the future longitudinal driving state of the target vehicle is continuously deduced at all preset prediction time nodes, and the motion state of the obstacles in front and behind the lane to be merged is matched synchronously. The bidirectional collision time between the target vehicle and the obstacles in front and behind is calculated in real time at each moment. The collision time data of all prediction time are strictly verified, and only candidate gaps with collision times in the entire time domain that are not lower than the preset safe collision time threshold are included in the set of candidate safe gaps, and unqualified gaps with instantaneous collision risks are eliminated. After completing the initial screening of the safety of all gaps, all qualified acceleration samples corresponding to candidate safe gaps are traversed, and the parameter with the smallest absolute value of acceleration is selected as the target acceleration. Finally, the candidate safe gap corresponding to the target acceleration is determined as the final compliant safe gap.

[0110] Through the implementation of the above embodiments, acceleration sampling is performed for each candidate gap, and a safety assessment is conducted on the collision time between the target vehicle and the obstacles in front and behind based on different prediction times. This enables the safety gap judgment to comprehensively consider the future motion state of the vehicle and the dynamic changes of surrounding traffic participants, thereby improving the accuracy of the safety gap evaluation. Furthermore, the gap corresponding to the acceleration sample with the smallest absolute value is selected from the candidate gaps that meet the safety conditions at all prediction times as the target safety gap. This allows the vehicle to complete the lane change with a smaller speed adjustment cost, improving the smoothness of the lane change process and reducing the problem of decreased driving comfort caused by excessive acceleration and deceleration.

[0111] In some embodiments, the aforementioned control of the target vehicle to further decelerate according to the target deceleration parameter to construct a safety gap that meets preset safety conditions may include: determining a minimum additional deceleration based on the longitudinal velocity of a reference obstacle, constrained by a preset safe collision time threshold; determining the larger of the minimum additional deceleration and a preset upper limit of comfortable deceleration as the target deceleration parameter; predicting the longitudinal positions of the target vehicle and the reference obstacle in the lane to be merged into based on the target deceleration parameter, obtaining a first predicted longitudinal position of the target vehicle and a second predicted longitudinal position of the reference obstacle after a preset construction time; constructing a candidate gap interval based on the first and second predicted longitudinal positions, combined with a preset safety buffer distance; and determining the candidate gap interval as a safety gap that meets preset safety conditions when the longitudinal length of the candidate gap interval is greater than the preset gap length.

[0112] In some examples, the minimum additional deceleration is the minimum deceleration adjustment amount obtained by back-calculating the real-time longitudinal velocity of the reference obstacle, with a preset safe collision time threshold as the hard safety bottom line. It is the basic theoretical deceleration parameter for achieving safe gap creation. It is used to ensure that the relative motion relationship between the vehicle and the reference obstacle always meets the safe collision requirements after the vehicle speed is adjusted, and at the same time, the gap is expanded with the minimum braking amplitude to avoid excessive deceleration and loss of traffic efficiency. For example, if the reference obstacle to be merged into the lane is in a low-speed constant speed state, in order to meet the safe collision time requirement, the minimum additional deceleration that needs to be applied is -1.2 meters per second squared, which is the minimum additional deceleration under the current working condition.

[0113] The process of determining the minimum additional deceleration based on the longitudinal velocity of a reference obstacle, constrained by a preset safe collision time threshold, involves using the preset safe collision time threshold as an insurmountable constraint boundary. This is combined with the real-time longitudinal velocity of the reference obstacle collected by the perception system, matched with the current motion state of the target vehicle, and through iterative calculations of relative motion, the minimum additional deceleration that just meets the safety constraints and minimizes braking amplitude is calculated. This ensures dynamic driving safety throughout the gap-creating process while maximizing vehicle throughput efficiency. For example, if the preset safe collision time threshold is 2.5 seconds, combined with the longitudinal velocity of the reference obstacle at 3 meters per second, the minimum additional deceleration that just meets the safe duration constraint is calculated iteratively, providing a theoretical benchmark for subsequent gap-creating deceleration.

[0114] The process of determining the larger of the minimum additional deceleration and the preset comfort deceleration upper limit as the target deceleration parameter involves comparing the theoretically solved minimum additional deceleration with the preset comfort deceleration upper limit, and selecting the parameter with the larger value and smoother braking as the final target deceleration parameter. This ensures that the actual deceleration force is always within the comfort range, and avoids driving discomfort caused by overly rigid theoretical clearance parameters. For example, if the minimum additional deceleration is -1.2 m / s² and the preset comfort deceleration upper limit is -2.0 m / s², selecting the larger value of -1.2 m / s² as the target deceleration parameter will achieve safe clearance through smooth braking.

[0115] The first predicted longitudinal position is the longitudinal driving coordinate of the target vehicle after the preset construction time, which is derived from the target deceleration parameters. It serves as the vehicle's position reference for constructing candidate gap intervals. The first predicted longitudinal position can be calculated by the vehicle kinematics prediction model. Using the vehicle's current longitudinal position, real-time speed, and locked target deceleration parameters as inputs, the accurate driving position after a fixed time is derived, which can truly reflect the changes in vehicle displacement during the gap-building process. For example, if the preset construction time is set to 4 seconds, the longitudinal coordinate of the target vehicle after 4 seconds is derived from the target deceleration parameters to be 320 meters. This position is the first predicted longitudinal position.

[0116] The second predicted longitudinal position is the longitudinal driving coordinate of the reference obstacle obtained by synchronous simulation after the preset construction time ends. It serves as the obstacle position benchmark for constructing candidate gap intervals. The second predicted longitudinal position can be obtained by combining the real-time longitudinal speed of the reference obstacle, its current position, and the preset construction time for uniform simulation. This accurately matches the movement trend of the obstacle and ensures the accuracy of the gap construction prediction. For example, if the longitudinal coordinate of the reference obstacle is 360 meters after 4 seconds of synchronous simulation, this position is the second predicted longitudinal position.

[0117] The preset safety buffer distance is a fixed longitudinal margin parameter set in advance to ensure safe lane change and cut-in. It is used to avoid gap safety hazards caused by prediction errors and vehicle size deviations. It is used as a boundary compensation amount during the gap construction process to ensure that the final gap has sufficient safety redundancy to meet the requirements of the vehicle's complete body cutting in. For example, in this embodiment, the preset safety buffer distance is 3 meters. During the gap interval calculation process, this parameter is used to compensate and correct the gap boundary to eliminate the risk of scraping caused by critical gaps.

[0118] The candidate gap interval is a new, accessible longitudinal space constructed by combining the predicted positions of vehicles and obstacles with the safety buffer margin. It is the initial gap sample output by the active gap-creating logic. The candidate gap interval no longer depends on the original traffic flow interval, but creates a new space by actively adjusting the vehicle's speed and displacement difference. It is adapted to the extreme working condition of dense traffic flow without natural gaps and provides a new and feasible entry area for lane changing in congested scenarios. For example, by combining the first predicted longitudinal position, the second predicted longitudinal position and the 3-meter safety buffer distance, a continuous longitudinal passage space with a length of 22 meters is constructed. This space is the candidate gap interval for the current working condition.

[0119] The preset gap length is the calibrated minimum longitudinal length threshold for determining whether a gap can support a safe lane change. It serves as the dimensional benchmark for judging the effectiveness of candidate gaps and fully covers the target vehicle's body dimensions and dynamic safety margin for lane changes. Only when the gap length meets the standard can the vehicle smoothly complete a single lateral cutting maneuver. For example, in this embodiment, the preset gap length is 15 meters. Only when the length of the candidate gap interval is greater than 15 meters can the gap size be considered sufficient and the conditions for lane changing be met. After the candidate gap interval is constructed, the effective longitudinal length of the interval can be calculated in real time and compared with the preset gap length threshold to filter out effective gaps with sufficient size and safety redundancy. For example, if the length of the candidate gap interval is 22 meters, which is greater than the preset gap length of 15 meters, then the candidate gap interval is determined to be a compliant and effective safe gap.

[0120] For example, a preset safe collision time threshold can be retrieved as the benchmark for global safety constraints. Combined with the longitudinal velocity and motion state of the reference obstacle collected in real time by the perception system, the minimum additional deceleration that just meets the safety constraints can be calculated through reverse iterative calculation using a relative kinematics model, thus determining the basic deceleration amplitude required to create a gap. In order to take into account driving comfort, the minimum additional deceleration is further compared and screened with the preset comfort deceleration upper limit. The parameter with a larger value and smoother braking is selected as the final executable target deceleration parameter to avoid driving jerks caused by high-intensity braking. After determining the target deceleration parameters, a preset construction time is retrieved, and longitudinal kinematic predictions are performed on the target vehicle and the reference obstacle respectively. The first predicted longitudinal position of the target vehicle and the second predicted longitudinal position of the reference obstacle after the preset construction time are obtained. Based on the displacement difference between the two sets of predicted positions, a preset safety buffer distance is superimposed to complete the gap boundary compensation, and a new candidate gap interval is constructed. Finally, the actual longitudinal length of the candidate gap interval is detected and compared with the preset gap length threshold. If the interval length is greater than the preset gap length, the active gap creation is determined to be successful, and the candidate gap interval is determined as a standard safety gap that meets the preset safety conditions, providing sufficient and compliant longitudinal space support for subsequent vehicles to laterally cut into the merging lane and complete a one-time safe lane change.

[0121] Through the implementation of the above embodiments, when there is no safe gap that meets the safety conditions, the minimum additional deceleration is determined by using a preset safe collision time threshold as a constraint, and the target deceleration parameters are determined by combining the comfort deceleration limit. This allows the vehicle to actively adjust its speed state under the premise of meeting safety constraints, avoiding reliance solely on natural traffic flow to form gaps. Furthermore, by predicting the future longitudinal positions of the target vehicle and reference obstacles, and constructing candidate gap intervals by combining them with safe buffer distances, the vehicle can actively create usable space that meets the safety conditions. When the candidate gap interval meets the preset length requirement, it is used as a safe gap for subsequent lane change planning, thereby improving the success rate of lane changes in congested scenarios and enhancing the vehicle's adaptability to complex traffic environments.

[0122] Furthermore, as an implementation of the aforementioned method embodiments, this application also provides a vehicle lane change control device for implementing the aforementioned method embodiments. This device embodiment corresponds to the aforementioned method embodiments. For ease of reading, this vehicle lane change control device embodiment will not repeat the details of the aforementioned method embodiments one by one, but it should be understood that the device in this application embodiment can correspondingly implement all the contents of the aforementioned method embodiments. For example... Figure 2As shown, the vehicle lane change control device 20 includes: a deceleration control unit 201, a gap judgment unit 202, a gap construction unit 203, and a lane change control unit 204. The deceleration control unit 201 is used to decelerate the target vehicle based on the remaining distance if the remaining distance between the target vehicle and the latest lane change point is less than a preset distance threshold. The gap judgment unit 202 is used to determine, during the deceleration control process, whether a safe gap meeting preset safety conditions exists based on obstacle movement information in the lane to be merged. The gap construction unit 203 is used to control the target vehicle to further decelerate according to target deceleration parameters if no safe gap exists, in order to construct a safe gap meeting preset safety conditions. The lane change control unit 204 is used to plan the lane change path of the target vehicle to laterally cut into the lane to be merged from the current lane based on the longitudinal interval of the safe gap, in order to control the target vehicle to complete the lane change.

[0123] In some embodiments, the deceleration control unit 201 is further configured to determine a target deceleration based on the remaining distance and the current speed of the target vehicle; if the remaining distance is greater than the target distance threshold, a first acceleration limit command is generated based on the target deceleration and a preset comfort deceleration upper limit, wherein the lower limit of acceleration limited by the first acceleration limit command is the larger of the target deceleration and the preset comfort deceleration upper limit; if the remaining distance is less than or equal to the target distance threshold and the absolute value of the target deceleration is less than a preset hard stop threshold, a hard stop line command is generated, wherein the hard stop line command is used to instruct the target vehicle to stop at a preset stopping buffer distance from the latest lane change point; when the remaining distance is less than or equal to the target distance threshold and the absolute value of the target deceleration is greater than or equal to the preset hard stop threshold, a second acceleration limit command is generated, wherein the second acceleration limit command is used to instruct the target vehicle to continue decelerating at the target deceleration.

[0124] In some embodiments, the deceleration control unit 201 is further configured to acquire the current speed of the target vehicle and the obstacle operation information in the lane to be merged into; based on the current speed, the longitudinal position of the latest lane change point and the obstacle operation information, classify and identify the merging endpoint scenario, the stationary obstacle scenario and the ordinary congestion scenario; if any scenario is identified and the remaining distance is less than a preset distance threshold, then deceleration control is performed on the target vehicle based on the remaining distance.

[0125] In some embodiments, the deceleration control unit 201 is further configured to determine the entry threshold set as the lag threshold set when the congestion scene identification status of the previous determination period is a non-congestion state, and to determine the exit threshold set as the lag threshold set when the congestion scene identification status of the previous determination period is a congestion state, wherein the value of each threshold in the exit threshold set is greater than the corresponding value in the entry threshold set; and to classify and identify merging endpoint scenarios, stationary obstacle scenarios, and ordinary congestion scenarios based on the lag threshold set, the current vehicle speed, the longitudinal position of the latest lane change point, and obstacle operation information.

[0126] In some embodiments, the deceleration control unit 201 is further configured to determine that the merging endpoint scenario identification is successful when the longitudinal distance between the latest lane change point and the target vehicle is less than a first preset buffer distance and the current vehicle speed is less than a first preset speed threshold; determine that the stationary obstacle scenario identification is successful when there is a stationary obstacle with a longitudinal speed less than a preset stationary threshold in the lane to be merged, the longitudinal distance between the stationary obstacle and the target vehicle is less than the safe following distance determined based on the current vehicle speed, and the current vehicle speed is less than a second preset speed threshold; and determine that the ordinary congestion scenario identification is successful when the global congestion indicator is valid, the speed of the vehicle in front in this lane is less than a preset slow driving threshold and the following distance is less than a preset following distance, and the current vehicle speed is less than a third preset speed threshold; wherein, the values ​​of the first preset speed threshold, the second preset speed threshold and the third preset speed threshold in the entry threshold set are respectively less than the corresponding values ​​in the exit threshold set.

[0127] In some embodiments, the gap determination unit 202 is further configured to: filter out a set of target obstacles located in the lane to be merged from the obstacles in the lane to be merged; sort the obstacles in the set of target obstacles according to their longitudinal positions to determine a reference obstacle, wherein the reference obstacle is the target obstacle in the set of target obstacles whose longitudinal position is closest to the target vehicle; generate candidate following gaps and candidate leading gaps based on the longitudinal intervals of the reference obstacle, wherein the candidate following gap is the longitudinal space between the tail of the adjacent obstacle in front of the reference obstacle and the head of the reference obstacle, and the candidate leading gap is the longitudinal space between the tail of the reference obstacle and the head of the adjacent obstacle behind the reference obstacle; and perform safety assessments on the candidate following gaps and candidate leading gaps respectively to determine whether there are any safe gaps that meet preset safety conditions.

[0128] In some embodiments, the gap determination unit 202 is further configured to generate multiple acceleration samples for each candidate gap using a preset acceleration sampling set; based on each acceleration sample, perform longitudinal kinematic prediction on the target vehicle at multiple preset prediction times to obtain the collision time between the target vehicle and the corresponding front and rear obstacles at each preset prediction time; determine the candidate gaps whose collision times at all preset prediction times are not less than a preset safe collision time threshold as candidate safe gaps; select the acceleration sample with the smallest absolute value from the acceleration samples corresponding to the candidate safe gaps as the target acceleration; and determine the candidate safe gaps corresponding to the target acceleration as safe gaps that meet preset safety conditions.

[0129] In some embodiments, the gap construction unit 203 is further configured to: determine a minimum additional deceleration based on the longitudinal velocity of a reference obstacle, constrained by a preset safe collision time threshold; determine the larger of the minimum additional deceleration and a preset upper limit of comfortable deceleration as a target deceleration parameter; predict the longitudinal positions of the target vehicle and the reference obstacle in the lane to be merged based on the target deceleration parameter, and obtain a first predicted longitudinal position of the target vehicle and a second predicted longitudinal position of the reference obstacle after a preset construction time; construct a candidate gap interval based on the first predicted longitudinal position and the second predicted longitudinal position, combined with a preset safe buffer distance; and determine the candidate gap interval as a safe gap that meets the preset safety conditions when the longitudinal length of the candidate gap interval is greater than the preset gap length.

[0130] This application also provides a computer-readable storage medium storing computer-executable instructions or computer programs, which, when executed by a processor, will cause the processor to perform any step of the vehicle lane change control method provided in this application.

[0131] In some embodiments, the computer-readable storage medium may be a random access memory (RAM), a read-only memory (ROM), flash memory, a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM); or it may be a variety of devices that include one or any combination of the above-mentioned memories.

[0132] In some embodiments, computer-executable instructions may take the form of programs, software, software modules, scripts, or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as stand-alone programs or as modules, components, subroutines, or other units suitable for use in a computing environment.

[0133] In some embodiments, computer-executable instructions may, but do not necessarily, correspond to files in a file system, and may be stored as part of a file that holds other programs or data, for example, in one or more scripts in a HyperText Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple co-located files (e.g., files that store one or more modules, subroutines, or code sections).

[0134] like Figure 3 As shown, this application also provides a vehicle 30, including a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor. When the processor 320 executes the computer program 311, it implements any step of the above-described vehicle lane change control method.

[0135] This application also provides a computer program product comprising a computer program or computer-executable instructions stored in a computer-readable storage medium. A vehicle's processor reads the computer program or computer-executable instructions from the computer-readable storage medium and executes the computer program or computer-executable instructions, causing the vehicle to perform any step of the vehicle lane change control method described above.

[0136] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A vehicle lane change control method, characterized in that, include: If the remaining distance between the target vehicle and the latest lane change point is less than a preset distance threshold, then the target vehicle is decelerated based on the remaining distance. During the deceleration control process, based on the obstacle movement information in the lane to be merged, it is determined whether there is a safe gap that meets the preset safety conditions; If the safety gap does not exist, the target vehicle is controlled to decelerate further according to the target deceleration parameters in order to construct a safety gap that meets the preset safety conditions. Based on the longitudinal interval of the safety gap, a lane-changing path is planned for the target vehicle to laterally cut into the lane to be merged from the current lane, so as to control the target vehicle to complete the lane change.

2. The vehicle lane change control method according to claim 1, characterized in that, The deceleration control of the target vehicle based on the remaining distance includes: Based on the remaining distance and the current speed of the target vehicle, determine the target deceleration; If the remaining distance is greater than the target distance threshold, a first acceleration limit command is generated based on the target deceleration and the preset comfort deceleration upper limit, wherein the lower limit of acceleration limited by the first acceleration limit command is the larger of the target deceleration and the preset comfort deceleration upper limit; If the remaining distance is less than or equal to the target distance threshold, and the absolute value of the target deceleration is less than the preset hard stop threshold, a hard stop line instruction is generated, wherein the hard stop line instruction is used to instruct the target vehicle to stop at a preset stopping buffer distance from the latest lane change point. When the remaining distance is less than or equal to the target distance threshold and the absolute value of the target deceleration is greater than or equal to the preset hard stop threshold, a second acceleration limit command is generated, wherein the second acceleration limit command is used to instruct the target vehicle to continue decelerating at the target deceleration.

3. The vehicle lane change control method according to claim 1, characterized in that, If the remaining distance between the target vehicle and the latest lane change point is less than a preset distance threshold, then deceleration control is applied to the target vehicle based on the remaining distance, including: Obtain the current speed of the target vehicle and the movement information of obstacles in the lane to be merged into; Based on the current vehicle speed, the longitudinal position of the latest lane change point, and the obstacle operation information, the merging endpoint scenario, the stationary obstacle scenario, and the ordinary congestion scenario are classified and identified. If any scene is identified and the remaining distance is less than the preset distance threshold, then the target vehicle is decelerated based on the remaining distance.

4. The vehicle lane change control method according to claim 3, characterized in that, The classification and identification of merging endpoint scenarios, stationary obstacle scenarios, and ordinary congestion scenarios based on the current vehicle speed, the longitudinal position of the latest lane change point, and the obstacle movement information includes: When the congestion scene identification status of the previous judgment period is non-congestion, the entry threshold set is determined as the lag threshold set. When the congestion scene identification status of the previous judgment period is congestion, the exit threshold set is used as the lag threshold set. The values ​​of each threshold in the exit threshold set are greater than the corresponding values ​​in the entry threshold set. Based on the set of hysteresis thresholds, the current vehicle speed, the longitudinal position of the latest lane change point, and the obstacle operation information, the merging endpoint scenario, the stationary obstacle scenario, and the ordinary congestion scenario are classified and identified.

5. The vehicle lane change control method according to claim 4, characterized in that, The classification and identification of merging endpoint scenarios, stationary obstacle scenarios, and ordinary congestion scenarios based on the hysteresis threshold set, the current vehicle speed, the longitudinal position of the latest lane change point, and the obstacle operation information includes: When the longitudinal distance between the latest lane change point and the target vehicle is less than the first preset buffer distance and the current vehicle speed is less than the first preset speed threshold, the merging endpoint scene recognition is determined to be successful. When there is a stationary obstacle in the lane to be merged into with a longitudinal speed less than a preset stationary threshold, the longitudinal distance between the stationary obstacle and the target vehicle is less than the safe following distance determined based on the current vehicle speed, and the current vehicle speed is less than a second preset speed threshold, the stationary obstacle scene recognition is determined to be successful. When the global congestion indicator is valid, the speed of the vehicle in front in this lane is less than the preset slow speed threshold and the following distance is less than the preset following distance, and the current speed is less than the third preset speed threshold, the ordinary congestion scenario identification is determined to be successful. Wherein, the values ​​of the first preset speed threshold, the second preset speed threshold, and the third preset speed threshold in the entry threshold set are respectively less than the corresponding values ​​in the exit threshold set.

6. The vehicle lane change control method according to claim 1, characterized in that, The determination of whether a safe gap exists that meets preset safety conditions based on the obstacle movement information in the lane to be merged includes: Select a set of target obstacles located within the lane to be merged from the obstacles within the lane to be merged; Based on the longitudinal position of each obstacle in the target obstacle set, a reference obstacle is determined, wherein the reference obstacle is the target obstacle in the target obstacle set that is closest to the target vehicle in terms of longitudinal position. Based on the longitudinal interval of the reference obstacle, candidate following gaps and candidate leading gaps are generated, wherein the candidate following gap is the longitudinal space between the tail of the adjacent obstacle in front of the reference obstacle and the head of the reference obstacle, and the candidate leading gap is the longitudinal space between the tail of the reference obstacle and the head of the adjacent obstacle behind the reference obstacle. The candidate following gap and the candidate leading gap are respectively assessed for safety to determine whether there is a safe gap that meets the preset safety conditions.

7. The vehicle lane change control method according to claim 6, characterized in that, The step of performing safety assessments on the candidate following gap and the candidate leading gap respectively to determine whether there is a safe gap that meets the preset safety conditions includes: For each candidate gap, multiple acceleration samples are generated using a preset acceleration sampling set; Based on each acceleration sample, longitudinal kinematics prediction is performed on the target vehicle at multiple preset prediction times to obtain the collision time between the target vehicle and the corresponding front and rear obstacles at each preset prediction time. Candidate gaps whose collision times at all preset prediction times are not less than the preset safe collision time threshold are identified as potential safe gaps. From the acceleration samples corresponding to the candidate safety gaps, select the acceleration sample with the smallest absolute value as the target acceleration; The candidate safety clearance corresponding to the target acceleration is determined as the safety clearance that meets the preset safety conditions.

8. The vehicle lane change control method according to claim 7, characterized in that, The step of controlling the target vehicle to further decelerate according to the target deceleration parameters to create a safety gap that meets the preset safety conditions includes: The minimum additional deceleration is determined based on the longitudinal velocity of the reference obstacle, constrained by a preset safe collision time threshold. The larger of the minimum additional deceleration and the preset upper limit of comfortable deceleration is determined as the target deceleration parameter; Based on the target deceleration parameters, the longitudinal positions of the target vehicle and the reference obstacle in the lane to be merged are predicted to obtain the first predicted longitudinal position of the target vehicle and the second predicted longitudinal position of the reference obstacle after a preset construction time. Based on the first predicted longitudinal position and the second predicted longitudinal position, and combined with a preset safety buffer distance, a candidate gap interval is constructed; When the longitudinal length of the candidate gap interval is greater than the preset gap length, the candidate gap interval is determined as a safe gap that meets the preset safety conditions.

9. A vehicle lane change control device, characterized in that, include: The deceleration control unit is used to decelerate the target vehicle based on the remaining distance if the remaining distance between the target vehicle and the latest lane change point is less than a preset distance threshold. The gap determination unit is used to determine whether there is a safe gap that meets the preset safety conditions based on the obstacle running information in the lane to be merged during the deceleration control process. A gap construction unit is used to control the target vehicle to further decelerate according to the target deceleration parameters if the safety gap does not exist, so as to construct a safety gap that meets the preset safety conditions. The lane change control unit is used to plan the lane change path of the target vehicle to laterally enter the lane to be merged from the current lane based on the longitudinal interval of the safety gap, so as to control the target vehicle to complete the lane change.

10. A vehicle comprising: The memory and processor are characterized in that the processor is used to implement the steps of the vehicle lane change control method as described in any one of claims 1 to 8 when executing a computer program stored in the memory.

Images