Vehicle lane changing method and device, electronic equipment and storage medium

Abstract

The invention discloses a vehicle lane changing method and device, electronic equipment and a storage medium, the vehicle lane changing method can be applied to the technical field of automatic driving, and the method comprises the steps that in the lane changing process of a vehicle, in response to updating of a current control period, the vehicle is controlled based on a trajectory planning result of the vehicle in the current control period; based on the time-space interaction time sequence of the own vehicle and the other vehicle, trajectory planning is conducted on the own vehicle, a trajectory planning result of the own vehicle in the next control period is obtained, trajectory prediction is conducted on the other vehicle, and a trajectory prediction result of the other vehicle in the next control period is obtained; the other vehicle is a vehicle located on a target adjacent lane within the sensing range of the own vehicle, and the target adjacent lane is a lane changing target of the own vehicle; and updating the time-space interaction time sequence based on the trajectory planning result of the own vehicle in the next control period and the trajectory prediction result of the other vehicle in the next control period. The collision risk in the lane changing process is reduced, and the overall stability and reliability of the automatic driving system are improved.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving technology, and in particular to a vehicle lane changing method, device, electronic device, and storage medium. Background Art

[0002] With the continuous development of intelligent driving technology, trajectory planning and behavior prediction, two core modules in autonomous driving systems, respectively assume the important functions of generating the vehicle's future driving path and predicting the movement trends of surrounding traffic participants. In practical applications, the two should theoretically complement each other, jointly supporting the autonomous driving system's ability to understand and respond to complex traffic environments. An ideal system architecture should enable information fusion and coordinated optimization between trajectory planning and behavior prediction, thereby improving the safety and reliability of overall decision-making.

[0003] However, in existing technologies, trajectory planning and behavior prediction are often designed as relatively independent functional modules, lacking an effective coordination mechanism. This fragmented approach results in limited information exchange between the two. Trajectory planning struggles to fully account for dynamic environmental evolution in the prediction results, while the prediction module is unable to accurately grasp the vehicle's planning intentions. This limits the system's overall ability to cope with complex traffic scenarios, such as multi-vehicle intersections and tight lane changes. This can lead to the output of unreasonable trajectory planning lines and a certain probability of increasing the risk of collision during lane changes. Summary of the Invention

[0004] In order to solve the problems of the prior art, the embodiments of the present application provide a vehicle lane changing method, device, electronic device, and storage medium. The technical solution is as follows: In one aspect, a vehicle lane changing method is provided, the method comprising: During a lane change of the ego vehicle, in response to an update of a current control cycle, controlling the ego vehicle based on a trajectory planning result of the ego vehicle in the current control cycle; Based on the spatiotemporal interaction sequence between the ego vehicle and the other vehicle, a trajectory of the ego vehicle is planned to obtain a trajectory planning result for the ego vehicle in the next control cycle, and a trajectory of the other vehicle is predicted to obtain a trajectory prediction result for the other vehicle in the next control cycle; the other vehicle is a vehicle located in a target adjacent lane within the perception range of the ego vehicle, and the target adjacent lane is the lane change target of the ego vehicle; The spatiotemporal interaction sequence is updated based on the trajectory planning result of the self-vehicle in the next control cycle and the trajectory prediction result of the other vehicle in the next control cycle.

[0005] In another aspect, a vehicle lane changing device is provided, the device comprising: an ego vehicle control module, configured to control the ego vehicle based on a trajectory planning result of the ego vehicle in the current control cycle in response to an update of the current control cycle during the lane change process of the ego vehicle; a planning and prediction module configured to plan a trajectory for the ego vehicle based on a spatiotemporal interaction sequence between the ego vehicle and another vehicle, thereby obtaining a trajectory planning result for the ego vehicle in a next control cycle, and to predict a trajectory for the other vehicle, thereby obtaining a trajectory prediction result for the other vehicle in the next control cycle; the other vehicle being a vehicle within the ego vehicle's perception range and located in a target adjacent lane, the target adjacent lane being the ego vehicle's lane change target; A timing update module is used to update the spatiotemporal interaction timing based on the trajectory planning result of the vehicle in the next control cycle and the trajectory prediction result of the other vehicle in the next control cycle.

[0006] In an exemplary embodiment, the timing update module includes: a distance determination module, configured to determine a proximity distance between a trajectory planning result of the self-vehicle in the next control cycle and a trajectory prediction result of the other vehicle in the next control cycle; an interaction point determination module, configured to determine, when the approach distance is less than or equal to a preset distance threshold, a spatiotemporal interaction point between the ego vehicle and the other vehicle in the next control cycle based on a trajectory planning result of the ego vehicle in the next control cycle and a trajectory prediction result of the other vehicle in the next control cycle; The timing determination module is used to determine the time sequence of the self-vehicle and the other vehicle arriving at the space-time interaction point to update the space-time interaction timing sequence.

[0007] In an exemplary embodiment, the device also includes a non-interaction module, which includes: when the approach distance is greater than the preset distance threshold, determining that the self-vehicle and the other vehicle have no interaction in the next control cycle to update the spatiotemporal interaction timing.

[0008] In an exemplary embodiment, the planning and prediction module includes: a first other vehicle prediction module, configured to, when the spatiotemporal interaction sequence indicates that the other vehicle arrives at the spatiotemporal interaction point before the own vehicle, perform trajectory prediction on the other vehicle to obtain a trajectory prediction result of the other vehicle in the next control cycle; The first ego vehicle prediction module is configured to perform trajectory planning for the ego vehicle based on the trajectory prediction result of the other vehicle in the next control cycle, and obtain a trajectory planning result of the ego vehicle in the next control cycle.

[0009] In an exemplary embodiment, the planning and prediction module further includes: a second ego vehicle prediction module, configured to, when the spatiotemporal interaction sequence indicates that the ego vehicle arrives at the spatiotemporal interaction point before the other vehicle, perform trajectory planning on the ego vehicle and obtain a trajectory planning result for the ego vehicle in the next control cycle; The second other vehicle prediction module is used to perform trajectory prediction on the other vehicle based on the trajectory planning result of the own vehicle in the next control cycle, and obtain the trajectory prediction result of the other vehicle in the next control cycle.

[0010] In an exemplary embodiment, the planning and prediction module further includes: The ego vehicle and other vehicle prediction module is configured to, when the spatiotemporal interaction sequence indicates that the ego vehicle and the other vehicle have no interaction in the next control cycle, perform trajectory planning for the ego vehicle to obtain a trajectory planning result for the ego vehicle in the next control cycle, and perform trajectory prediction for the other vehicle to obtain a trajectory prediction result for the other vehicle in the next control cycle.

[0011] On the other hand, an electronic device is provided, including a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the vehicle lane changing method of any of the above aspects.

[0012] On the other hand, a computer-readable storage medium is provided, in which at least one instruction or at least one program is stored. The at least one instruction or the at least one program is loaded and executed by a processor to implement the vehicle lane changing method as described in any of the above aspects.

[0013] In another aspect, a computer program product or computer program is provided. The computer program product or computer program includes computer instructions stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the electronic device to perform any of the above-described vehicle lane change methods.

[0014] The embodiment of the present application not only controls the driving of the vehicle based on the trajectory planning results of the current cycle within each control cycle, but also combines the spatiotemporal interaction sequence between the vehicle and other vehicles to perform trajectory planning for the next control cycle and predict the trajectories of other vehicles, thereby achieving real-time perception and response to environmental dynamics. On this basis, the updated trajectory planning and prediction results are further used to continuously iterate the spatiotemporal interaction sequence, enabling the system to more accurately reflect the evolution trend of traffic scenarios and improve the ability to predict potential conflicts in complex traffic environments. This enhances the rationality and safety of trajectory planning, reduces the risk of collision during lane changes, and improves the overall stability and reliability of the autonomous driving system. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0016] Figure 1 This is a flowchart of a driver's lane change decision-making process provided by an embodiment of the present application; Figure 2 This is a flow chart of a vehicle lane changing method provided in an embodiment of the present application; Figure 3 This is a schematic diagram of the time-space interaction during the vehicle lane change process provided by an embodiment of the present application; Figure 4 This is a flowchart of another vehicle lane changing method provided by an embodiment of the present application; Figure 5 This is a structural block diagram of a vehicle lane changing device provided in an embodiment of the present application; Figure 6 This is a hardware structure block diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0017] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0018] It should be noted that the terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products, or devices.

[0019] It is understandable that in the specific implementation of this application, related data such as user information is involved. When the above embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0020] Improper lane changes are a major cause of traffic accidents. Research shows that improper lane changes account for 30% of all accidents and cause approximately 10% of all collision-related delays. Predicting a vehicle's lane change intention and taking appropriate action can significantly reduce the risk of accidents. With the advancement of autonomous driving technology and artificial intelligence, modern vehicles are able to collect information and perceive their surroundings while driving, predicting the behavior of nearby vehicles based on real-time data. However, due to differences in driver style and lane change environments, vehicles exhibit varying characteristics when executing lane changes. Most current autonomous driving systems employ a modular design, with prediction and planning modules often developed by separate teams. While trajectory prediction algorithms can provide relatively accurate multimodal predictions, they fail to effectively integrate dynamic interaction information (such as inter-vehicle interactions and the impact of lane changes on the surrounding dynamic environment) and overlook the role of the ego vehicle's trajectory in the lane change process. This limits their performance in complex traffic scenarios and can sometimes lead to potentially risky and irrational trajectory plans.

[0021] In light of this, this application, based on research into human driving principles, conducts an in-depth analysis of the behavioral interaction mechanisms between vehicles, particularly the interaction between the active vehicle's trajectory planning and the predicted trajectory of other vehicles in lane-changing scenarios, to propose a safer lane-changing method. This method aims to improve the artificial intelligence level of autonomous vehicles, the rationality of trajectory planning, driving safety, and the comfort experience of passengers. Specifically, drawing on the lane-changing decision-making process of human drivers, a lane-changing method is proposed. Specifically, this method includes a method for identifying the active and passive relationships between the active vehicle and other vehicles during lane changes. Based on this, a lane-changing method that focuses more on safety margins and is closer to human driving habits is proposed. By enabling data sharing and collaboration between the prediction and planning modules, this method can improve the quality of trajectory planning and enhance the functional effectiveness in lane-changing scenarios. Specifically, it integrates the future path of the active vehicle into the behavioral characteristics of the passive vehicle, thereby generating more reasonable and accurate prediction results. This not only enhances the intelligence level of autonomous vehicles, the accuracy of trajectory planning, and driving safety, but also improves the overall driver and passenger experience and enhances the stability and safety of the autonomous driving system.

[0022] See also Figure 1 , which shows a flowchart of a driver's lane change decision provided by an embodiment of the present application. Figure 1The driver's lane-changing decision-making process is analyzed. Lane-changing decisions involve a multi-factor, comprehensive judgment process based on a dynamic traffic environment. Its core goal is to optimize traffic efficiency and control safety risks through spatial position adjustment. During driving, the vehicles involved in lane changes primarily include the driver and other vehicles in adjacent lanes. While driving, the driver continuously observes the surrounding environment and gathers information. When the speed of the vehicle ahead is consistently lower than their own, they may initiate a lane-changing attempt to achieve higher traffic efficiency. They then observe traffic in adjacent lanes and seek a more efficient lane. Based on the speed and spacing of other vehicles in adjacent lanes, they determine whether their vehicle is in an active lane-changing state, meaning that the vehicle meets a set safe distance from the vehicle to the right in the adjacent lane during the lane-changing process. If so, the lane change is initiated. If the vehicle is in a passive state, meaning that the vehicle does not currently meet a safe distance from the vehicle to the right in the adjacent lane, the driver continues to observe the active vehicle in the adjacent lane and only changes lanes when a safe distance is established.

[0023] Analyzing human drivers' lane-changing behavior reveals that when making lane-changing decisions, drivers are essentially assessing the relative temporal and spatial relationships between their own vehicle and other vehicles, thereby determining whether their own vehicle is in the active position relative to the other vehicle. Therefore, this application, based on an analysis of how human drivers assess the active and passive relationships between their own vehicle and other vehicles, simulates and restores the human logic for lane-changing decisions in complex traffic environments.

[0024] See also Figure 2 , which shows a flow chart of a vehicle lane changing method provided by an embodiment of the present application. It should be noted that this specification provides method operation steps such as the embodiments or flow charts, but may include more or fewer operation steps based on conventional or non-creative work. The order of steps listed in the embodiments is only one way of executing the steps among many, and does not represent the only execution order. When the actual system or product is executed, it can be executed in the order shown in the embodiments or the drawings or in parallel (for example, in a parallel processor or multi-threaded processing environment). Specifically, Figure 2 As shown, the method may include: S201 , during a lane change process of the ego vehicle, in response to an update of a current control cycle, controlling the ego vehicle based on a trajectory planning result of the ego vehicle in the current control cycle.

[0025] Specifically, during lane changes, the system operates using a control cycle to achieve real-time control and dynamic adjustments. Each control cycle has a fixed duration, for example, 100 milliseconds, during which the vehicle status is updated, trajectory planning is performed, and control instructions are output. The current control cycle refers to the cycle in which control operations are being executed, while the next control cycle is the cycle immediately following the current one.

[0026] S203 , based on the spatiotemporal interaction sequence between the ego vehicle and the other vehicle, perform trajectory planning for the ego vehicle to obtain the trajectory planning result of the ego vehicle in the next control cycle, and perform trajectory prediction for the other vehicle to obtain the trajectory prediction result of the other vehicle in the next control cycle.

[0027] The spatiotemporal interaction sequence refers to the spatial location (e.g., coordinates in the road coordinate system) of the intersection of the ego vehicle's and the other vehicle's trajectories during their future travel, as well as the order in which both vehicles will arrive at that point. This sequence describes the dynamic interaction relationship by continuously calculating the coordinates of the intersection of the ego vehicle's planned trajectory and the other vehicle's predicted trajectory, as well as the estimated time it takes for both vehicles to arrive at that point.

[0028] The term "other vehicle" specifically refers to vehicles within the ego vehicle's perception range that are located in the target adjacent lane. The target adjacent lane is the lane the ego vehicle plans to change into. Specifically, there may be one or more other vehicles, and predictions are made for each other vehicle that is located in the target adjacent lane and that may interact with the ego vehicle.

[0029] Trajectory planning involves planning a desired driving path for the ego vehicle during lane changes. This path must take into account factors such as the vehicle's dynamic characteristics, road boundary conditions, traffic regulations, and the safe distance from surrounding vehicles. Furthermore, it must generate a reasonable, safe, efficient, and smooth driving trajectory based on various lane change scenarios and objectives, such as overtaking or lane avoidance. In practice, to improve driving comfort and reduce wear on vehicle components, the trajectory planning results for the next control cycle can be optimized and smoothed. The optimization process can consider factors such as trajectory curvature and acceleration rate to ensure trajectory smoothness. For example, a spline curve can be used to fit the trajectory to ensure continuous curvature changes and avoid abrupt changes. Furthermore, the trajectory planning results must satisfy vehicle dynamic constraints, such as maximum steering angle and maximum acceleration, to ensure the vehicle can actually execute the trajectory. These optimizations and processing can ensure smoother driving, enhance passenger comfort, and extend the vehicle's service life.

[0030] Trajectory prediction involves estimating the possible paths of other vehicles over the next period of time. By analyzing and modeling information such as the vehicle's current speed, acceleration, direction, and intention, and incorporating other relevant factors in the traffic scenario, such as road curvature, traffic signals, and interference from surrounding vehicles, the vehicle's likely trajectory is predicted, providing a basis for subsequent decision-making and control. In practice, various prediction models and methods can be employed to improve the accuracy of other vehicle trajectory prediction. For example, statistical models based on historical data can analyze the other vehicle's past driving behavior to predict its future trajectory. Alternatively, machine learning models, such as recurrent neural networks and long-short-term memory networks, can be used to predict the other vehicle's driving intentions and trajectories. Furthermore, traffic regulations and road structure information can be incorporated to further constrain the predicted trajectory range of other vehicles. For example, at an intersection, the possible directions of other vehicles are limited. The system can narrow the prediction range based on the intersection's shape and traffic signs, thereby improving prediction accuracy.

[0031] In specific implementations, when the ego vehicle is expected to arrive at a trajectory intersection before another vehicle, the system considers the ego vehicle to be the active vehicle. A smooth lane change path is generated for the ego vehicle, maintaining a safe distance. This also suppresses any sudden obstacle avoidance behavior of the other vehicle caused by the ego vehicle's lane change. If the other vehicle is expected to arrive at the intersection first, the ego vehicle is considered the passive vehicle. The ego vehicle's planned trajectory is used as an environmental variable to modify the other vehicle's trajectory prediction (e.g., the other vehicle may slow down or adjust its direction due to the ego vehicle's lane change). Based on this modified prediction of the other vehicle, the ego vehicle's path is adjusted to avoid overtaking conflicts.

[0032] S205 , based on the trajectory planning result of the own vehicle in the next control cycle and the trajectory prediction result of the other vehicle in the next control cycle, update the spatiotemporal interaction sequence.

[0033] Specifically, the closest intersection point between the planned trajectory of the ego vehicle and the predicted trajectory of each other vehicle in the current cycle, as well as the time difference between the two vehicles reaching this point, are calculated. The active and passive relationship between the ego vehicle and the other vehicles is determined based on the time difference.

[0034] Specifically, after the ego vehicle begins changing lanes, the system enters a lane change control loop. Each time the current control cycle is updated, the system obtains the trajectory planning results for the ego vehicle during that cycle. This result includes information such as the vehicle's driving path and speed profile during that cycle. Based on this information, control instructions are generated to drive the ego vehicle's actuators, such as the steering system and drive system, to ensure that the ego vehicle follows the planned trajectory. Simultaneously, the system begins processing the trajectory planning and prediction tasks for the next control cycle. First, the system obtains the current spatiotemporal interaction sequence between the ego vehicle and other vehicles. This sequence is based on the spatiotemporal interaction sequence updated in the previous control cycle. Based on this sequence, the system performs trajectory planning for the ego vehicle and trajectory prediction for other vehicles.

[0035] When the ego vehicle completes the lane change, fully entering the target adjacent lane and maintaining a safe distance from surrounding vehicles, the system determines the lane change is complete. At this point, the system exits lane change control mode and enters normal lane keeping or other driving modes. The lane change completion determination takes into account factors such as the vehicle's position, speed, and relative distance to surrounding vehicles. For example, the lane change is considered successful if the deviation between the ego vehicle's centerline and the centerline of the target adjacent lane is less than a certain threshold for a sustained period (e.g., 2 seconds), while the distance to the vehicle behind is greater than a safe distance.

[0036] As can be seen from the technical solutions described above in the embodiments of this application, the embodiments fully consider the dynamic nature of the active and passive relationships, employing a continuous frame tracking strategy based on control cycles. This strategy achieves continuous dynamic allocation of lane change strategies by cyclically determining the active and passive relationships between the vehicle and other vehicles within each control cycle. This mechanism significantly enhances the system's intelligence and anthropomorphism by simulating the decision-making logic of a human driver who continuously observes, judges, and adjusts during lane changes, bringing the decision-making performance of autonomous vehicles in lane change scenarios closer to that of experienced human drivers.

[0037] In an exemplary embodiment, the above step S205 may include: Determine the proximity between the trajectory planning result of the ego vehicle in the next control cycle and the trajectory prediction result of the other vehicle in the next control cycle; When the approach distance is greater than the preset distance threshold, it is determined that the ego vehicle and the other vehicle have no interaction in the next control cycle, so as to update the spatiotemporal interaction sequence.

[0038] Specifically, if the calculated proximity distance exceeds a preset distance threshold, it indicates that the trajectories of the ego vehicle and the other vehicle will not significantly interact within the foreseeable next control cycle, and their driving paths are relatively independent. In this case, it is determined that there will be no interaction between the ego vehicle and the other vehicle in the next control cycle, and the spatiotemporal interaction sequence is updated accordingly.

[0039] Specifically, the interaction relationship between the ego vehicle and other vehicles in the adjacent lane is divided into three categories: the first category is that the ego vehicle is active and the other vehicle is passive; the second category is that the ego vehicle is passive and the other vehicle is active; the third category is that there is no spatiotemporal interaction between the two vehicles. The specific judgment logic is as follows: Combined Figure 3 The spatiotemporal interaction diagram shown in the figure obtains the planned trajectory S2 of the ego vehicle at future time step t and the predicted trajectory S1 of the other vehicle at the same time. The minimum Euclidean distance d1 between the two trajectories is calculated. A preset distance threshold d is set based on the ego vehicle length parameter (for example, 1.5 times the vehicle length) and can be adaptively adjusted based on the user-defined lane change aggressiveness coefficient (for example, between 0.8 and 1.2 times). If d1 is greater than d, the two vehicles are determined to have no spatiotemporal intersection within the planning period, corresponding to the third type of no-interaction relationship.

[0040] In this case, when planning and predicting the trajectory for the next control cycle, the vehicle can focus less on the other vehicle's behavior and more on its own lane change objectives and driving conditions. For example, the vehicle can change lanes along a more ideal trajectory without frequently adjusting speed or direction, thus improving lane change efficiency.

[0041] The preset distance threshold is dynamically adjusted based on actual driving conditions. For example, when the vehicle is traveling at a high speed, the preset distance threshold can be automatically increased to provide more safety time. In poor weather conditions, such as rain or fog, where visibility is reduced, the threshold can also be appropriately increased to improve safety. This dynamic adjustment of the preset distance threshold is calculated and updated using a preset algorithm based on factors such as the vehicle's real-time speed, environmental sensor data (such as camera and radar perception), and traffic conditions. This dynamic adjustment mechanism enables the system to better adapt to varying driving conditions and improve the rationality of lane change decisions.

[0042] Accordingly, the above step S203 may include: When the spatiotemporal interaction sequence indicates that the ego vehicle and the other vehicle have no interaction in the next control cycle, trajectory planning is performed on the ego vehicle to obtain the trajectory planning result of the ego vehicle in the next control cycle, and trajectory prediction is performed on the other vehicle to obtain the trajectory prediction result of the other vehicle in the next control cycle.

[0043] Specifically, when the spatiotemporal interaction sequence indicates that the ego vehicle and the other vehicle will not interact in the next control cycle, the planning process for the ego vehicle primarily considers its own lane change needs and road conditions, without paying much attention to the behavior of the other vehicle. For the other vehicle, the prediction process produces independent trajectory predictions based on its own driving state and environmental factors. In this case, the trajectory planning and prediction for the ego vehicle and the other vehicle are relatively independent.

[0044] When the approach distance is less than or equal to a preset distance threshold, the time-space interaction point between the ego vehicle and the other vehicle in the next control cycle is determined based on the trajectory planning result of the ego vehicle in the next control cycle and the trajectory prediction result of the other vehicle in the next control cycle; Determine the time sequence of the self-vehicle and other vehicles arriving at the space-time interaction point to update the space-time interaction sequence.

[0045] Continue reading Figure 3 If d1≤d, it is determined that the vehicle and the other vehicle may intersect in the space-time trajectory, and the time order of the two vehicles arriving at the interaction point needs to be further determined. The specific determination method is as follows: Assume that the actual driving trajectory of the vehicle and the other vehicle forms a space-time interaction point at point F, and the time step of the two vehicles reaching the nearest interaction point is used to make the judgment. The time step when the vehicle reaches the nearest space-time interaction point is , the time step when the other car reaches the nearest space-time interaction point , where V1 and V2 are the tangential velocities of the other vehicle and the ego vehicle in the trajectory direction, respectively. When t2 ≤ t1, the ego vehicle arrives at the interaction point before the other vehicle. At this point, the ego vehicle takes the initiative in right-of-way allocation, and this is considered a Type I interaction (active, passive). The planned trajectory of the ego vehicle can be directly output. When t2 > t1, this is considered a Type II interaction (passive, active). In this case, the planned trajectory of the ego vehicle must use the predicted trajectory of the other vehicle as an interaction reference, significantly improving the rationality of trajectory planning.

[0046] Accordingly, the above step S203 may include: When the time-space interaction sequence indicates that the other vehicle arrives at the time-space interaction point before the self-vehicle, the trajectory of the other vehicle is predicted to obtain the trajectory prediction result of the other vehicle in the next control cycle; Based on the trajectory prediction results of other vehicles in the next control cycle, the trajectory of the own vehicle is planned to obtain the trajectory planning result of the own vehicle in the next control cycle.

[0047] Specifically, if the spatiotemporal interaction sequence indicates that the ego vehicle will reach the spatiotemporal interaction point before the other vehicle, trajectory planning is performed first. This planning considers the ego vehicle's lane change objective, speed requirements, and road conditions, resulting in a trajectory that reaches the interaction point on time. Then, based on the ego vehicle's planned trajectory, the trajectory of the other vehicle is predicted. This prediction takes into account the possibility that the other vehicle may react to the ego vehicle's lane change, such as slowing down to yield or adjusting its trajectory, thereby predicting the other vehicle's trajectory for the next control cycle. In this case, the ego vehicle can prioritize passing the interaction point according to the planned trajectory, while the other vehicle adjusts based on the ego vehicle's actions, ensuring a smooth lane change.

[0048] When the time-space interaction sequence indicates that the ego vehicle arrives at the time-space interaction point before the other vehicles, the trajectory of the ego vehicle is planned to obtain the trajectory planning result of the ego vehicle in the next control cycle; Based on the trajectory planning result of the self-vehicle in the next control cycle, the trajectory of the other vehicle is predicted to obtain the trajectory prediction result of the other vehicle in the next control cycle.

[0049] Specifically, when the spatiotemporal interaction sequence indicates that the other vehicle arrives at the spatiotemporal interaction point before the ego vehicle, a trajectory prediction is performed on the other vehicle. This prediction process considers the other vehicle's possible driving intentions, such as staying in the current lane, accelerating, or decelerating, and combines road environmental factors, such as the presence of vehicles ahead and intersections, to arrive at a predicted trajectory for the other vehicle in the next control cycle. Based on this prediction, the ego vehicle's trajectory is planned. This planning process ensures that the ego vehicle does not collide with the other vehicle during driving. Strategies such as slowing down, waiting, or adjusting the lane change path may be necessary. For example, the ego vehicle could continue changing lanes after the other vehicle passes the interaction point, or arrive at the interaction point at a later time to avoid arriving at the same time as the other vehicle.

[0050] In real-world traffic, multiple other vehicles may be in the lane adjacent to the ego vehicle's target. In this case, the system needs to process each potential interaction with the ego vehicle. Specifically, the system obtains the spatiotemporal interaction sequence of each other vehicle, calculates the approach distance between each other vehicle and the ego vehicle, and determines the corresponding spatiotemporal interaction points and their temporal order. For each other vehicle, trajectory planning and prediction are performed according to the aforementioned method, and the corresponding spatiotemporal interaction sequence is updated. When handling multiple other vehicles, the impact of each other vehicle must be comprehensively considered to ensure that the ego vehicle's trajectory planning avoids conflicts with all other vehicles. For example, when multiple other vehicles pass through an interaction point, the ego vehicle needs to adjust its trajectory based on the arrival time sequence of each vehicle to ensure safe passage.

[0051] It can be seen from the above technical solutions of the embodiment of the present application that the embodiment of the present application proposes a method for distinguishing the active and passive relationship between the vehicle and other vehicles based on the analysis of the lane-changing behavior of human drivers, abstracts the vehicle-to-vehicle interaction during the lane-changing process into a game decision-making process, and constructs a multi-dimensional interaction feature vector by simulating the judgment logic of human drivers on road priority to realize dynamic identification of the active and passive relationship; on this basis, differentiated lane-changing strategies are implemented for different active and passive relationships. When it is determined to be a non-interaction state or an active state of the vehicle, the optimized trajectory is directly output without relying on the prediction results of the other vehicle, which reduces computing power consumption and significantly improves planning efficiency. When the vehicle is in a passive state, a multimodal predicted trajectory of the other vehicle is generated based on the environmental perception data, and it is used as a hard constraint to generate an obstacle avoidance trajectory through the interactive optimization algorithm. At the same time, a two-way information collaborative interaction mechanism for ego-vehicle trajectory planning and other-vehicle trajectory prediction was constructed. By realizing the forward transmission and reverse feedback of planning results and prediction data in each control cycle, it effectively solved the technical defects of traditional trajectory prediction algorithms in dealing with the multimodal uncertainty of other-vehicle driving behavior. This strategy formed a closed-loop interactive system by integrating the ego-vehicle's lane-changing intention and the dynamic response data of other vehicles in real time, improving the prediction accuracy of the other-vehicle's motion trajectory during the lane-changing process and significantly enhancing the system's ability to predict the behavior of other vehicles in complex traffic environments.

[0052] See also Figure 4 , which is a flow chart of another vehicle lane changing method provided by an embodiment of the present application. In order to facilitate a full understanding of the solution of the present application, the following Figure 4 Take the example to describe the vehicle lane changing method of the present application as a whole.

[0053] The active relationship between the ego vehicle and all other vehicles in the adjacent lanes in the current frame scenario is judged. Differentiated lane-changing strategies are assigned based on the three existing relationships between the ego vehicle and other vehicles: "ego vehicle active, other vehicles passive", "ego vehicle passive, other vehicles active", and "no interaction between the two vehicles". If the relationship is "no interaction between the two vehicles" or "ego vehicle active, other vehicles passive", since there is no collision risk in lane changing, the conventional planning module is directly called to plan the ego vehicle trajectory and output the planned trajectory for the current frame. If the relationship is "ego vehicle passive, other vehicles active", only predicting the trajectory of other vehicles when changing lanes may cause the planning result to have a collision risk. Therefore, the trajectories of both the ego vehicle and other vehicles need to be predicted at the same time. The prediction module completes the prediction of the trajectory of other vehicles and inputs it into the planning module. At this time, the planning module is upgraded to an interactive planning module. When planning the ego vehicle trajectory, the predicted result of the other vehicle trajectory is incorporated into the vectorized representation of the ego vehicle encoding as an additional vector, and the planning result for the current frame is output through the interactive planning module.

[0054] Considering the continuity of trajectory prediction, the trajectory prediction results of the current frame serve as historical data for the next frame, influencing the determination of active interaction relationships in the next frame. Therefore, the active interaction relationships between vehicles must be determined in advance for the next frame. The current frame's trajectory planning results are input into the trajectory prediction module for the ego vehicle and the passive vehicle, upgrading the prediction module to an interaction prediction module. This module encodes the semantics of the ego vehicle's passive vehicle environment by incorporating the ego vehicle's planned trajectory in the current frame to generate its predicted trajectory. Subsequently, based on the active interaction relationship determination rules, the predicted passive vehicle trajectory is compared with the ego vehicle's planned trajectory in the current frame, calculating the minimum spatiotemporal distance d1 between the two. If d1 < d1 (indicating that the two vehicles will reach the same location in spatiotemporal space), an active interaction relationship is determined according to the aforementioned conditions, and a relationship label is assigned to the other vehicle in the next frame. This breaks the traditional upstream and downstream relationship between the prediction and planning modules, achieving synergy, reducing the uncertainty in predicting the driving behavior of other vehicles, and significantly improving the accuracy of trajectory prediction for surrounding vehicles.

[0055] Corresponding to the vehicle lane changing methods provided in the above-mentioned embodiments, an embodiment of the present application also provides a vehicle lane changing device. Since the vehicle lane changing device provided in the embodiment of the present application corresponds to the vehicle lane changing methods provided in the above-mentioned embodiments, the implementation methods of the aforementioned vehicle lane changing methods are also applicable to the vehicle lane changing device provided in this embodiment and will not be described in detail in this embodiment.

[0056] See also Figure 5, which shows a schematic structural diagram of a vehicle lane changing device provided by an embodiment of the present application, the device has the function of implementing the vehicle lane changing method in the above method embodiment, and the function can be implemented by hardware or by hardware executing corresponding software. Figure 5 As shown, the device may include: The ego vehicle control module 510 is configured to control the ego vehicle based on the trajectory planning result of the ego vehicle in the current control cycle in response to the update of the current control cycle during the ego vehicle's lane change process; The planning and prediction module 520 is configured to plan the trajectory of the ego vehicle based on the spatiotemporal interaction sequence between the ego vehicle and other vehicles, obtaining the planned trajectory of the ego vehicle for the next control cycle, and to predict the trajectory of other vehicles, obtaining the predicted trajectory of other vehicles for the next control cycle. The other vehicles are vehicles within the ego vehicle's perception range and located in the target adjacent lane, and the target adjacent lane is the lane change target of the ego vehicle. The timing update module 530 is used to update the spatiotemporal interaction timing based on the trajectory planning result of the own vehicle in the next control cycle and the trajectory prediction result of the other vehicle in the next control cycle.

[0057] In an exemplary embodiment, the timing update module includes: A distance determination module is used to determine the proximity distance between the trajectory planning result of the ego vehicle in the next control cycle and the trajectory prediction result of the other vehicle in the next control cycle; An interaction point determination module is configured to determine the spatiotemporal interaction point between the ego vehicle and the other vehicle in the next control cycle based on the trajectory planning result of the ego vehicle in the next control cycle and the trajectory prediction result of the other vehicle in the next control cycle when the approach distance is less than or equal to a preset distance threshold; The timing determination module is used to determine the time sequence of the vehicle and other vehicles arriving at the spatiotemporal interaction point to update the spatiotemporal interaction timing.

[0058] In an exemplary embodiment, the device further includes a non-interaction module, which includes: when the approach distance is greater than a preset distance threshold, determining that the vehicle and the other vehicle have no interaction in the next control cycle to update the spatiotemporal interaction sequence.

[0059] In an exemplary embodiment, the planning and prediction module includes: The first other vehicle prediction module is used to predict the trajectory of the other vehicle when the time-space interaction sequence indicates that the other vehicle arrives at the time-space interaction point before the own vehicle, and obtain the trajectory prediction result of the other vehicle in the next control cycle; The first ego vehicle prediction module is used to perform trajectory planning for the ego vehicle based on the trajectory prediction result of the other vehicle in the next control cycle, and obtain the trajectory planning result of the ego vehicle in the next control cycle.

[0060] In an exemplary embodiment, the planning and prediction module further includes: The second ego vehicle prediction module is used to plan the trajectory of the ego vehicle when the spatiotemporal interaction sequence indicates that the ego vehicle arrives at the spatiotemporal interaction point before the other vehicle, and obtain the trajectory planning result of the ego vehicle in the next control cycle; The second other vehicle prediction module is used to predict the trajectory of the other vehicle based on the trajectory planning result of the own vehicle in the next control cycle, and obtain the trajectory prediction result of the other vehicle in the next control cycle.

[0061] In an exemplary embodiment, the planning and prediction module further includes: The ego vehicle and other vehicle prediction module is used to plan the trajectory of the ego vehicle and obtain the trajectory planning result of the ego vehicle in the next control cycle when the spatiotemporal interaction sequence indicates that the ego vehicle and other vehicles have no interaction in the next control cycle, and to predict the trajectory of the other vehicles and obtain the trajectory prediction result of the other vehicles in the next control cycle.

[0062] It should be noted that the apparatus provided in the above embodiments, when implementing its functions, is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the apparatus and method embodiments provided in the above embodiments are based on the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0063] An embodiment of the present application provides an electronic device, which includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement any one of the vehicle lane changing methods provided in the above method embodiments.

[0064] The memory can be used to store software programs and modules. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory. The memory can mainly include a program storage area and a data storage area. The program storage area can store the operating system, application programs required for functions, etc.; the data storage area can store data created based on the use of the device, etc. In addition, the memory can include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory can also include a memory controller to provide the processor with access to the memory.

[0065] The method embodiments provided in the embodiments of the present application can be executed in a computer terminal, a server or a similar computing device, that is, the above-mentioned electronic device can include a computer terminal, a server or a similar computing device. Figure 6 This is a hardware structure block diagram of a computer device for running a vehicle lane changing method provided by an embodiment of the present invention, such as Figure 6 As shown, the internal structure of the computer device may include but is not limited to: a processor, a network interface and a memory. The processor, network interface and memory in the computer device may be connected via a bus or other means. Figure 6 The bus connection is taken as an example.

[0066] Among them, the processor (or CPU (Central Processing Unit)) is the computing core and control core of the computer device. The network interface may optionally include a standard wired interface, a wireless interface (such as WI-FI, a mobile communication interface, etc.). Memory is a memory device in a computer device for storing programs and data. It is understandable that the memory here can be a high-speed RAM storage device or a non-volatile memory device (non-volatile memory), such as at least one disk storage device; optionally, it can also be at least one storage device located away from the aforementioned processor. The memory provides a storage space, which stores the operating system of the electronic device, which may include but is not limited to: Windows system (an operating system), Linux (an operating system), Android (Android, a mobile operating system) system, IOS (a mobile operating system) system, etc., and the present invention is not limited to this; and, in the storage space, one or more instructions suitable for being loaded and executed by the processor are also stored, and these instructions can be one or more computer programs (including program codes). In the embodiment of this specification, the processor loads and executes one or more instructions stored in the memory to implement the vehicle lane changing method provided in the above method embodiment.

[0067] An embodiment of the present application also provides a computer-readable storage medium, which can be set in an electronic device to store at least one instruction or at least one program related to implementing a vehicle lane changing method. The at least one instruction or the at least one program is loaded and executed by the processor to implement any one of the vehicle lane changing methods provided in the above method embodiments.

[0068] Optionally, in this embodiment, the above-mentioned storage medium may include but is not limited to: a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and other media that can store program codes.

[0069] It should be noted that the order of the embodiments of the present application described above is for descriptive purposes only and does not represent the superiority or inferiority of the embodiments. The above description is of specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0070] The various embodiments in this specification are described in a progressive manner. Similar parts between the various embodiments can be referred to in conjunction with each other. Each embodiment focuses on the differences from other embodiments. In particular, the device embodiments are generally similar to the method embodiments, so the description is relatively simple. For relevant parts, refer to the description of the method embodiments.

[0071] Those skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware, or may be accomplished by a program instructing the relevant hardware, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk, or an optical disk, etc.

[0072] The above are only preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. A vehicle lane changing method, characterized in that: The method comprises: During a lane change of the ego vehicle, in response to an update of a current control cycle, controlling the ego vehicle based on a trajectory planning result of the ego vehicle in the current control cycle; Based on the spatiotemporal interaction sequence between the ego vehicle and the other vehicle, a trajectory of the ego vehicle is planned to obtain a trajectory planning result for the ego vehicle in the next control cycle, and a trajectory of the other vehicle is predicted to obtain a trajectory prediction result for the other vehicle in the next control cycle; the other vehicle is a vehicle located in a target adjacent lane within the perception range of the ego vehicle, and the target adjacent lane is the lane change target of the ego vehicle; The spatiotemporal interaction sequence is updated based on the trajectory planning result of the self-vehicle in the next control cycle and the trajectory prediction result of the other vehicle in the next control cycle.

2. The vehicle lane changing method according to claim 1, characterized in that: The updating of the spatiotemporal interaction sequence based on the trajectory planning result of the self-vehicle in the next control cycle and the trajectory prediction result of the other vehicle in the next control cycle includes: Determining a proximity distance between a trajectory planning result of the self-vehicle in the next control cycle and a trajectory prediction result of the other vehicle in the next control cycle; When the approach distance is less than or equal to a preset distance threshold, determining a spatiotemporal interaction point between the ego vehicle and the other vehicle in the next control cycle based on a trajectory planning result of the ego vehicle in the next control cycle and a trajectory prediction result of the other vehicle in the next control cycle; The time sequence in which the self-vehicle and the other vehicle arrive at the space-time interaction point is determined to update the space-time interaction sequence.

3. The vehicle lane changing method according to claim 2, characterized in that: The method further comprises: When the approach distance is greater than the preset distance threshold, it is determined that the self-vehicle and the other vehicle have no interaction in the next control cycle, so as to update the spatiotemporal interaction sequence.

4. The vehicle lane changing method according to claim 2, characterized in that: The process of performing trajectory planning for the ego vehicle based on the spatiotemporal interaction sequence between the ego vehicle and the other vehicle to obtain a trajectory planning result for the ego vehicle in the next control cycle, and performing trajectory prediction for the other vehicle to obtain a trajectory prediction result for the other vehicle in the next control cycle, includes: When the spatiotemporal interaction sequence indicates that the other vehicle arrives at the spatiotemporal interaction point before the own vehicle, performing trajectory prediction on the other vehicle to obtain a trajectory prediction result of the other vehicle in the next control cycle; Based on the trajectory prediction result of the other vehicle in the next control cycle, the trajectory of the own vehicle is planned to obtain the trajectory planning result of the own vehicle in the next control cycle.

5. The vehicle lane changing method according to claim 2, characterized in that: The process of performing trajectory planning for the ego vehicle based on the spatiotemporal interaction sequence between the ego vehicle and the other vehicle to obtain a trajectory planning result for the ego vehicle in the next control cycle, and performing trajectory prediction for the other vehicle to obtain a trajectory prediction result for the other vehicle in the next control cycle, includes: When the spatiotemporal interaction sequence indicates that the ego vehicle arrives at the spatiotemporal interaction point before the other vehicle, performing trajectory planning on the ego vehicle to obtain a trajectory planning result of the ego vehicle in the next control cycle; Based on the trajectory planning result of the own vehicle in the next control cycle, the trajectory of the other vehicle is predicted to obtain the trajectory prediction result of the other vehicle in the next control cycle.

6. The vehicle lane changing method according to claim 3, characterized in that: The process of performing trajectory planning for the ego vehicle based on the spatiotemporal interaction sequence between the ego vehicle and the other vehicle to obtain a trajectory planning result for the ego vehicle in the next control cycle, and performing trajectory prediction for the other vehicle to obtain a trajectory prediction result for the other vehicle in the next control cycle, includes: When the spatiotemporal interaction sequence indicates that the ego vehicle and the other vehicle have no interaction in the next control cycle, trajectory planning is performed on the ego vehicle to obtain a trajectory planning result of the ego vehicle in the next control cycle, and trajectory prediction is performed on the other vehicle to obtain a trajectory prediction result of the other vehicle in the next control cycle.

7. A vehicle lane changing device, characterized in that: The device comprises: an ego vehicle control module, configured to control the ego vehicle based on a trajectory planning result of the ego vehicle in the current control cycle in response to an update of the current control cycle during the lane change process of the ego vehicle; a planning and prediction module configured to plan a trajectory for the ego vehicle based on a spatiotemporal interaction sequence between the ego vehicle and another vehicle, thereby obtaining a trajectory planning result for the ego vehicle in a next control cycle, and to predict a trajectory for the other vehicle, thereby obtaining a trajectory prediction result for the other vehicle in the next control cycle; the other vehicle being a vehicle within the ego vehicle's perception range and located in a target adjacent lane, the target adjacent lane being the ego vehicle's lane change target; A timing update module is used to update the spatiotemporal interaction timing based on the trajectory planning result of the vehicle in the next control cycle and the trajectory prediction result of the other vehicle in the next control cycle.

8. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the vehicle lane changing method according to any one of claims 1 to 6.

9. A computer-readable storage medium, wherein the storage medium stores at least one instruction or at least one program, wherein the at least one instruction or the at least one program is loaded and executed by a processor to implement the vehicle lane changing method according to any one of claims 1 to 6.

10. A computer program, characterized in that When the computer program is executed by a processor, the vehicle lane changing method according to any one of claims 1 to 6 is implemented.

Images