A lateral lane changing method considering multi-vehicle interaction, intelligent vehicle and storage medium

Abstract

The application provides a lateral lane-changing decision-making method considering multi-vehicle interaction, an intelligent vehicle and a readable storage medium, the lateral lane-changing decision-making method comprising: an environment perception module perceiving vehicle driving environment information and vehicle operating state information, and transmitting the information to a lane-changing trigger; the lane-changing trigger receiving the information, performing data analysis, deciding whether the ego vehicle performs lane changing, and if the ego vehicle performs lane changing, transmitting the lane-changing decision to a decision-making module; the decision-making module receiving the lane-changing decision, and performing lane-changing decision processing; a trajectory generation module generating a trajectory according to the lane-changing decision processing, and transmitting the trajectory to a motion control module; and the motion control module receiving the trajectory, performing lane-changing operation according to the trajectory, and realizing lane changing. The application solves the technical problem of low vehicle lane-changing efficiency in a congested road section, and achieves the technical effect of improving the vehicle lane-changing efficiency in the congested road section.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic driving, in particular to a lane changing method considering multi-vehicle interaction, an intelligent vehicle and a storage medium. BACKGROUND

[0002] With the increase in the number of vehicles, the research on intelligent vehicles has gradually been valued, and how to realize efficient and safe lane changing has become a research focus. With the development of computer technology, communication technology, pattern recognition and other technologies, supplemented by correct decision-making and accurate control, the active lane changing of intelligent vehicles can be realized efficiently and safely.

[0003] However, in the actual driving process, there is a problem: the current technical solution excessively pursues safety and ignores driving efficiency, and does not fully utilize the social mechanism of multi-vehicle interaction game and cooperation, so that the overall intelligent degree is not high. The performance bottleneck of the existing intelligent driving assistance system is mainly limited in a few scenes, such as the merging of auxiliary lanes into main lanes in cities and lane changing in congested road sections, which is difficult to complete lane changing, causing the problem of low vehicle lane changing efficiency. SUMMARY

[0004] The problem solved by the present application is the technical problem of low vehicle lane changing efficiency in congested road sections, and the technical effect of improving the vehicle lane changing efficiency in congested road sections is realized.

[0005] To solve the above problems, the present application provides a lane changing decision-making method considering multi-vehicle interaction, comprising:

[0006] Step S100: The environment perception module perceives vehicle driving environment information and vehicle running state information, and transmits the information to the lane changing trigger;

[0007] Step S200: The lane changing trigger receives the information, performs data analysis, and decides whether the ego vehicle performs lane changing. If lane changing is performed, the lane changing decision is transmitted to the decision-making module;

[0008] Step S300: The decision-making module receives the lane changing decision and performs lane changing decision processing. The trajectory generation module generates a trajectory according to the lane changing decision processing and transmits the trajectory to the motion control module;

[0009] Step S400: The motion control module receives the trajectory, performs lane changing operation according to the trajectory, and realizes lane changing;

[0010] The lane changing trigger includes a lane changing safety evaluation module, the decision-making module includes an interactive decision-making unit and a traditional decision-making unit, and the trajectory generation module includes an interactive trajectory generation unit and a traditional trajectory generation unit.

[0011] Compared with the prior art, the technical effects achieved by the technical scheme are as follows: the environment perception module is used for perceiving vehicle driving environment information and vehicle running state information and collecting in real time, analyzing and processing the surrounding environment information of the ego vehicle and the running state of the ego vehicle, and transmitting the collected information to the safety evaluation module of the lane changing trigger; the safety evaluation module performs data analysis on the information transmitted by the environment perception module, the quantifiable safety evaluation module is used for ensuring the safety of the intelligent vehicle when changing lanes laterally, and the shortcomings of other decision-making methods, such as non-transparency and difficulty in quantitative evaluation, are solved, and the calculation process is simple, does not involve a large number of complex solving and repeated iteration processes, and can efficiently obtain an optimal solution, and the overall scheme has strong landing performance; the lane changing trigger is used for receiving the lane changing decision obtained by the safety evaluation module and transmitting the lane changing decision to the trajectory module; the decision module is used for performing lane changing decision processing, selecting a suitable decision unit to plan a trajectory, and transmitting the trajectory to the motion control module after the corresponding trajectory is generated, and finally the motion control module executes the lane changing operation according to the received trajectory to realize lane changing. The interactive decision unit is used for providing a decision unit that makes the vehicle more efficient and more proactive in lateral lane changing when the traditional decision unit fails to successfully execute the lane changing action.

[0012] In an example of the present application, the lane changing trigger receives information, performs data analysis, and decides whether the ego vehicle changes lanes. If the ego vehicle changes lanes, the lane changing decision is transmitted to the decision module, which includes:

[0013] Step S210: The lane changing safety evaluation module receives information, establishes a first mathematical model d 安全距离1 and a second mathematical model d 安全距离2 .

[0014] Step S220: The lateral and longitudinal distance S1 between the current position of the ego vehicle and the target lane rear vehicle is compared with the first mathematical model d 安全距离1 .

[0015] Step S230: The lateral distance S2 between the current position of the ego vehicle and the vehicle in the adjacent lane of the target lane is compared with the second mathematical model d 安全距离2 .

[0016] Step S240: If the lane changing condition is met, the lane changing decision is transmitted to the decision module through the lane changing trigger; if the lane changing condition is not met, steps S210-S240 are continued.

[0017] Wherein, the lane changing condition is S1>d 安全距离1 and S2>d 安全距离2 .

[0018] Compared with the prior art, the technical effects reached by adopting the technical scheme are that the lane changing safety evaluation module fully quantitatively evaluates the safety and potential collision risk of the current lane changing behavior in the form of a mathematical model, and ensures that each lane changing decision made by the lane changing trigger is reasonable and meets the basic safety condition.

[0019] In one example of the present application, the first mathematical model d 安全距离1 = d 纵向响应距离1 + d 纵向制动距离1 + d 纵向冗余 + d 横向冗余 , and the second mathematical model d 安全距离2 = d 横向响应距离 + d 横向制动距离 + d 横向响应距离2 + d 横向制动距离2 + d 横向冗余 .

[0020] Wherein, d 安全距离1 is the safety distance between the current position of the ego vehicle and the rear vehicle of the target lane, d 纵向响应距离1 is the longitudinal response distance of the rear vehicle of the target lane within the response time after the rear vehicle of the target lane perceives the collision danger, d 纵向制动距离1 is the longitudinal braking distance of the rear vehicle of the target lane when the rear vehicle of the target lane performs longitudinal deceleration braking at a relatively small deceleration after the response time, d 纵向冗余 is the longitudinal safety redundancy distance, d 横向冗余 is the lateral safety redundancy distance, d 安全距离2 is the safety distance between the current position of the ego vehicle and the vehicle of the adjacent lane of the target lane, d 横向响应距离 is the lateral response distance of the ego vehicle, d 横向制动距离 is the lateral braking distance of the ego vehicle, d 横向响应距离2 is the lateral response distance of the vehicle of the adjacent lane of the target lane within the response time after the vehicle of the adjacent lane of the target lane perceives the collision danger, and d 横向制动距离2 is the lateral braking distance of the vehicle of the adjacent lane of the target lane when the vehicle of the adjacent lane of the target lane performs lateral deceleration braking at a relatively small deceleration after the response time.

[0021] Compared with the prior art, the technical effects reached by adopting the technical scheme are that the first mathematical model and the second mathematical model improve the safety of the overall decision system by assuming dangerous working conditions to evaluate the safety distance, and the longitudinal safety redundancy distance and the lateral safety redundancy distance can be dynamically set according to the user's preferences, thereby ensuring the flexibility of the system; the quantifiable lane changing safety evaluation model is used to ensure the safety of the intelligent vehicle when the intelligent vehicle changes lanes, and also solves the disadvantages of other decision methods, such as being not transparent and difficult to quantitatively evaluate.

[0022] In an example of the present application, the decision module receives the lane-changing decision and performs the lane-changing decision processing, the trajectory generation module generates a trajectory according to the lane-changing decision processing and transmits the trajectory to the motion control module, comprising:

[0023] Step S310: The traditional decision unit receives the lane-changing decision and performs the lane-changing decision processing, if the lane-changing action can be successfully performed, the traditional trajectory generation unit generates a traditional trajectory and transmits the traditional trajectory to the motion control module; if the lane-changing action cannot be successfully performed, the interactive decision unit receives the lane-changing decision;

[0024] Step S320: The interactive decision unit receives the lane-changing decision and performs the lane-changing decision processing, if the lane-changing action can be successfully performed, the interactive trajectory generation unit generates an interactive trajectory and transmits the interactive trajectory to the motion control module.

[0025] Compared with the prior art, the technical effects achieved by adopting the technical scheme are: the decision module comprises a traditional decision unit and an interactive decision unit. The interactive decision unit is added to provide a decision unit that makes the vehicle more efficient and more proactive in lateral lane changing when the traditional decision unit cannot successfully perform the lane-changing action. Through the proactive interactive behavior, the efficiency of the whole vehicle in merging and lane changing is greatly improved, and the passengers have a more personalized intelligent experience.

[0026] In an example of the present application, the interactive decision unit receives the lane-changing decision and performs the lane-changing decision processing, if the lane-changing action can be successfully performed, the interactive trajectory generation unit generates an interactive trajectory and transmits the interactive trajectory to the motion control module, comprising:

[0027] Step S321: The interactive decision unit receives the lane-changing decision;

[0028] Step S322: A social rule module is set, and if the social rule module is met, a behavior trial stage is entered, and the behavior trial is performed on the other vehicle;

[0029] Step S323: A counter vehicle behavior analysis stage is entered, whether the counter vehicle has a yielding intention after the behavior trial is analyzed, if it is judged that the counter vehicle has a yielding intention, a lane-changing execution stage is entered; if it is judged that the counter vehicle does not have a yielding intention, a lane-changing waiting stage is entered, and step S322 is repeated;

[0030] Step S324: After the vehicle body of the ego vehicle completely enters the target lane, a straightening adjustment is triggered, and a straightening adjustment stage is entered;

[0031] Step S325: After the vehicle body of the ego vehicle is flush with the target lane, the whole lane-changing process is ended.

[0032] Compared with the prior art, the technical effects reached by adopting the technical scheme are as follows: the interactive decision unit is divided into five stages, i.e., behavior exploration, opponent vehicle behavior analysis, lane changing execution, lane changing waiting, and back to normal adjustment. During the interactive decision unit, the essence is a negotiation driving strategy. In the behavior exploration stage, after the ego vehicle as a road right requester initiates a road right request to a road right owner (a behavior exploration trajectory is generated), the ego vehicle enters the opponent vehicle behavior analysis stage. If the road right owner's permission (such as deceleration, avoidance, and other behaviors) is obtained, the ego vehicle can enter the lane changing execution stage, and a fast lateral lane changing trajectory is generated. If the road right owner's attitude is not clear or is refused, the ego vehicle enters the lane changing waiting stage, and degenerates into a defensive and conservative driving strategy, and continues to perform behavior exploration until the road right owner's attitude is updated or a new road right owner comes.

[0033] In an example of the present application, the behavior exploration refers to commanding the ego vehicle to make a lane changing tendency pre-action, and the pre-action includes adjusting a vehicle head direction, turning on a turn signal, and partially invading a target lane.

[0034] Compared with the prior art, the technical effects reached by adopting the technical scheme are as follows: in the behavior exploration stage, the ego vehicle is commanded to make a lane changing tendency pre-action under the premise of safety, feasibility, and compliance with social rules, such as adjusting a vehicle head direction, turning on a turn signal, and partially invading a target lane. The behavior exploration stage is mainly used for judging the avoidance intention of the opponent vehicle.

[0035] In an example of the present application, the social rule module includes driving specifications and road right criteria, and is used for providing a basis for the interactive decision unit to perform lane changing decision processing.

[0036] Compared with the prior art, the technical effects reached by adopting the technical scheme are as follows: the social rule module is internally provided with some driving specifications and road right criteria commonly followed by experienced drivers when changing lanes and converging, and the driving specifications and road right criteria internally provided in the social rule module provide a decision basis for the interactive decision module.

[0037] In an example of the present application, the interactive decision unit receives a lane changing decision and performs lane changing decision processing. If the lane changing action can be successfully performed, the lane changing operation is performed, and the interactive trajectory generation unit generates a series of future driving trajectories of the ego vehicle in the behavior exploration stage, the opponent vehicle behavior analysis stage, the lane changing execution stage, and the lane changing waiting stage.

[0038] Compared with the prior art, the technical effects reached by adopting the technical scheme are as follows: the interactive trajectory generation unit generates a series of future driving trajectories of the ego vehicle according to the instruction of the upstream decision, and finally outputs the trajectories to the downstream motion control module for corresponding control execution; for example, a trajectory with a relatively short length, which enables the front side of the vehicle head to slowly intrude into the target lane, is planned in the behavior exploration stage; and the trajectory generation principle in the lane changing execution stage tends to be efficient rather than comfortable, so as to ensure that the vehicle of the user can quickly and timely complete the lane changing demand.

[0039] In still another aspect, the embodiment of the present application provides an intelligent vehicle, which comprises a processor, a memory, and a program or instruction stored in the memory and executable on the processor, and the program or instruction is executed by the processor to implement the method for lateral lane changing decision considering multi-vehicle interaction according to any embodiment of the present application.

[0040] The intelligent vehicle in the embodiment is used to implement the method for lateral lane changing decision considering multi-vehicle interaction according to any embodiment of the present application, and thus has all the beneficial effects of the method for lateral lane changing decision considering multi-vehicle interaction according to any embodiment of the present application, which will not be repeated here.

[0041] In still another aspect, the embodiment of the present application provides a readable storage medium, which comprises a stored computer program, and when the computer program is executed by a processor, the readable storage medium controls the device where the readable storage medium is located to perform the steps of the method for lateral lane changing decision considering multi-vehicle interaction according to any embodiment of the present application.

[0042] The readable storage medium in the embodiment is used to store the method for lateral lane changing decision considering multi-vehicle interaction according to any embodiment of the present application, and thus has all the beneficial effects of the method for lateral lane changing decision considering multi-vehicle interaction according to any embodiment of the present application, which will not be repeated here.

[0043] After the technical scheme of the present application is adopted, the following technical effects can be achieved:

[0044] (1) The interactive decision unit is added to provide a decision unit that is more efficient and more proactive in lateral lane changing when the traditional decision unit cannot successfully execute the lane changing action, and to bring a more personalized intelligent experience to the driver and passenger;

[0045] (2) The lane changing safety evaluation module quantitatively evaluates the safety and potential collision risk of the current lane changing behavior by using a mathematical model, so as to ensure the safety of the intelligent vehicle in lateral lane changing;

[0046] (3) The longitudinal safety redundancy distance and the lateral safety redundancy distance in the lane changing safety evaluation module can be dynamically set according to the user's preference, so as to ensure the flexibility of the system;

[0047] (4) The lane change safety assessment module has a simple calculation process, does not involve a large number of complex solutions and repeated iterations, and can obtain the optimal solution relatively efficiently. The overall solution is highly feasible. Attached Figure Description

[0048] Figure 1 This is one of the flowcharts of a lateral lane change decision-making method considering multi-vehicle interaction provided in Embodiment 1 of the present invention.

[0049] Figure 2 This is the second flowchart of a lateral lane change decision-making method considering multi-vehicle interaction provided in Embodiment 1 of the present invention.

[0050] Figure 3 for Figure 2 A diagram illustrating an intelligent vehicle changing lanes.

[0051] Figure 4 This is the third step in the flowchart of a lateral lane change decision-making method considering multi-vehicle interaction provided in Embodiment 1 of the present invention.

[0052] Figure 5 This is a block diagram of an intelligent vehicle provided in the second embodiment of the present invention.

[0053] Figure 6 This is a block diagram of a readable storage medium provided in the third embodiment of the present invention.

[0054] Explanation of reference numerals in the attached figures:

[0055] 100 - Intelligent vehicle; 110 - Memory; 111 - Computer program; 120 - Processor; 200 - Readable storage medium; 210 - Computer-executable instructions. Detailed Implementation

[0056] To make the above-mentioned objectives, features, and advantages of the present invention more apparent and understandable, the technical solutions in the embodiments of the present invention are clearly and completely described. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0057] [First Embodiment]

[0058] See Figures 1-4 This invention provides a lateral lane change decision-making method considering multi-vehicle interaction, comprising:

[0059] Step S100: The environmental perception module senses the vehicle's driving environment information and vehicle operating status information, and transmits the information to the lane change trigger;

[0060] Step S200: The lane changing trigger receives information, performs data analysis, decides whether the ego vehicle changes lane, and if so, transmits the lane changing decision to the decision module;

[0061] Step S300: The decision module receives the lane changing decision and performs lane changing decision processing, the trajectory generation module generates a trajectory according to the lane changing decision processing and transmits the trajectory to the motion control module;

[0062] Step S400: The motion control module receives the trajectory, performs lane changing operation according to the trajectory, and realizes lane changing;

[0063] The lane changing trigger includes a lane changing safety assessment module, the decision module includes an interactive decision unit and a traditional decision unit, and the trajectory generation module includes an interactive trajectory generation unit and a traditional trajectory generation unit.

[0064] The scheme of the embodiment is applied to the scene of intelligent vehicle lane changing in a crowded road section. In the embodiment, the environment perception module needs to perceive vehicle driving environment information and vehicle operating state information for real-time collection, analyze and process the ego vehicle surrounding environment information and the ego vehicle operating state, and transmit the collected information to the safety assessment module of the lane changing trigger.

[0065] The safety assessment module performs data analysis on the information transmitted by the environment perception module, decides whether the ego vehicle changes lane through analysis, and if so, transmits the lane changing decision to the decision module; The safety assessment module uses a mathematical model to fully quantify the safety and potential collision risk of the current lane changing behavior, ensuring that every lane changing decision made by the lane changing trigger is reasonable and meets the basic safety conditions. The quantifiable safety assessment module is used to ensure the safety of the intelligent vehicle when changing lanes laterally, and also solves the shortcomings of other decision-making methods that are not transparent and difficult to quantify.

[0066] The decision module includes an interactive decision unit and a traditional decision unit. In the decision of lane changing, the decision module performs lane changing decision processing, selects whether to plan a trajectory through the interactive decision unit or the traditional decision unit, and transmits the trajectory to the motion control module after generating the corresponding trajectory. Finally, the motion control module performs lane changing operation according to the received trajectory to realize lane changing.

[0067] Further, the lane changing trigger receives information, performs data analysis, decides whether the ego vehicle changes lane, and if so, transmits the lane changing decision to the decision module, including:

[0068] Step S210: The lane changing safety assessment module receives information, establishes a first mathematical model d 安全距离1 And a second mathematical model d 安全距离2 ;

[0069] Step S220: Compare the lateral and longitudinal distances S1 between the current position of the vehicle and the vehicle behind it in the target lane with the first mathematical model d. 安全距离1 Perform a comparison;

[0070] Step S230: Combine the lateral distance S2 between the current position of the vehicle and the vehicles in the adjacent lane of the target lane with the second mathematical model d. 安全距离2 Perform a comparison;

[0071] Step S240: If the lane change conditions are met, the lane change decision is transmitted to the decision module through the lane change trigger; if the lane change conditions are not met, continue with steps S210-S240.

[0072] The lane change condition is S1>d. 安全距离1 and S2>d 安全距离2 .

[0073] In this embodiment, the lane change safety assessment module uses a mathematical model to fully quantify and evaluate the safety and potential collision risks of the current lane change behavior, ensuring that every lane change decision made by the lane change trigger is reasonable and meets basic safety conditions. During the lane change process, the main considerations are the collision risks arising from the intersection of the lateral and longitudinal distances between the vehicle and the vehicle following in the target lane (other vehicle 1); and the lateral collision risks caused by potential lane change behaviors from vehicles in adjacent lanes (other vehicle 2) of the target lane. To ensure the absolute safety of the vehicle by the decision command, the lane change safety assessment module adopts a worst-case assumption in its risk assessment; that is, a decision command that meets the model assumptions must have an extremely high safety factor.

[0074] The lateral and longitudinal distances S1 between the current position of the vehicle and the vehicle behind it in the target lane are compared with the first mathematical model d. 安全距离1 The comparison is performed by comparing the lateral distance S2 between the current position of the vehicle and the vehicles in the adjacent lane of the target lane with the second mathematical model d. 安全距离2 The comparison is performed, and if the worst-case assumption is met: that is, S1>d 安全距离1 and S2>d 安全距离2 If the lane change condition is met, it means the next step can be safely performed.

[0075] Further, see Figure 3 The first mathematical model d 安全距离1 =d 纵向响应距离1 +d 纵向制动距离1 +d 纵向冗余 +d 横向冗余 The second mathematical model d 安全距离2 =d 横向响应距离 +d 横向制动距离 +d 横向响应距离2 +d 横向制动距离2 +d 横向冗余 ;

[0076] wherein d 安全距离1 is the safe distance from the current position of the ego vehicle to the rear vehicle in the target lane, d 纵向响应距离1 is the longitudinal response distance of the rear vehicle in the target lane within the response time after sensing the collision danger, d 纵向制动距离1 is the longitudinal braking distance of the rear vehicle in the target lane when decelerating at a relatively small deceleration after the response time, d 纵向冗余 is the longitudinal safety redundancy distance, d 横向冗余 is the lateral safety redundancy distance, d 安全距离2 is the safe distance from the current position of the ego vehicle to the vehicle in the adjacent lane of the target lane, d 横向响应距离 is the lateral response distance of the ego vehicle, d 横向制动距离 is the lateral braking distance of the ego vehicle, d 横向响应距离2 is the lateral response distance of the vehicle in the adjacent lane of the target lane within the response time after sensing the collision danger, d 横向制动距离2 is the lateral braking distance of the vehicle in the adjacent lane of the target lane when decelerating at a relatively small deceleration after the response time.

[0077] In this embodiment, the worst-case assumption is adopted in both the longitudinal and lateral response distances, i.e., the driver / vehicle is accelerating at a relatively large acceleration (a 纵向加速，大 ,a 横向加速,大 ) in the longitudinal and lateral directions before reacting to the collision danger after a period of time t 响应 . Then, the ego vehicle decelerates at a relatively small deceleration (a 纵向减速，小 ,a 横向减速,小 ) in the longitudinal and lateral directions, and appropriate longitudinal and lateral safety redundancy distances (d 纵向冗余 ,d 横向冗余 ) are added respectively.

[0078] The first mathematical model is d 安全距离1 = d 纵向响应距离1 + d 纵向制动距离1 + d 纵向冗余 + d 横向冗余 . Wherein, d 纵向响应距离1 represents the distance traveled by the ego vehicle in the accelerating motion before sensing the danger, v 他车1纵向 represents the longitudinal velocity of the other vehicle 1 assuming that it moves at a constant speed within the response time t 响应 , t 响应 represents the time required for the ego vehicle to sense the danger and take action, a 纵向加速，大 represents the longitudinal acceleration of the ego vehicle and the other vehicle 1 assuming that they accelerate within the response time t 响应 , d 纵向制动距离1represents the longitudinal distance of the ego vehicle in the deceleration motion after sensing the danger and taking action, a 纵向减速，小 represents the longitudinal deceleration of the ego vehicle and the other vehicle 1 in the deceleration process after the response time t 响应 represents the longitudinal deceleration of the ego vehicle and the other vehicle 1 in the deceleration process after the response time t 纵向冗余 represents the longitudinal safety redundancy distance, d 横向冗余 represents the lateral safety redundancy distance, wherein the longitudinal safety redundancy distance and the lateral safety redundancy distance can be dynamically set by the user's preference, ensuring the flexibility of the system and giving the driver a more personalized intelligent experience.

[0079] The second mathematical model d 安全距离2 = d 横向响应距离 + d 横向制动距离 + d 横向响应距离2 + d 横向制动距离2 + d 横向冗余 , wherein

[0080] d 横向响应距离 represents the lateral distance of the ego vehicle in the acceleration motion when changing lanes, v 自车横向 represents the lateral speed of the ego vehicle in the uniform motion during the driving process within the response time t 响应 represents the lateral speed of the ego vehicle in the uniform motion during the driving process within the response time t 横向加速，大 represents the lateral acceleration of the ego vehicle and the other vehicle 2 in the acceleration process after the response time t 响应 represents the lateral distance of the ego vehicle in the deceleration motion after sensing the danger and taking action, a 横向制动距离 represents the lateral distance of the ego vehicle in the deceleration motion after sensing the danger and taking action, a 横向减速，小 represents the lateral deceleration of the ego vehicle and the other vehicle 2 in the deceleration process after the response time t 响应 represents the lateral deceleration of the ego vehicle and the other vehicle 2 in the deceleration process after the response time t 横向响应距离2 represents the lateral distance of the other vehicle 2 in the acceleration motion when changing lanes, v 他车2横向 represents the lateral speed of the other vehicle 2 in the uniform motion during the driving process within the response time t 响应 represents the lateral speed of the other vehicle 2 in the uniform motion during the driving process within the response time t 横向制动距离2 represents the lateral distance of the ego vehicle in the deceleration motion after sensing the danger and taking action.

[0081] The first mathematical model and the second mathematical model improve the safety of the overall decision system by evaluating the safety distance under the assumption of dangerous working conditions, and the longitudinal safety redundancy distance and the lateral safety redundancy distance can be dynamically set by the user's preference, ensuring the flexibility of the system.

[0082] The embodiment proposes a quantifiable lane changing safety evaluation model to ensure the safety of the intelligent vehicle when changing lanes laterally, and also solves the shortcomings of other decision-making methods that are not transparent and difficult to quantify.

[0083] Further, the decision module receives the lane-changing decision and performs lane-changing decision processing, and the trajectory generation module generates a trajectory according to the lane-changing decision processing and transmits the trajectory to the motion control module, including:

[0084] Step S310: The traditional decision unit receives the lane-changing decision and performs lane-changing decision processing. If the lane-changing action can be successfully performed, the traditional trajectory generation unit generates a traditional trajectory and transmits the traditional trajectory to the motion control module. If the lane-changing action cannot be successfully performed, the interactive decision unit receives the lane-changing decision.

[0085] Step S320: The interactive decision unit receives the lane-changing decision and performs lane-changing decision processing. If the lane-changing action can be successfully performed, the interactive trajectory generation unit generates an interactive trajectory and transmits the interactive trajectory to the motion control module.

[0086] In this embodiment, the decision module includes a traditional decision unit and an interactive decision unit. The traditional decision unit is more safety-oriented and has lower efficiency than the interactive decision unit, and it is more difficult to implement vehicle lane-changing on a congested road. The interactive decision unit is added to provide a more efficient and proactive decision unit for vehicle lane-changing when the traditional decision unit cannot successfully perform the lane-changing action. Through proactive interactive behavior, the efficiency of the entire vehicle in merging and lane-changing is greatly improved, and the driver and passenger have a more personalized intelligent experience.

[0087] Therefore, the lane-changing decision can be transmitted to the traditional decision unit for lane-changing decision processing. If the traditional decision unit can successfully perform the lane-changing action, a traditional trajectory is generated by the traditional trajectory generation unit and transmitted to the motion control module. If the traditional decision unit cannot successfully perform the lane-changing action, the lane-changing decision is transmitted to the interactive decision unit, which decides whether the lane-changing action can be successfully performed. If the lane-changing action can be successfully performed, an interactive trajectory is generated by the interactive trajectory generation unit and transmitted to the motion control module.

[0088] Preferably, the lane-changing decision can be transmitted to the traditional decision unit and the interactive decision unit simultaneously for lane-changing decision processing, saving transmission time and further improving the efficiency of lane-changing decision processing.

[0089] Further, referring to Figure 4 , the interactive decision unit receives the lane-changing decision and performs lane-changing decision processing. If the lane-changing action can be successfully performed, the interactive trajectory generation unit generates an interactive trajectory and transmits the interactive trajectory to the motion control module, including:

[0090] Step S321: The interactive decision unit receives the lane-changing decision.

[0091] Step S322: Set the social rule module, and enter the behavior trial stage to perform behavior trial on the other vehicle under the condition of conforming to the social rule module;

[0092] Step S323: Enter the opponent vehicle behavior analysis stage to analyze whether the opponent vehicle has a yielding intention after behavior trial, if it is judged that the opponent vehicle has a yielding intention, enter the lane changing execution stage, if it is judged that the opponent vehicle has no yielding intention, enter the lane changing waiting stage, and repeat step S322;

[0093] Step S324: Trigger the back-to-normal adjustment after the vehicle body of the ego vehicle completely enters the target lane, and enter the back-to-normal adjustment stage;

[0094] Step S325: End the entire lane changing process when the vehicle body of the ego vehicle is flush with the target lane.

[0095] In the embodiment, the interactive decision unit is divided into five stages, i.e., behavior trial, opponent vehicle behavior analysis, lane changing execution, lane changing waiting, and back-to-normal adjustment. The behavior trial stage commands the ego vehicle to make a pre-action with a lane changing tendency under the premise of safety, feasibility, and conformity to social rules. After the opponent vehicle behavior analysis stage analyzes that the opponent vehicle decides to yield, the lane changing execution stage is entered. After the vehicle body of the ego vehicle completely enters the target lane, the back-to-normal adjustment is triggered to make the vehicle body of the ego vehicle flush with the target lane, and the entire lane changing process is ended; if the opponent vehicle has no yielding intention, the lane changing waiting stage is entered, and the yielding intention of a subsequent opponent vehicle is continuously judged through further behavior trial at a suitable time.

[0096] In the interactive decision unit, the essence is a consultative driving strategy. When the ego vehicle as a road right requester initiates a road right request to a road right owner (generates a trial behavior trajectory), if the road right owner's permission (such as deceleration, avoidance, etc.) is obtained, the lane changing execution stage is entered, and a fast lateral lane changing trajectory is generated. If the road right owner's attitude is not clear or is refused, the lane changing waiting stage is entered, and a defensive and conservative driving strategy is degenerated, and the behavior trial is continued until the road right owner's attitude is updated or a new road right owner comes.

[0097] Further, the behavior trial commands the ego vehicle to make a pre-action with a lane changing tendency, and the pre-action includes adjusting the vehicle head direction, turning on the turn signal, and partially invading the target lane with the vehicle head.

[0098] In the embodiment, the behavior trial stage commands the ego vehicle to make a pre-action with a lane changing tendency under the premise of safety, feasibility, and conformity to social rules, such as adjusting the vehicle head direction, turning on the turn signal, and partially invading the target lane with the vehicle head. The behavior trial stage is mainly used to judge the avoidance intention of the opponent vehicle.

[0099] Further, the social rule module includes lane changing rules, road right rules, etc., to provide a basis for the interactive decision unit to make lane changing decision processing.

[0100] In this embodiment, the social rule module includes some lane changing rules and road right rules commonly followed by experienced drivers when changing lanes and merging, such as:

[0101] When merging, the vehicle that is expected to reach the target intersection first has the right of way and can go first;

[0102] When merging, the vehicle on the main road goes first under the same conditions, and the ego vehicle can request the right of way through negotiation;

[0103] When changing lanes, the lane changing initiator is fully responsible for the collision (the right of way requester);

[0104] When changing lanes, the vehicle is considered to have completed the lane change and has the right of way as long as the entire vehicle body is inside the target lane.

[0105] The lane changing rules and road right rules built into the social rule module provide a basis for the interactive decision module to make decisions.

[0106] Further, the interactive decision unit receives the lane changing decision and makes lane changing decision processing, and if the lane changing action can be successfully executed, the lane changing operation is performed, and the interactive trajectory generation unit generates a series of future driving trajectories of the ego vehicle in the behavior exploration stage, the opponent vehicle behavior analysis stage, the lane changing execution stage, and the lane changing waiting stage.

[0107] In this embodiment, the interactive trajectory generation unit generates a series of future driving trajectories of the ego vehicle according to the instructions of the upstream decision, and finally outputs them to the downstream motion control module for corresponding control execution. The trajectory generation principle in the lane changing execution stage tends to be efficient rather than comfortable, ensuring that the user's vehicle can quickly and first complete the lane changing requirement, so the generated trajectory sequence will have a relatively large curvature, heading change, and high lateral speed; the behavior exploration stage will plan a relatively short length trajectory that allows the front side of the vehicle to slowly invade the target lane.

[0108]

Second Embodiment

[0109] Referring to Figure 5 , the present embodiment provides an intelligent vehicle 100, which includes a processor 120, a memory, and a program or instruction stored in the memory 110 and executable on the processor 120, and the program or instruction is executed by the processor to implement the lane changing decision method considering multi-vehicle interaction according to any embodiment of the present application.

[0110] It is a structural schematic diagram of an intelligent vehicle 100 provided by the second embodiment of the present application, and the intelligent vehicle 100 includes a processor 120 and a memory 110 electrically connected to the processor 120, for example, and the memory 110 stores a computer program 111, and the processor 120 loads the computer program 111 to implement the lane changing decision method considering multi-vehicle interaction as in the first embodiment.

[0111] The intelligent vehicle of the embodiment of the present application is used to implement the lane changing decision method considering multi-vehicle interaction as in any embodiment of the present application, and therefore has all the beneficial effects of the lane changing decision method considering multi-vehicle interaction as in any embodiment of the present application, which will not be repeated here.

[0112]

Third Embodiment

[0113] Referring to Figure 6 The embodiment provides a readable storage medium 200, which includes a stored computer program 111, wherein when the computer program 111 is run by a processor 120, the readable storage medium 200 controls the device where the readable storage medium 200 is located to perform the steps of the lane changing decision method considering multi-vehicle interaction as in any embodiment of the present application.

[0114] The embodiment also provides a readable storage medium 200, which stores computer executable instructions 210, and when the computer executable instructions 210 are read and run by a processor, the readable storage medium 200 controls the intelligent vehicle 100 where the readable storage medium 200 is located to implement the lane changing decision method considering multi-vehicle interaction as in the first embodiment.

[0115] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can also be implemented in other manners. The above described device embodiments are only schematic, for example, the various functional modules in the various embodiments of the present application can be integrated in one independent part, or exist in separate parts, or two or more modules can be integrated in one independent part.

[0116] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the parts of the present application that essentially contribute to the prior art or the parts of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the embodiments of the present application. The aforementioned readable storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0117] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A lateral lane change decision-making method considering multi-vehicle interaction, characterized in that, The lateral lane change decision-making method includes: Step S100: The environmental perception module perceives the vehicle driving environment information and vehicle operating status information, and transmits the information to the lane change trigger; Step S200: The lane change trigger receives the information, performs data analysis, and decides whether the vehicle should change lanes. If it should change lanes, the lane change decision is transmitted to the decision module. Step S300: The decision module receives the lane change decision and performs lane change decision processing; the trajectory generation module generates a trajectory based on the lane change decision processing and transmits the trajectory to the motion control module. Step S400: The motion control module receives the trajectory and performs a lane change operation according to the trajectory to realize the lane change; The lane change trigger includes a lane change safety assessment module, the decision module includes an interactive decision unit and a traditional decision unit, and the trajectory generation module includes an interactive trajectory generation unit and a traditional trajectory generation unit. The decision-making module receives the lane change decision and processes it. The trajectory generation module generates a trajectory based on the lane change decision processing and transmits the trajectory to the motion control module, including: Step S310: The conventional decision-making unit receives the lane change decision and processes it. If the lane change action can be successfully executed, the conventional trajectory generation unit generates a conventional trajectory and transmits the conventional trajectory to the motion control module. If the lane change action cannot be successfully executed, the interactive decision-making unit receives the lane change decision. Step S320: The interactive decision unit receives the lane change decision and processes it. If the lane change action can be successfully executed, the interactive trajectory generation unit generates an interactive trajectory and transmits the interactive trajectory to the motion control module. The interactive decision-making unit receives the lane change decision and processes it. If the lane change action is successfully executed, the interactive trajectory generation unit generates an interactive trajectory and transmits the interactive trajectory to the motion control module, including: Step S321: The interactive decision-making unit receives the lane change decision; Step S322: Set the social rules module, and if the behavior is in compliance with the social rules module, enter the behavior probing stage and conduct behavior probing on other vehicles; Step S323: Enter the opponent vehicle behavior analysis stage, analyze whether the opponent vehicle intends to yield after the behavior probing. If it is determined that the opponent vehicle intends to yield, then enter the lane change execution stage; if it is determined that the opponent vehicle does not intend to yield, then enter the lane change waiting stage and repeat step S322. Step S324: Once the vehicle body has fully entered the target lane, trigger the centering adjustment and enter the centering adjustment phase; Step S325: Once the vehicle body is aligned with the target lane, the entire lane change process ends.

2. The lateral lane change decision-making method according to claim 1, characterized in that, The lane change trigger receives the information, performs data analysis, and decides whether the vehicle should change lanes. If a lane change is required, the lane change decision is transmitted to the decision module, including: Step S210: The lane change safety assessment module receives the information and establishes a first mathematical model. Second mathematical model ; Step S220: Calculate the lateral and longitudinal distances between the current position of the vehicle and the vehicle behind it in the target lane. With the first mathematical model Perform a comparison; Step S230: Calculate the lateral distance between the current position of the vehicle and the vehicles in the adjacent lane of the target lane. With the second mathematical model Perform a comparison; Step S240: If the lane change conditions are met, the lane change decision is transmitted to the decision module through the lane change trigger; if the lane change conditions are not met, continue with steps S210-S240. Wherein, the lane change condition is and .

3. The lateral lane change decision-making method according to claim 2, characterized in that, The first mathematical model The second mathematical model ; in, The safe distance between the current position of the vehicle and the vehicle behind in the target lane. This refers to the longitudinal response distance of the vehicle following in the target lane within the response time after perceiving the collision hazard. This refers to the longitudinal braking distance traveled by a vehicle following in the target lane when it undergoes longitudinal deceleration and braking with a relatively small deceleration after the response time. For longitudinal safety redundancy distance, For lateral safety redundancy distance, This refers to the safe distance between the current position of the vehicle and vehicles in the adjacent lane of the target lane. for , for , This refers to the lateral response distance of vehicles in adjacent lanes within the response time after perceiving a collision hazard. It is the lateral braking distance traveled by vehicles in adjacent lanes of the target lane when they undergo lateral deceleration braking with a relatively small deceleration after the response time.

4. The lateral lane change decision-making method according to claim 1, characterized in that, The proposed behavior is to instruct the vehicle to make a preliminary action that indicates a tendency to change lanes. The preliminary action includes adjusting the direction of the vehicle, turning on the turn signal, and partially encroaching into the target lane.

5. The lateral lane change decision-making method according to claim 1, characterized in that, The social rules module includes driving regulations and right-of-way rules, which provide a basis for the interactive decision-making unit to make lane-changing decisions.

6. The lateral lane change decision-making method according to claim 1, characterized in that, The interactive decision-making unit receives the lane-changing decision and processes it. If the lane-changing action can be successfully executed, the lane-changing operation is performed. The system also includes: During the behavior exploration phase, the opponent vehicle behavior analysis phase, the lane change execution phase, and the lane change waiting phase, the interactive trajectory generation unit generates a series of future driving trajectories for the vehicle.

7. An intelligent vehicle, characterized in that, The intelligent vehicle includes: a processor, a memory, and a program or instructions stored in the memory and capable of running on the processor, wherein when the program or instructions are executed by the processor, they implement the steps of the lateral lane change decision method considering multi-vehicle interaction as described in any one of claims 1 to 6.

8. A readable storage medium, characterized in that, The readable storage medium stores a program or instructions that, when executed by a processor, implement the steps of the lateral lane change decision method considering multi-vehicle interaction as described in any one of claims 1 to 6.

Images